Guardian Procedure Errors and Messages Manual

Guardian Procedure
Errors and Messages

Manual
Abstract
This manual describes the Guardian messages for HP systems that use the HP
NonStopTM operating system. The manual covers the following types of messages:
error codes and error lists associated with Guardian procedure calls, the interprocess
messages sent to application programs by the operating system and the command
interpreter, traps, and signals.
Product Version
N/A
Supported Release Version Updates (RVUs)
This publication supports J06.03 and all subsequent J-series RVUs, H06.03 and all
subsequent H-series RVUs, and G06.15 and all subsequent G-series RVUs, until
otherwise indicated by its replacement publications. Additionally, all considerations for
H-series throughout this manual will hold true for J-series also, unless mentioned
otherwise.
Part Number Published

522628-011 February 2011
Document History
Part Number Product Version Published
522628-006 N/A August 2007
522628-008 N/A May 2008
522628-009 N/A February 2010
522628-010 N/A August 2010
522628-011 N/A February 2011
Legal Notices
© Copyright 2011 Hewlett-Packard Development Company L.P.
Confidential computer software. Valid license from HP required for possession, use or copying.
Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software
Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under
vendor's standard commercial license.
The information contained herein is subject to change without notice. The only warranties for HP
products and services are set forth in the express warranty statements accompanying such products
and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be
liable for technical or editorial errors or omissions contained herein.
Export of the information contained in this publication may require authorization from the U.S.
Department of Commerce.
Microsoft, Windows, and Windows NT are U.S. registered trademarks of Microsoft Corporation.
Intel, Itanium, Pentium, and Celeron are trademarks or registered trademarks of Intel Corporation or its
subsidiaries in the United States and other countries.
Java® is a U.S. trademark of Oracle and/or its affiliates.
Motif, OSF/1, UNIX, X/Open, and the "X" device are registered trademarks and IT DialTone and The
Open Group are trademarks of The Open Group in the U.S. and other countries.
Open Software Foundation, OSF, the OSF logo, OSF/1, OSF/Motif, and Motif are trademarks of the
Open Software Foundation, Inc.
OSF MAKES NO WARRANTY OF ANY KIND WITH REGARD TO THE OSF MATERIAL PROVIDED
HEREIN, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
OSF shall not be liable for errors contained herein or for incidental consequential damages in
connection with the furnishing, performance, or use of this material.
© 1990, 1991, 1992, 1993 Open Software Foundation, Inc. This documentation and the software to
which it relates are derived in part from materials supplied by the following:
© 1987, 1988, 1989 Carnegie-Mellon University. © 1989, 1990, 1991 Digital Equipment Corporation.
© 1985, 1988, 1989, 1990 Encore Computer Corporation. © 1988 Free Software Foundation, Inc.
© 1987, 1988, 1989, 1990, 1991 Hewlett-Packard Company. © 1985, 1987, 1988, 1989, 1990, 1991,
1992 International Business Machines Corporation. © 1988, 1989 Massachusetts Institute of
Technology. © 1988, 1989, 1990 Mentat Inc. © 1988 Microsoft Corporation. © 1987, 1988, 1989,
1990, 1991, 1992 SecureWare, Inc. © 1990, 1991 Siemens Nixdorf Informationssysteme AG. © 1986,
1989, 1996, 1997 Sun Microsystems, Inc. © 1989, 1990, 1991 Transarc Corporation.
This software and documentation are based in part on the Fourth Berkeley Software Distribution
under license from The Regents of the University of California. OSF acknowledges the following
individuals and institutions for their role in its development: Kenneth C.R.C. Arnold,
Gregory S. Couch, Conrad C. Huang, Ed James, Symmetric Computer Systems, Robert Elz. © 1980,
1981, 1982, 1983, 1985, 1986, 1987, 1988, 1989 Regents of the University of California.
Printed in the US
Guardian Procedure Errors and
Messages Manual
Index Figures Tables
Legal Notices
What’s New in This Manual vii
Manual Information vii
New and Changed Information vii
About This Manual xi
Purpose of This Manual xi
Who Should Read This Manual xi
How This Manual Is Organized xi
Related Reading xiii
Notation Conventions xiii
HP Encourages Your Comments xvii
1. Introduction
Procedure Errors 1-1
G-Series and Later Procedure Errors 1-1
C-Series Procedure Errors 1-2
Interprocess Messages 1-3
Interprocess Command Interpreter Messages 1-3
System Messages 1-3
Error Lists 1-3
Error List Content 1-3
Error Lists and Non-SPI Subsystems 1-4
Traps and Signals 1-4
2. File-System Errors
Error Codes 2-1
C-Series and D-Series Error Handling 2-2
Error Mapping to C Language errno Values 2-3
Error Code Descriptions 2-5
Error Lists 2-96
1: ZFIL-VAL-AWAITIO 2-99
Hewlett-Packard Company—522628-011
i
Contents 2. File-System Errors (continued)
2. File-System Errors (continued)

2: ZFIL-VAL-CHECKCLOSE 2-100
3: ZFIL-VAL-CHECKMONITOR 2-101
4: ZFIL-VAL-CHECKOPEN 2-103
5: ZFIL-VAL-CHECKPOINT 2-104
6: ZFIL-VAL-CHECKPOINTMANY 2-106
7: ZFIL-VAL-CHECKSWITCH 2-108
8: ZFIL-VAL-CONTROL 2-109
9: ZFIL-VAL-CREATE 2-111
10: ZFIL-VAL-KEYPOSITION 2-112
11: ZFIL-VAL-OPEN 2-113
12: ZFIL-VAL-PURGE 2-114
13: ZFIL-VAL-POSITION 2-115
14: ZFIL-VAL-READ 2-116
15: ZFIL-VAL-READLOCK 2-117
16: ZFIL-VAL-READUPDATE 2-118
17: ZFIL-VAL-READUPDATELOCK 2-119
18: ZFIL-VAL-REPLY 2-120
19: ZFIL-VAL-SETMODE 2-121
20: ZFIL-VAL-SETMODENOWAIT 2-122
21: ZFIL-VAL-WRITE 2-124
22: ZFIL-VAL-WRITEREAD 2-125
23: ZFIL-VAL-WRITEUPDATE 2-127
24: ZFIL-VAL-WRITEUPDATEUNLOCK 2-128
25: ZFIL-VAL-AWAITIOX 2-129
26: ZFIL-VAL-CHECKPOINTMANYX 2-130
27: ZFIL-VAL-CHECKPOINTX 2-132
28: ZFIL-VAL-KEYPOSITIONX 2-133
29: ZFIL-VAL-READLOCKX 2-134
30: ZFIL-VAL-READUPDATELOCKX 2-136
31: ZFIL-VAL-READUPDATEX 2-137
32: ZFIL-VAL-READX 2-138
33: ZFIL-VAL-REPLYX 2-139
34: ZFIL-VAL-WRITEREADX 2-140
35: ZFIL-VAL-WRITEUPDATEUNLOCKX 2-141
36: ZFIL-VAL-WRITEUPDATEX 2-143
37: ZFIL-VAL-WRITEX 2-144
38: ZFIL-VAL-CLOSE 2-145
39: ZFIL-VAL-DEVICEINFO2 2-146
Guardian Procedure Errors and Messages Manual—522628-011

ii
Contents 2. File-System Errors (continued)
2. File-System Errors (continued)

65: ZFIL-VAL-FILE-OPEN-CHKPT 2-147
66: ZFIL-VAL-FILE-CREATELIST 2-147
67: ZFIL-VAL-FILE-OPEN 2-148
68: ZFIL-VAL-FILE-PURGE 2-149
69: ZFIL-VAL-FILE-CLOSE 2-150
70: ZFIL-VAL-FILE-GETINFOBYNAME 2-150
71: ZFIL-VAL-FILE-GETRECEIVEINFO 2-151
72: ZFIL-VAL-FILENAME-COMPARE 2-152
73: ZFIL-VAL-FILE-GETOPENINFO 2-152
74: ZFIL-VAL-DISK-REFRESH 2-153
75: ZFIL-VAL-FILE-RENAME 2-154
76: ZFIL-VAL-FILENAME-FINDSTART 2-155
77: ZFIL-VAL-FILENAME-FINDNEXT 2-155
78: ZFIL-VAL-FILENAME-FINDFINISH 2-156
80: ZFIL-VAL-FILE-CREATE 2-157
3. Sequential I/O Errors

Message Descriptions 3-1
4. DEFINE Errors
Error Codes 4-1
5. NEWPROCESS AND NEWPROCESSNOWAIT Errors

Error Codes 5-1
Error Lists 5-20
3: ZGRD-VAL-NEWPROCESS 5-21
4: ZGRD-VAL-NEWPROCESSNOWAIT 5-22
6. Process Creation Errors

Error Codes 6-1
Error Lists 6-36
22: ZGRD-VAL-PROCESS-CREATE 6-37
25: ZGRD-VAL-PROCESS-LAUNCH 6-39
7. PROCESS_GETINFOLIST_ Errors
Error Codes 7-1
8. PROCESS_GETPAIRINFO_ Errors
Error Codes 8-1

iii
Contents 9. PROCESS_SPAWN_ Open System Services
(OSS) Errors
9. PROCESS_SPAWN_ Open System Services (OSS) Errors

Guardian Error Codes 9-1
OSS Error Codes 9-2
Error Lists 9-6
24: ZGRD-VAL-PROCESS-SPAWN 9-7
10. ALLOCATESEGMENT Errors

Error Codes 10-1
Error Lists 10-4
1: ZGRD-VAL-ALLOCATESEGMENT 10-5
11. SEGMENT_ALLOCATE_ Errors

Error Codes 11-1
Error Lists 11-6
21: ZGRD-VAL-SEGMENT-ALLOCATE 11-7
12. USESEGMENT Errors

Error Lists 12-1
2: ZGRD-VAL-USESEGMENT 12-2
13. SEGMENT_USE_ Errors

Error Lists 13-1
23: ZGRD-VAL-SEGMENT-USE 13-2
14. Subsystem Programmatic Interface (SPI) Errors

Error Codes 14-1
Error Lists 14-9
Error From SSGET or SSGETTKN 14-10
Error From SSINIT 14-11
Error From SSMOVE or SSMOVETKN 14-12
Error From SSNULL 14-13
Error From SSPUT or SSPUTTKN 14-14
15. EDITREAD and EDITREADINIT Errors

Error Codes 15-1
16. IOEdit Errors

Error Message Forms 16-1
File-System Error Codes 16-2
Error Codes Unique to IOEdit 16-5
Catastrophic Error Messages 16-8

iv
Contents 17. Formatter Errors
17. Formatter Errors

Error Codes 17-1
18. INITIALIZER Errors

Error Messages 18-1
19. Interprocess Command Interpreter Messages

20. System Messages

Application Conversion 20-1
Error Return Conventions 20-1
21. Traps and Signals

Trap Handling 21-2
Signal Handling 21-2
Signals and Trap Numbers 21-2
Error Lists 21-9
5: ZGRD-VAL-TRAPOUT 21-10
22. OSS Error Information

Brief OSS Error Information 22-1
Detailed OSS Error Information 22-1
Index
Figures
Figure 1-1. Error List Format 1-4
Figure 19-1. Command Interpreter Messages 19-1
Tables
Table 1-1. Error Designations 1-2
Table 2-1. File-System Error Categories 2-1
Table 2-2. D-Series Error Information Procedures 2-2
Table 2-3. Guardian File-System Errors Mapped to C Errors 2-3
Table 2-4. Conditional Tokens That Can Appear in Any File-System Error
List 2-96
Table 5-1. NEWPROCESS Error 6 Error Subcodes 5-4
Table 6-1. Error Subcodes for PROCESS_LAUNCH_ and PROCESS_SPAWN_
Errors 2 and 3 6-3

v
Contents Tables (continued)
Tables (continued)
Table 6-2. Error Subcodes for Process Creation Errors 12, 13, 70, 76, 84, and
3xx 6-7
Table 6-3. Error Subcodes for Process Creation Error 64 6-23
Table 9-1. PROCESS_SPAWN_ Error Fields 9-1
Table 14-1. Supplementary Status Values 14-7
Table 19-1. Command Interpreter Messages 19-2
Table 20-1. D-Series-Format System Messages 20-2
Table 20-2. C-Series and D-Series System Messages Compared 20-3
Table 20-3. Completion Codes 20-21
Table 21-1. TNS/R Native Signal Names, Signal Numbers, Trap Numbers, and Trap
Descriptions 21-1

vi
What’s New in This Manual
Manual Information
Guardian Procedure Errors and Messages Manual
Abstract
NonStopTM operating system. The manual covers the following types of messages:
error codes and error lists associated with Guardian procedure calls, the interprocess
messages sent to application programs by the operating system and the command
interpreter, traps, and signals.
Product Version
N/A
Supported Release Version Updates (RVUs)
This publication supports J06.03 and all subsequent J-series RVUs, H06.03 and all
subsequent H-series RVUs, and G06.15 and all subsequent G-series RVUs, until
otherwise indicated by its replacement publications. Additionally, all considerations for
H-series throughout this manual will hold true for J-series also, unless mentioned
otherwise.
Part Number Published

522628-011 February 2011
Document History
Part Number Product Version Published
522628-006 N/A August 2007
522628-008 N/A May 2008
522628-009 N/A February 2010
522628-010 N/A August 2010
522628-011 N/A February 2011
New and Changed Information

Changes to the H06.22/J06.11 manual:
• Added error detail 19 for error 77 on pages 6-27 and 6-29.
• Updated the format of Remote Password on page 19-17.
• Added error code description for Error 1092 on page 2-86.

vii
What’s New in This Manual Changes to the H06.21/J06.10 Manual
Changes to the H06.21/J06.10 Manual

• Added a description for error 35 on page 2-16.
• Added file-system errors 758 through 766 on page 2-84.
• Updated Note under Signals and Trap Numbers on page 21-2.
• Updated the description of error code 72 on page 6-10.
• Removed the instance of “-instance_data data2protected” from recovery action 10
of error code 77 on page 6-28.

• Added the following new error code descriptions:
° Errors 700 on page 2-76 to 711 on page 2-79.
° Error 734 on page 2-83.

• Supported release statements have been updated to include J-series RVUs.
Changes to the H06.11 Manual

• Updated the name of error 733 (%1335) FEBRANCHISFAILED on page 2-83.
• Added these new errors:
° 3502 (%6656) FEREQUESTALLOCATIONFAILURE on page 2-87
° 538 SIOERRÊXTSIZEÔVERFLOW: Extent size is greater than 65535 pages

on page 3-7
° 539 SIOERR^PAGEWRITEÔVERFLOW: The highest possible page in EDIT

file has been written. on page 3-8
° 2081 (%004041) Bad internal format. on page 4-9

• Added subcode 75 on page 6-10 in the Error Subcodes for Process Creation
Errors 12, 13, 70, 76, 84, and 3xx table.

viii
About This Manual
Purpose of This Manual
NonStop operating system. It covers the following types of messages:
• The error codes and error lists associated with Guardian procedure calls
• The interprocess messages sent to application programs by the operating system
and the command interpreter
• The trap numbers, signals, and the error list for traps and signals
Each message description explains the cause of the message, states the effect on the
system, and provides a recommended recovery action.
Who Should Read This Manual

This manual is intended mainly for application programmers whose programs call
Guardian procedures directly. Applications can check for errors returned by these
procedures, and subsequent action can be initiated based on the error returned.
Other users who can benefit from this manual include interactive users of HP NonStop
systems, system managers, and system operators.
How This Manual Is Organized

This manual has the following organization.
• Section 1, Introduction, provides an overview of the various messages described in
this manual.
• Section 2, File-System Errors, describes the file-system error codes and error lists.
• Section 3, Sequential I/O Errors, describes the error codes produced by the
sequential I/O (SIO) procedures.
• Section 4, DEFINE Errors, describes the errors that relate specifically to DEFINE
attribute sets.
• Section 5, NEWPROCESS AND NEWPROCESSNOWAIT Errors, describes the
error codes and error lists produced by the process-control procedures
NEWPROCESS and NEWPROCESSNOWAIT.
• Section 6, Process Creation Errors, describes the error codes and error lists
produced by the process-creation procedures PROCESS_LAUNCH_ and
PROCESS_CREATE_. It also describes Guardian error codes returned by the
PROCESS_SPAWN_ procedure. Open System Services (OSS) error codes and

ix
About This Manual How This Manual Is Organized
the error list returned by the PROCESS_SPAWN_ procedure are listed in

Section 9, PROCESS_SPAWN_ Open System Services (OSS) Errors.
• Section 7, PROCESS_GETINFOLIST_ Errors, describes the error codes produced
by the PROCESS_GETINFOLIST_ procedure.
• Section 8, PROCESS_GETPAIRINFO_ Errors, describes the error codes produced
by the PROCESS_GETPAIRINFO_ procedure.
• Section 9, PROCESS_SPAWN_ Open System Services (OSS) Errors, describes
the OSS error codes and error list produced by the process-creation procedure
PROCESS_SPAWN_. Guardian error codes returned by both the
PROCESS_SPAWN_ and PROCESS_CREATE_ procedures are documented in
Section 6, Process Creation Errors.
• Section 10, ALLOCATESEGMENT Errors, describes the error codes and error lists
produced by the ALLOCATESEGMENT procedure.
• Section 11, SEGMENT_ALLOCATE_ Errors, describes the error codes and error
lists produced by the SEGMENT_ALLOCATE_ procedure.
• Section 12, USESEGMENT Errors, describes the error list associated with the
USESEGMENT procedure.
• Section 13, SEGMENT_USE_ Errors, describes the error list associated with the
SEGMENT_USE_ procedure.
• Section 14, Subsystem Programmatic Interface (SPI) Errors, describes the error
codes issued by the SPI procedures and the error lists associated with those
procedures.
• Section 15, EDITREAD and EDITREADINIT Errors, describes the errors returned
by the EDITREAD and EDITREADINIT procedures.
• Section 16, IOEdit Errors, describes the error messages returned by the IOEdit
procedures. These errors include file-system errors that have specific meaning for
IOEdit.
• Section 17, Formatter Errors, describes the error codes produced by the
FORMATDATA[X] procedures.
• Section 18, INITIALIZER Errors, describes the error messages produced by the
INITIALIZER procedure.
• Section 19, Interprocess Command Interpreter Messages, describes the command
interpreter system messages that an application can receive through its
$RECEIVE file.
• Section 20, System Messages, describes the operating-system messages that an
application can receive through its $RECEIVE file.
• Section 21, Traps and Signals, describes trap codes, TNS/R native signals, and
the error lists used to report trap and signal conditions.

x
About This Manual Related Reading
• Section 22, OSS Error Information, describes how to find information on Open
System Services (OSS) errors.
Related Reading
This manual assumes that you are familiar with the HP system architecture and the
NonStop operating system. The following manuals provide information about the
hardware architecture and the operating system:
• Introduction to Tandem NonStop Systems
• The introductions and system description manuals for individual HP NonStop
systems
While using this manual, you might need to refer to the following related programming
manuals.
• Guardian Programmer’s Guide
• Guardian Procedure Calls Reference Manual
• Guardian Programming Reference Summary
• Open System Services System Calls Reference Manual
• Open System Services System Calls Reference Manual
Notation Conventions
General Syntax Notation
The following list summarizes the notation conventions for syntax presentation in this
manual.
UPPERCASE LETTERS. Uppercase letters indicate keywords and reserved words; enter
these items exactly as shown. Items not enclosed in brackets are required. For
example:
MAXATTACH
lowercase italic letters. Lowercase italic letters indicate variable items that you supply.
Items not enclosed in brackets are required. For example:
file-name
computer type. Computer type letters within text indicate C and Open System Services
(OSS) keywords and reserved words. Type these items exactly as shown. Items not
enclosed in brackets are required. For example:
myfile.c

xi
About This Manual General Syntax Notation
italic computer type. Italic computer type letters within text indicate C and Open
System Services (OSS) variable items that you supply. Items not enclosed in brackets
are required. For example:
pathname
[ ] Brackets. Brackets enclose optional syntax items. For example:

TERM [\system-name.]$terminal-name
INT[ERRUPTS]
A group of items enclosed in brackets is a list from which you can choose one item or
none. The items in the list may be arranged either vertically, with aligned brackets on
each side of the list, or horizontally, enclosed in a pair of brackets and separated by
vertical lines. For example:
LIGHTS [ ON ]
[ OFF ]
[ SMOOTH [ num ] ]
K [ X | D ] address-1
{ } Braces. A group of items enclosed in braces is a list from which you are required to
choose one item. The items in the list may be arranged either vertically, with aligned
braces on each side of the list, or horizontally, enclosed in a pair of braces and
separated by vertical lines. For example:
LISTOPENS PROCESS { $appl-mgr-name }
{ $process-name }
ALLOWSU { ON | OFF }
| Vertical Line. A vertical line separates alternatives in a horizontal list that is enclosed in
brackets or braces. For example:
INSPECT { OFF | ON | SAVEABEND }
… Ellipsis. An ellipsis immediately following a pair of brackets or braces indicates that you
can repeat the enclosed sequence of syntax items any number of times. For example:
M address-1 [ , new-value ]...
[ - ] {0|1|2|3|4|5|6|7|8|9}...
An ellipsis immediately following a single syntax item indicates that you can repeat that
syntax item any number of times. For example:
"s-char..."
Punctuation. Parentheses, commas, semicolons, and other symbols not previously

described must be entered as shown. For example:
error := NEXTFILENAME ( file-name ) ;
LISTOPENS SU $process-name.#su-name

xii
About This Manual Notation for Messages
Quotation marks around a symbol such as a bracket or brace indicate that the symbol
is a required character that you must enter as shown. For example:
"[" repetition-constant-list "]"
Item Spacing. Spaces shown between items are required unless one of the items is a
punctuation symbol such as a parenthesis or a comma. For example:
CALL STEPMOM ( process-id ) ;
If there is no space between two items, spaces are not permitted. In the following
example, there are no spaces permitted between the period and any other items:
$process-name.#su-name
Line Spacing. If the syntax of a command is too long to fit on a single line, each
continuation line is indented three spaces and is separated from the preceding line by
a blank line. This spacing distinguishes items in a continuation line from items in a
vertical list of selections. For example:
ALTER [ / OUT file-spec / ] CONTROLLER
[ , attribute-spec ]...
Notation for Messages

The following list summarizes the notation conventions for the presentation of
displayed messages in this manual.
Nonitalic text. Nonitalic letters, numbers, and punctuation indicate text that is displayed or
returned exactly as shown. For example:
Backup Up.
lowercase italic letters. Lowercase italic letters indicate variable items whose values are
displayed or returned. For example:
p-register
process-name
[ ] Brackets. Brackets enclose items that are sometimes, but not always, displayed. For
example:
Event number = number [ Subject = first-subject-value ]
A group of items enclosed in brackets is a list of all possible items that can be
displayed, of which one or none might actually be displayed. The items in the list might
be arranged either vertically, with aligned brackets on each side of the list, or
horizontally, enclosed in a pair of brackets and separated by vertical lines. For
example:
LDEV ldev [ CU %ccu | CU %... ] UP [ (cpu,chan,%ctlr,%unit) ]

xiii
About This Manual Notation for Management Programming Interfaces
{ } Braces. A group of items enclosed in braces is a list of all possible items that can be
displayed, of which one is actually displayed. The items in the list might be arranged
either vertically, with aligned braces on each side of the list, or horizontally, enclosed in
a pair of braces and separated by vertical lines. For example:
LBU { X | Y } POWER FAIL
process-name State changed from old-objstate to objstate
{ Operator Request. }
{ Unknown. }
| Vertical Line. A vertical line separates alternatives in a horizontal list that is enclosed in
brackets or braces. For example:
Transfer status: { OK | Failed }
% Percent Sign. A percent sign precedes a number that is not in decimal notation. The
%þnotation precedes an octal number. The %Bþnotation precedes a binary number.
The %Hþnotation precedes a hexadecimal number. For example:
%005400
P=%p-register E=%e-register
Notation for Management Programming Interfaces

The following list summarizes the notation conventions used in the boxed descriptions
of error lists in this manual.
UPPERCASE LETTERS. Uppercase letters indicate names from definition files; enter these
names exactly as shown. For example:
ZCOM-TKN-SUBJ-SERV
lowercase letters. Words in lowercase letters are words that are part of the notation,
including Data Definition Language (DDL) keywords. For example:
token-type
Change Bar Notation

Change bars are used to indicate substantive differences between this edition of the
manual and the preceding edition. Change bars are vertical rules placed in the right
margin of changed portions of text, figures, tables, examples, and so on. Change bars
highlight new or revised information. For example:
The message types specified in the REPORT clause are different in the COBOL85
environment and the Common Run-Time Environment (CRE).
The CRE has many new message types and some new message type codes for
old message types. In the CRE, the message type SYSTEM includes all messages
except LOGICAL-CLOSE and LOGICAL-OPEN.

xiv
About This Manual HP Encourages Your Comments
HP Encourages Your Comments

HP encourages your comments concerning this document. We are committed to
providing documentation that meets your needs. Send any errors found, suggestions
for improvement, or compliments to docsfeedback@hp.com.
Include the document title, part number, and any comment, error found, or suggestion
for improvement you have concerning this document.

xv
About This Manual HP Encourages Your Comments

xvi
1 Introduction
This manual describes the Guardian messages associated with the NonStop operating
system. This manual is dual-threaded, covering both G-series and H-series releases.
For each message, the description provides an explanation of the cause, a discussion
of the effect on the system, and suggestions for corrective action or response.
Several types of messages can be returned to your program:
• Procedure errors
• Interprocess messages
• Error lists
• Traps and signals
This introductory section explains each of these message categories.
Note that Sections 2-18 describe the messages associated with specific procedures.
Each of these sections describes procedure errors or error lists or both. Section 19 and
Section 20 describe interprocess messages. Section 21 describes traps and signals.
Section 22 describes how to find more information on Open System Services (OSS)
errors.
Procedure Errors
Procedure errors are values returned to your program after your program calls
procedures such as file-system procedures. Your application program should test for
possible errors and take appropriate action when necessary.
Your program detects procedure errors in two ways, depending on the procedure being
called. Errors returned by D-series and later procedures are not in the same format as
those returned by C-series procedures. The names of D-series and later procedures
can be recognized by their trailing underscore; the names of most C-series procedures
do not contain the underscore character.
G-Series and Later Procedure Errors

Many G-series and later procedures return procedure error values through an output
parameter of the procedure. For these procedures, you simply check the value of the
error or status parameter to determine whether an error occurred. For example, an
error returned in the error parameter of the procedure FILE_CREATE_ is a simple
file-system error code giving the status of the operation.
Some G-series and later procedures return additional information through an error-
item or error-detail parameter. Depending on the procedure, this information can
supply the ordinal number of the parameter in error or other detailed information about
the error encountered.
All G-series and later operating-system procedures have an integer returned value
called error or status. If the error might have associated information, the extra
information is returned in an integer error-detail output parameter. If error is a

1 -1
Introduction C-Series Procedure Errors
returned value, it is always possible to obtain it or return it. Bounds checking is not
required.
When error-detail is used, most error values share the error designations
shown in Table 1-1.
Table 1-1. Error Designations

Value Meaning
0 Successful operation
1 File-system error other than 0, 2, or 3; error-detail contains the file-system
error number
2 Parameter error
3 Bounds error on reference parameter
4-n Other error; error-detail can contain additional information
Four G-series file information procedures are available for checking on error
conditions:
• FILE_GETINFO_
• FILE_GETINFOBYNAME_
• FILE_GETINFOLIST_
• FILE_GETINFOLISTBYNAME_
These procedures are described in the Guardian Procedure Calls Reference Manual.
See Section 2, File-System Errors, for information regarding their use in obtaining file-
system error information.
C-Series Procedure Errors

C-series procedures continue to use condition codes; condition codes are not used by
D-series and later procedures. A condition code of“equal”(CCE) indicates success; a
condition code of “less than” (CCL) or “greater than” (CCG) indicates that a problem
occurred and that the error should be retrieved. The completion of a read associated
with a C-series system message returns a condition code of “greater than” (CCG) and
file-system error 6 from the FILEINFO procedure.
As has been the case for previous releases of the operating system, you use
FILEINFO to obtain the error code. You should use the file number form of the
FILEINFO procedure to obtain information about errors relating to operations involving
open files; the last error associated with the file number is returned.
A typical method of testing for an error and calling FILEINFO to get the error value
might appear as follows:
IF <> THEN ! If CCL or CCG occurred, then
CALL FILEINFO ! call this procedure to get the error value.

1 -2
Introduction Interprocess Messages
Interprocess Messages
Interprocess messages are data structures that are exchanged between processes or
between the operating system and a process. Interprocess message are received by a
user process through its $RECEIVE file. A user process sends an interprocess
message by opening the process to which the message is to be sent.
This manual describes two types of interprocess messages:
• Interprocess command interpreter messages
• System messages
Interprocess Command Interpreter Messages

An interprocess command interpreter message is an interprocess message that is
exchanged between the command interpreter and an application process. The Tandem
Advanced Command Language (TACL) command interpreter is supplied by HP for use
on the operating system.
Use of interprocess command interpreter messages is documented in the Guardian
Programmer’s Guide.
System Messages
A system message is an interprocess message that is sent from the operating system
to an application process. Use of these messages is documented in the Guardian
Error Lists
Error lists are Subsystem Programmatic Interface (SPI) buffers returned to an
application program by another process. Error lists can be returned to your application
process if you are using SPI to send requests to another process. If the other process
encounters a procedure error, it returns the error information in an error list to your
application process.
Error List Content

Unlike the error messages described elsewhere in this manual, error list information is
returned in the form of tokens that are meaningful to application processes rather than
in the form of displayable text. The content of error lists can vary; for example:
• A response record can contain both response data and error or warning
information.
• A single response record can contain multiple error lists, especially if warnings
occur.

1 -3
Introduction Error Lists and Non-SPI Subsystems
• A single error can actually be a pass-through error, an error that originated in

another subsystem that was called by the subsystem to which the command was
sent.
Figure 1-1 shows the format of an error list.
Figure 1-1. Error List Format

header ZSPI-TKN- ZSPI-TKN- n error error ZSPI-TKN-
RETCODE ERRLIST token token ENDLIST
The value of the return token, ZSPI-TKN-RETCODE, indicates the content of the error
list:
• If zero and no error list follows, no error occurred.
• If zero but one or more error lists follow, the error lists contain only warning
messages: unusual conditions that might be of interest to the requester but do not
prevent the server from completing the command.
• If nonzero, the error list must contain at least one token with an error number that
matches the value of the return token. Each subsystem defines its own set of error
numbers.
Error Lists and Non-SPI Subsystems

Some HP software facilities do not have a programmatic command interface based on
SPI but do define standard error lists. If a HP subsystem that has an SPI-based
command interface calls one of these facilities and receives an error, the subsystem
usually tries to recover from the error. If that is not possible, the subsystem reports the
error in an SPI error token (ZSPI-TKN-ERROR), embeds it in an error list of the
prescribed form, and encloses this error list in its response.
For information about creating error lists, additional information about tokens and token
types, and definitions of tokens whose names begin with ZSPI-, refer to the SPI
Programming Manual.
Traps and Signals

Certain critical problems can cause a process to be unable to continue executing.
These are typically the result of coding errors, but other conditions, such as the lack of
a system resource (for example, memory), can also prevent normal process execution.
Such conditions are reported as traps to TNS processes and as signals to TNS/R
native processes. A trap is a software mechanism that stops process execution. A
signal is a software interrupt that can notify a process of other events, such as timer
expiration, as well as of critical error conditions.
The set of signals that are used in the Guardian environment is known as TNS/R
native signals. This set is a subset of a larger set of signals used in the Open System
Services (OSS) environment. Most of the TNS/R native signals are caused by the

1 -4
Introduction Traps and Signals
same conditions that cause traps to occur in TNS processes. In this manual, equivalent
trap and signal conditions are described together.
Each description of a trap or signal condition contains the following information:
• The signal name or trap number
• The cause of the trap or signal
• The effect of the error on the system
• The recovery actions you can take
You can use the ARMTRAP procedure to specify a location in your application program
where execution should begin if a trap occurs. Your program can use information
passed to investigate the cause of the error. You can even reset the trap mechanism
and cause the program to restart.
Use of the ARMTRAP procedure and user-written trap processes is documented in the
Guardian Programmer’s Guide.

1 -5
Introduction Traps and Signals

1 -6
2 File-System Errors
Error Codes
The file system of the NonStop operating system returns a code to the calling
procedure to indicate errors and other special conditions. Because most programs use
the file system, these conditions can occur during execution of almost any user-written
application or any program supplied by HP.
Many programs display file-system error codes on the user’s terminal. These
messages typically contain the code number and a short message. For example:
WARNING - $VOLUME.SUBVOL.FILE ERR 11
To find out what the file-system error means, look up the explanation in this section
under the error code number.
You can obtain a short explanation of any file-system error code or OSS errno value at
the TACL prompt by entering:
ERROR number
If you want to scan a list of all the errors, enter:
ERROR -1
Table 2-1 lists the major categories of error numbers.
Table 2-1. File-System Error Categories

Number Category
0 No error: The operation executed successfully. For C-series procedures, the
condition code is set to “equal to” (CCE).
1-9 Warning: The operation has executed with the exception of the indicated
condition. For warning 6, data is returned in the application process buffer. For
C-series procedures, the condition code is set to “greater than” (CCG).
10-255, Error: The operation encountered an error or a special condition that the
512- application must recognize, for instance, an aborted transaction on an audited
32767 TMF file. For C-series procedures, the condition code is set to “less than”
(CCL).
300-511 Application-defined error: These error numbers are reserved for use by
(except application processes.
538)
538 FILE^CHECK procedure: This procedure could not return the primary or
secondary extent value because the extent size is greater than 65535 pages.
4000- Open System Services error. For information on how to find the meaning of an
4999 OSS error, see Section 22, OSS Error Information.
5000- NonStop Storage Management Foundation (SMF) error.
5999

2 -1
File-System Errors C-Series and D-Series Error Handling
C-Series and D-Series Error Handling

When an operating-system procedure call returns an error, it is sometimes necessary
to call a second procedure to obtain additional information about the error. Some
C-series procedures return only condition codes and require that a second procedure
be called to obtain the file-system error code; other procedures might require different
information to handle the error properly.
When using C-series procedures, additional information about an error on a file is
obtained by calling the FILEINFO procedure. The information returned includes the
type of device associated with the error if you include the devtype parameter in the
FILEINFO call and if a device type is applicable. (For some errors returned by
subsystems such as the TMF subsystem, there is no associated physical device.)
Refer to Appendix A in the Guardian Procedure Calls Reference Manual for a list of the
device types and subtypes.
On D-series and later releases, file-system error information is also available using one
of the procedures described in Table 2-2.
Table 2-2. D-Series Error Information Procedures

Procedure When to Use
FILE_GETINFO_ To obtain information about a file using the file number
(filenum).
FILE_GETINFOLIST_ To obtain information about a file using the file number
(filenum).
FILE_GETINFOBYNAME_ To obtain information about a file using the file name
(file-name).
If you call FILE_GETINFOBYNAME_ in a waited
manner, the system returns the information in the
procedure output parameters. However, if you call this
procedure in a nowait manner, the system returns the
information in the system message -108 (nowait
FILE_GETINFOBYNAME_ completion), which is
described in Section 20, System Messages.
FILE_GETINFOLISTBYNAME_ To obtain information about a file using the file name
(file-name).
For more information on errors related to terminals, line printers, tape drives, card
readers, interprocess communication, and the operator console, refer to the Guardian
Programmer’s Guide. For more information about interprocess communication errors
that can occur when applications use the Subsystem Programmatic Interface (SPI),
refer to the SPI Programming Manual.

2 -2
File-System Errors Error Mapping to C Language errno Values
File-system errors returned by the data communications subsystems can have special
meanings depending on the particular subsystem, access method, or protocol being
used. After obtaining the device type associated with the error, consult Appendix A of
the Guardian Procedure Calls Reference Manual to determine which subsystem is
involved and refer to the appropriate manual for the subsystem.
Note. In the error code descriptions, if a device-type number includes a dot (.), the digits to the
left of the dot are the device type and the digits to the right of the dot are the device subtype.
Error Mapping to C Language errno Values

Some Guardian file-system errors map to corresponding C language error numbers.
These error numbers are returned in the errno value when using the C programming
language. See Table 2-3 for a list of file-system errors and their corresponding C
errno values.
Table 2-3. Guardian File-System Errors Mapped to C Errors (page 1 of 2)

File-System Error C errno Symbolic Name
0-1 ENOERR
2 ENOENT
4 EIO
11 ENOENT
12 EGUARDIANOPEN
13-15 ENOENT
16 EBADF
18 ENOENT
21 EINVAL
22 EFAULT
24 EPERM
29 EINVAL
31-37 ENOMEM
43-44 ENOSPC
45 EFBIG
46 EIO
48 EACCES
49 EINVAL
50-51 EIO
53 EFSERR
54 EINVAL

2 -3
File-System Errors Error Mapping to C Language errno Values
Table 2-3. Guardian File-System Errors Mapped to C Errors (page 2 of 2)

File-System Error C errno Symbolic Name
55 EIO
59 EFILEBAD
60 EWRONGID
61 EIO
73 EGUARDIANLOCKED
103 EIO
232 EIO
549, 560, 564, 590 EINVAL

2 -4
File-System Errors Error Code Descriptions
Error Code Descriptions

0 (%0) The operation completed successfully.
Cause. The file-system procedure successfully completed the requested operation.

Effect. The file-system procedure returns to the calling procedure.
Recovery. Informative message only; no corrective action is needed.
1 (%1) A read procedure reached end-of-file or a

write procedure reached end of tape.
(device type: 3, 4, or 6)
Cause. A read procedure encountered a logical or physical end-of-file indicator, or a

write procedure reached the end of the physical medium.
Effect. The procedure returns.
Recovery. Recovery, if any, is application dependent.
2 (%2) The operation specified is not allowed on

this type of file. (device type: any)
Cause. The procedure is invalid for the given type of file. For example:
• PURGE of a nondisk file
• KEYPOSITION on an unstructured disk file
• Invalid SETMODE operation
• ACTIVATERECEIVETRANSID, READUPDATE, or REPLY attempted on
$RECEIVE opened with a receive-depth of zero.
This error is also returned from REPLYX in any of the three following cases:
• The address of a parameter is extended, but no segment is in use at the time of
the call or the segment in use is invalid.
• The address of a parameter is extended, but is an absolute address.
• The file system cannot use the user’s segment when needed.
Effect. The operation might return without performing all or part of the requested
operation. For example, FILE_GETINFOLIST[BYNAME]_ can return without all of the
requested information.
Recovery. Correct the request.

2 -5
3 (%3) An open or purge of a partition failed.

(device type: 3)
Cause. An OPEN or FILE_OPEN_ operation could not open one or more partitions
defined for the file, or a purge or other operation for a partitioned file could not operate
on one or more of the defined partitions.
Effect. The open operation succeeds. However, any future attempts to access the
nonexistent partition return an error 72 (attempt to access unmounted partition). The
purge operation deletes what it can.
Recovery. Correct the operation or parameters.
4 (%4) An open operation for an alternate key file

failed. (device type: 3)
Cause. An open operation for a structured file with alternate keys could not open an
alternate-key file.
Effect. The open succeeds. However, any subsequent attempts to access the file by
way of an alternate key stored in the unopened file result in file-system error 46 (invalid
key specified). When an insertion or update is made to a file having an unopened
alternate-key file, the insertion (that is, the write) finishes successfully but does not
update the unopened alternate-key file.
Recovery. Corrective action is application dependent.
5 (%5) Sequential buffering not used because no

space was available in the PFS for the
sequential block buffer. (device type: 3)
Cause. Space was not available in the process file segment (PFS) for the sequential
block buffer or the file is not a structured disk file.
Effect. Sequential-block buffering is not used.
Recovery. Informative warning message only; no corrective action is needed.
6 (%6) A system message was received from another

process. (device type: 2)
Cause. The process received a system message from the operating system.
Effect. Data is returned in the application process buffer.
Recovery. Informative message only. The process receiving the system message
should read it (from $RECEIVE) to determine what action is needed.

2 -6
7 (%7) This process cannot accept a CONTROL,

CONTROLBUF, SETMODE, or RESETSYNC because
$RECEIVE is not opened correctly.
(device type: 0)
Cause. The process receiving the error message called CONTROL, SETMODE,
RESETSYNC, or CONTROLBUF for a server process file, but the server process did
not open its $RECEIVE file to enable receipt of these messages either with the
procedure OPEN parameter flags.<1> set to 1 or with the FILE_OPEN_ procedure
parameter options.<15> set to 0.
Effect. The procedure returns without performing the requested operation.
Recovery. Either have the server process open $RECEIVE with the OPEN parameter
flags.<1> set to 1 or FILE_OPEN_ parameter options.<15> set to 0, correct the file
parameter on the procedure call, or eliminate the call.
8 (%10) The operation was successful (examine MCW

for additional status).
(device type: 11.40, 11.42)
Cause. An operation to an EnvoyACP/XF data-communication line (synchronous data

link control or advanced data communications control procedures) finished
successfully.
Effect. The procedure returns additional status information in the message control
word (MCW).
Recovery. Retrieve information from the MCW before proceeding.
9 (%11) A read through a locked record was

successful. (device type: 3)
Cause. A “read-thru-locks with warning” mode was specified, and the record returned
was locked by another user.
Effect. The procedure returns the requested data and the error code.
10 (%12) The new record or file could not be created

because a file by that name or a record with
that key already exists. (device type: 3)
Cause. A process requested creation of a new disk file, insertion of a new record in a
file, or insertion of a new record with a unique alternate key in a structured file, but a
file by that name or a record with that key already exists.
Effect. The procedure sets the error code and returns without performing the
requested operation.

2 -7
11 (%13) The file is not in the directory or the

record is not in the file, or the specified
tape file is not on a labeled tape.
Cause. An operation referred to a nonexistent disk file or record, or the indicated tape
file was not on a labeled tape.
Recovery. Corrective action is application dependent; for “record not in file,” it
depends on the positioning mode.
12 (%14) The file is in use. (device type: any)
Cause. The specified file was in use, with exclusive or protected access, by another
process.
If a backup open is being performed and the file number is currently opened by the
backup process, the open operation returns this file-system error number though the
file was not in use.
13 (%15) The filename was not specified in proper

form. (device type: any)
Cause. The specified file name was not in proper form.

Recovery. Correct the file name.
14 (%16) That device does not exist on this system.

(device type: any)
Cause. The specified device or process did not exist on the system.
Recovery. Correct the device name or process name.

2 -8
15 (%17) The volume or system specified for a file

RENAME operation does not match the name of
the volume or system of the file.
(device type: 3)
Cause. The disk volume name or system number specified for a FILE_RENAME_ or
RENAME procedure did not match the name of the volume or system number on
which the given file resides.
Recovery. Correct the device name or process name.
16 (%20) No file with that file number has been

opened. (device type: any)
Cause. A file number was supplied to a file-system procedure call but no file with that
number was open.
Recovery. Open the file using either the FILE_OPEN_ procedure or the OPEN
procedure; then make your other calls such as FILE_GETINFO_ or FILEINFO.
17 (%21) A paired-open was specified and the file is

not open by the primary process, the
parameters supplied do not match the
parameters supplied when the file was opened
by the primary, or the primary process is
not alive. (device type: any)
Cause. A backup process attempted an OPEN or FILE_OPEN_ operation, or the

primary process attempted a CHECKOPEN or FILE_OPEN_CHKPT_ operation, but
either the file was not opened by the primary process, the parameters supplied did not
match the parameters supplied when the file was opened by the primary, or the
primary process was not alive.

2 -9
18 (%22) The system specified does not exist in the

network. (device type: any)
Cause. The specified system does not exist in the network; no connection has been
made to this system since the last cold load.
Recovery. Correct the system name, or make sure that the lines between the local
system and the referenced system are up.
19 (%23) There is no more space for devices in the

logical device table. (device type: any
except 1 and 2)
Cause. The logical device table was full, so the file system could not add the specified
device or process.
20 (%24) File name is too long to be opened across a

network, or swap file or system name is not
acceptable. (device type: any except 2)
Cause. An unconverted process tried to gain network access either to a process that
has a name of more than five characters or to a device that has a name of more than
seven characters. Error 20 also occurs if the opener has a process name of more than
five characters or a home terminal name of more than seven characters when using a
procedure that requires names in internal format. Error 20 can also indicate an
incorrect swap file or system designation.
Recovery. If the opener’s process name is too long, run it under a shorter name. If the
opener’s home terminal name is too long, either run the process with the TACL RUN
command specifying “TERM name”; run the application from a terminal with a shorter
name, rename the terminal in the SYSGEN configuration file, and regenerate the
system. This problem should not occur if you use the procedures introduced in the
D-series release (such as FILE_OPEN_ ), because they do not use names in internal
format.

2- 10
21 (%25) An illegal <count> was specified in a

file-system call, or the operation attempted
to transfer too much or too little data.
(device type: any except 2)
Cause. A process specified an invalid count parameter in a file-system call, or the

operation tried to transfer too much or too little data. Typical causes are:
• A FILE_CREATE_ or CREATE procedure call for a structured disk file contains an
invalid record length or alternate-key length.
• A FILE_CREATE_ or CREATE procedure call for a disk file requested a size larger
than the system limit.
• A file access specifies an inconsistent key length or compare length.
• A disk I/O request spans more than two extents, is greater than 4096 bytes, or
exceeds eight sectors, or a request under long transfer mode is not a multiple of
2048.
• For a 5520 printer request, the size of the direct-access, vertical-format file buffer
is not an even number or is greater than 508 bytes, a line length exceeds the
maximum transfer size for the current print mode, or a print line is too long for the
specified printer width.
• A data line written to the Spooler is too long.
Recovery. This is a coding error; corrective action is device dependent and
application dependent.
22 (%26) The application parameter or buffer address

is out of bounds. (device type: any)
Cause. A process specified an out-of-bounds parameter or a buffer address

parameter in a file-system call; that is, a pointer to the parameter or the buffer has an
address that is greater than the memory associated with the data area of the process
or that is in the part of the stack used by the file system.
Recovery. This is a coding error; corrective action is application dependent.

2 -11
23 (%27) The disk address is out of bounds, or the

maximum number of blocks in an alternate-key
file is exceeded. (device type: 3)
Cause. A disk address specified in a file-system call was too large or too small, or the
maximum number of blocks in an alternate-key file was exceeded. This error indicates
corrupt data or a corrupt alternate-key file.
Recovery. Corrective action is application dependent. However, you might need to
increase the workfile size.
24 (%30) Privileged mode is required for this

operation. (device type: any)
Cause. A nonprivileged user or process attempted to perform an operation requiring

privileged mode. The OPEN procedure returns this error when flags.<1> is set to 1
and the file being opened is not $RECEIVE. The FILE_OPEN_ procedure returns this
error when either options.<14> or options.<15> is set to 1 and the file being
opened is not $RECEIVE.
Recovery. If you intended to use the privileged feature, have the system manager
license the program file.
If a call to OPEN returned this error, check the flags parameter. If a call to
FILE_OPEN_ returned this error, check the options parameter.
25 (%31) AWAITIO[X] or CANCEL attempted on file

opened for waited I/O. (device type: any)
Cause. A process called AWAITIO[X] or CANCEL for a file opened for waited I/O.
Recovery. Either open the file for nowait I/O, correct the file number, or make another
correction appropriate to the application.

2- 12
26 (%32) AWAITIO[X] or CANCEL or CONTROL 22 attempted

on a file with no outstanding I/O requests.
(device type: any)
Cause. A process called AWAITIO[X], CANCEL, or CONTROL 22, but no I/O requests
were outstanding on the file; or a process called FILE_COMPLETE_ with a timeout
value of -1 (wait indefinitely) but no I/O had been initiated.
27 (%33) An operation was attempted with outstanding

no-waited I/O requests pending.
(device type: any)
Cause. A process tried a waited operation on a file that was opened for nowait I/O,
and outstanding nowait I/O requests were pending on that file.
(Note that some operations cannot be performed nowait, such as SETMODE,
POSITION, KEYPOSITION, or SETPARAM.)

2- 13
28 (%34) The number of outstanding nowait operations

would exceed that specified; an attempt was
made to open a disk file or $RECEIVE with
the maximum number of concurrent operations
more than 1; an attempt to ADD more than the
configured maximum number of subdevices for
an I/O process; or sync depth exceeds number
the opener can handle; or trying to run more
than 254 processes from the same object
file. (device type: any)
Cause. This error can occur for any of the following reasons:
• A nowait I/O request brought the number of outstanding nowait requests on the
given file to a value greater than the maximum nowait depth specified when the file
was opened.
• An attempt was made to open a disk file or the $RECEIVE file with a maximum
number of concurrent nowait operations greater than 1.
• An attempt was made to add more than the configured maximum number of
subdevices for an I/O process.
• The sync depth exceeds the number the opener is prepared to handle.
29 (%35) A required parameter is missing in a

procedure call, or two mutually exclusive
parameters were supplied.
(device type: any)
Cause. A file-system procedure call was missing a required parameter or supplied two
mutually exclusive parameters. The string:maxlen and actual-length
parameters must both be present if either one is present. If one of these two
parameters is present but the other is missing, this error is returned.
Recovery. This is a coding error; correct the call.

2- 14
30 (%36) Message system is unable to obtain memory.

Cause. This message indicates that the system has run out of one of the following
resources:
• No entry was available in the message block pool to perform the specified
operation.
• A process is already using its maximum number of RECEIVE message blocks.
• A process is already using its maximum number of SEND message blocks.
Recovery. Check the system for processes that use too many messages, or for
processes whose message limits are too low. (A process can change its limits by using
the CONTROLMESSAGESYSTEM procedure call.) Retry the operation. Try to
determine what system process or application process is causing the fault. If the
problem persists, contact your HP representative.
31 (%37) Unable to obtain file-system buffer space.

(device type: any)
Cause. This message is returned for a privileged operating system call. Insufficient
space was available in the process file segment (PFS) for a file-system buffer needed
to perform the specified operation.
Recovery. Check the program to see if it uses too much buffer space, opens too many
files, or uses too many DEFINEs. Try reducing the workload assigned to the process.
32 (%40) Unable to obtain storage pool space (SYSPOOL

or EXTPOOL). (device type: any)
Cause. This message is returned for a privileged operating system call. Insufficient
space was available for a control block needed to perform the specified operation
(perhaps due to exceptionally long control blocks).
Recovery. For an insufficient space error, wait, then try again; check the system for
processes that are using too much memory.
If the problem persists, contact your HP representative. Configuration information for
the current system image should be saved for your HP representative.

2- 15
33 (%41) I/O process is unable to obtain sufficient

buffer space. (device type: any except 2)
Cause. Insufficient buffer space was available for the I/O process. This message is
returned for a privileged operating system call.
Recovery. Wait, then try again; check the system for processes that are using too
much memory for I/O or applications attempting to queue very large I/O operations.
For persistent network errors, provide more buffer space for the network process. If
the disk process ran out of lock space, try to reduce the number of concurrent locks on
this disk process.
If the problem persists, contact your HP representative.
34 (%42) Unable to obtain a file-system control

block. (device type: any)
Cause. All file-system control blocks are in use. This message is returned for a
privileged operating system call.
Recovery. Wait, then try again. Check the system for processes that use too many
open files.
If the problem persists, the system has run out of some resource. Contact your service
provider.
35 (%43) Unable to obtain an I/O process control

block, or the transaction or open lock unit
limit has been reached.
Note. The above error description is displayed only on systems running J06.08 and earlier
J-series RVUs and H06.19 and earlier H-series RVUs.
35 (%43) Unable to obtain a lock control block, or

the transaction or open lock unit limit has
been reached. (device type: any except 2)
Note. The above error description is displayed only on systems running J06.09 and later
J-series RVUs and H06.20 and later H-series RVUs.

2- 16
Cause. All I/O process control blocks are in use, or a requester tried to acquire too
many record locks or file locks. This message is returned for a privileged operating-
system call.
Recovery. Wait, then try again. Check the system for processes that are performing
too many concurrent I/O operations, or rewrite the application to request fewer locks. If
the problem persists, contact your service provider.
36 (%44) Unable to lock physical memory; not enough

memory available. (device type: any)
Cause. Insufficient physical memory was available to perform the specified operation.
Recovery. Wait, then try again. If the problem persists, check the system for
processes that use too much locked memory; run MEASURE to help determine which
processes are locking memory.
37 (%45) I/O process is unable to lock physical

memory. (device type: any except 2)
Cause. Insufficient physical memory was available to perform the specified I/O
operation. This message is returned for a privileged operating system call.
Recovery. Wait, then try again. Check the system for processes that use too much
locked memory; run MEASURE to help determine which processes are locking
memory. If the problem persists, contact your service provider.
38 (%46) Attempt to perform operation on wrong type

of TNS system. (device type: any except 2)
Cause. Program running on a NonStop system specified an operation available only

on a NonStop 1+ system.
Recovery. Recode the application to eliminate the invalid operation.
39 (%47) The server process received a request with a

sync ID older than the set of saved replies.
(device type: 0)

2- 17
Cause. A sync ID was encountered that is older than the current sync ID minus the
sync depth.
Recovery. Recovery is application dependent.

2- 18
40 (%50) The operation timed out. AWAITIO[X] or

FILE_COMPLETE_ did not complete within the
time specified by its <time-limit>
parameter.
Cause. One of the following occurred:

• A call to AWAITIO[X] or FILE_COMPLETE_ did not finish within the time specified
by its timelimit parameter.
• A message was added to the caller’s queue (such as a BREAK message) while
the procedure PROCESS_DEBUG_ or DEBUGPROCESS was being used.
Effect. For AWAITIO[X], if 0D was specified for the time limit (completion check) or if -
1 was specified for the file number (wait for completion of an operation on any file), the
operation did not finish. If a nonzero time limit and a particular file were specified, the
operation timed out and was automatically canceled; it is considered completed.
For FILE_COMPLETE_, this error is only returned when a time limit other than -1D
(wait indefinitely) was specified and the operation did not finish.
The requested action might have been performed.
Recovery. Corrective action is application dependent and depends on whether a read
or a write was involved or a PROCESS_DEBUG_ or DEBUGPROCESS operation was
being used.
41 (%51) A checksum error occurred on a file

synchronization block. (device type: 3)
Cause. A file synchronization block was in error, probably because a user program
modified the file-system sync buffer area.
Recovery. Correct the coding error.
42 (%52) Attempt to read from unallocated extent.

(device type: 3)
Cause. A process tried a read operation when the file was positioned to an
unallocated disk address.
Recovery. Correct the coding error.

2- 19
43 (%53) Unable to obtain disk space for file extent.

(device type: 3)
Cause. The specified volume did not contain enough contiguous free space to permit
allocating an extent of the size needed for the file or for the directory.
Recovery. Purge unneeded files residing on the volume.
Determine the amount of available space and the number and size of fragments on the
volume by using the DSAP FREESPACE command or (on D-series releases only) the
LISTFREE command. If there is excessive fragmentation on the disk, have the
operator consolidate the available space by using the Disk Compression Program
(DCOM).
For a temporary workaround, specify a different swapvol in the RUN command.
44 (%54) The disk directory or DCT is full.

(device type: 3)
Cause. The disk directory is full. For a named process, the destination control table
(DCT) is full.
Effect. The procedure sets the error code and returns without adding the entry.
Recovery. For disk files, purge some files, then try again, or have the operator relabel
the disk to allow for more directory entries. The directory structure might allow you to
add new files whose names correspond to other areas of the directory structure where
there is still space.
For process files, the system might not create any newly named process until at least
one existing named process has stopped.

2- 20
45 (%55) The file is full; or two entries for

<process-name> were already in the PPD.
(device type: 3)
Cause. This error can occur for the following reasons:

• The file system could not add any more records to the given disk file.
• During a call to PROCESS_CREATE_ , NEWPROCESSNOWAIT, or
NEWPROCESS, two entries (both the primary and the backup) for process-
name were already in the process-pair directory, so no new entry could be added.
Effect. For unstructured disk files, as much data as possible is written.
Otherwise, the procedure sets the error code and returns without adding to the file.
Recovery. Either create a new file with larger extents and reload it, or increase the
current file’s maximum extents by issuing, for example, the FUP ALTER command with
the MAXEXTENTS option.
For a PROCESS_CREATE_ , NEWPROCESS, or NEWPROCESSNOWAIT call, use
another process name.
46 (%56) An invalid key was specified; key length

passed to CREATE exceeds 255 bytes; or
application failed to open an alternate-key
file. (device type: 3)
Cause. An operation on a structured file specified an invalid key, or the key length
passed to CREATE exceeds 255 bytes. The application might have failed to open an
alternate-key file (see file-system error 4).
Effect. The procedure sets the error code and returns without performing the specified
operation.
Recovery. Corrective action is application dependent. See the Enscribe Programmer’s
Guide for details.

2- 21
47 (%57) The alternate key data is not consistent

with primary file data. (device type: 3)
Cause. An error was encountered partway through a set of updates to primary and
alternate files. The file system tried to undo what had been accomplished thus far, but
it encountered another error.
Effect. The alternate key data is not consistent with the primary data; the file-system
sets the error code and returns.
All further attempts to make the primary and alternate files consistent are abandoned.
Recovery. If you are operating under a Transaction Management Facility (TMF)
transaction, abort the transaction.
If you are not operating under a TMF transaction, rebuild the alternate-key files with the
FUP LOADALTFILE command.
48 (%60) Security violation; illegal operation

attempted. (device type: 3)
Cause. The specified operation (read, write, execute, or purge) was invalid because of
the way the file was secured or because of an invalid or nonexistent password in an
Expand network environment.
Recovery. Resecure the file or recode the application; if access is across a network,
establish matching user IDs and remote passwords at both nodes.

2- 22
49 (%61) Access violation; attempt to use an

unexpired labeled tape for output; or a
mismatch between a DEFINE Use attribute
(input or output/extend) and the current
operation (read or write).
Cause. The type of access (read, write, or execute) specified by the calling process
was invalid on the given file for one of the following reasons:
• The calling process did not open the file for the type of access attempted.
• Another process had the file open in protected or exclusive mode.
• There was an attempt to use a labeled tape for output that had not expired.
• There was a mismatch between the DEFINE USE attribute and the current
operation on a CLASS tape DEFINE. The DEFINE USE attribute specifies how
the tape is to be used: for example, input, output, or extend.
50 (%62) Directory error on a disk volume.

(device type: 3)
Cause. A severe problem occurred with the directory on a disk volume. The file
associated with the error is no longer accessible.
Recovery. Your system operator might be able to recover the file.
51 (%63) Directory on a disk volume is marked bad.

(device type: 3)
Cause. A severe problem occurred with the directory on a disk volume. The file

2- 23
52 (%64) Error in the disk free space table.

(device type: 3)
Cause. A severe problem occurred with the accessed disk volume. The file
53 (%65) File system internal error or CP6100 file

management interface error.
(device type: 3)

• There is an internal problem in the file system or I/O system.
• A process replied to an open message with a file-system error code warning in the
range 1 through 9.
Effect. The procedure sets the error code and returns without continuing the
Recovery. If this error resulted from a process reply to an open message, correct the
process. Otherwise, contact your service provider. Note any console messages that
result from this error. Attempt to run a CMI or SCF TRACE if the problem is
repeatable.
54 (%66) I/O error in disk free space table or in DP2

undo area. (device type: 3)
55 (%67) I/O error in disk directory; the file is no

longer accessible. (device type: 3)
Cause. A severe problem occurred on a disk volume used by the file system. The file

2- 24
56 (%70) I/O error on disk volume label; the volume

is no longer accessible. (device type: 3)
Cause. A severe problem occurred on a disk volume used by the file system. The
volume associated with the error is no longer accessible.
57 (%71) The disk free space table is full.

(device type: 3)
Cause. A severe problem occurred on a disk volume used by the file system. There
might not be enough contiguous free disk space to enlarge the disk free space table.
requested operation. The file associated with the error is no longer accessible.
58 (%72) The disk free space table is marked bad.

(device type: 3)
59 (%73) The disk file is bad; there is a mismatch in

the internal FCB, or the file structure in a
structured file is inconsistent.
(device type: 3)
This error can occur during an attempt to open a file, or during an attempt to read or
write to a structured file. It might indicate that the file or table is broken.
If an application uses unstructured access to write to a DP2 structured file, it is
possible that future attempts to read or write to this file using unstructured access will
cause this error. User applications should not use unstructured access to write to a
structured file.
Another example of this error occurring follows a RESTORE operation on a key-
sequenced file that was backed up while it was being modified. The error eventually
occurs because the file is inconsistent.

2- 25
Error Occurs During File Open

Cause. The file you attempted to open is marked as corrupt. A utility command such
as FUP DUP, FUP LOAD, SQLCI DUP, SQLCI LOAD, or RESTORE has not finished.
requested operation. The file or table cannot be opened.
Recovery. Wait until the utility operation finished before attempting to open the file. If
the file is not currently open, then the utility abended before finishing; purge the file or
table.
Error Occurs During Read/Write Activity

Cause. The structured file or table has an inconsistent structure.
requested operation. The portion of the file or table with the inconsistent structure
cannot be accessed or updated.
Recovery. Examine the EMS event logs for BAD FILE DETECTED events. Run the
FILCHECK utility to determine whether the file is broken. If FILCHECK reports a
broken file or table, use the TMF subsystem or the RESTORE program to recover the
data. If TMF or RESTORE recovery is not available, and recovery is required, contact
your service provider for assistance.
60 (%74) The file resides on a removed volume, the

device is downed or not open, or a server
has failed and a process has been replaced
by a different process with the same name
since the server was opened.
(device type: any except 1 and 2)

• The specified file resides on a volume that has been removed from the system or
on a device that was brought down after the file was opened.
• The file is a device or process on a remote system whose link has gone down.
• If a server returned this error, the server received a message from a process that it
does not recognize as its opener. This can happen if a FILE_OPEN_ procedure
was used to open a file but the file name did not include the optional sequence
number.
• Certain sequences of DDL and DML operations can result in an error 60 (lost
open). Refer to the NonStop SQL/MP Reference Manual for more information on
opening SQL files.

2- 26
Recovery. Ensure that the device (if any) is up, close the file and reopen it, then try
again. If that is not successful, abort the operation and start over.
61 (%75) No more file opens are permitted on this

volume or device; the system operator issued
a STOPOPENS command or the number of open
files reached the maximum allowed.
(device type: 3)
Cause. The number of open files on this volume reached the maximum allowed, or the
system operator used a PUP STOPOPENS command (on D-series systems) or SCF
STOPOPENS (on G-series systems) to stop any more files from being opened on the
volume.
Recovery. Retry when files can be opened.
62 (%76) The volume was mounted but no mount order

was given, so the file open is not
permitted. (device type: 3)
Cause. The specified disk volume was physically mounted, but the system has not
received a mount request.
Recovery. Retry the operation after issuing a mount command.
63 (%77) The volume was mounted and the mount is in

progress, so a file open is not permitted.
(device type: 3)
Cause. The specified disk volume was physically mounted and the mount command
has been given, but the mount has not finished.
Recovery. Retry after the mount finishes.
64 (%100) The volume was mounted and the mount is in

progress, so a file open is not permitted.
(device type: 3)
Cause. A file open operation was tried while a mount operation was in progress.

2- 27
Recovery. Retry after the mount completes.
65 (%101) Only special requests permitted--or special

request to mirrored disk pair attempted with
only one device in special state.
(device type: 3)

• A system operator designated the specified disk volume for special request mode.
• A read or write operation with the special bit set was attempted on a mirrored-disk
pair that has one device in the up state, the other device in the special state.
Recovery. Retry after the volume is available for normal operations.
66 (%102) The device is downed, the LIU is not yet

downloaded, a hard failure occurred on the
controller, the disk and controller are not
compatible (DP1/DP2), or both halves of a
mirrored disk are down. (device type: any
except 2)

• The system operator brought down the specified device.
• A hard error occurred on the device controller.
• A path failure error in the range {210:226} occurred in a network.
• A DP2 disk process discovered a DP1 volume label or other label consistency
problem.
• Both halves of a mirrored disk are down.
If a path failed but other lines are active, the network brings down the defective line.
If there is a DP2-DP1 label consistency problem, the disk goes down, but the system
does not freeze.
If the problem occurs just after a cold load, $SYSTEM goes down, and then the
monitor process cannot open the OSIMAGE file and the operator process cannot open
the log file. Console messages to this effect are issued and, a minute later, the system
freezes with code %002002.
Recovery. If there is a DP2-DP1 consistency problem, correct the problem.

2- 28
For other devices, retry the failed operation after the device comes up. If the problem
recurs, notify your system operator.
70 (%106) Continue the file operation.

(device type: 0)
Cause. To indicate that the process file is ready to receive ASSIGN and PARAM
messages, the receiving process replies with this error after a write of the command
interpreter startup message.
Effect. The initial write of the startup message is completed.
Recovery. Continue startup sequence. Send the ASSIGN and PARAM messages if
appropriate.
71 (%107) A duplicate record was encountered.

(device type: 3)
Cause. The specified structured file record is a duplicate.

72 (%110) An attempt was made to access an unmounted

or nonexistent partition, or to access a
secondary partition. (device type: 3)
Cause. The specified operation on a structured file tried to access a nonexistent

partition or a partition to which access is invalid. For instance, if $B.X is the secondary
partition of $A.X, then “FUP COPY $B.X” fails with this error.
73 (%111) The disk file or record is locked.

(device type: 3)
Cause. The operation requested access to a locked disk file or record. This error
occurs only if the calling process accessed the file in alternate locking mode.
Recovery. Wait, then retry.

2- 29
74 (%112) Number of READUPDATES without replies

exceeds <receive-depth>, or
ACTIVATERECEIVETRANSID or REPLY called with
an invalid <message tag>. (device type: 2)
Cause. This error can occur for two reasons:

• A process called READUPDATE to read a message from $RECEIVE, but the
number of outstanding messages read but not replied to equals the receive depth.
• A process called ACTIVATERECEIVETRANSID or REPLY either with an invalid
message tag or when no outstanding message exists.
75 (%113) Requesting process has no current process

transaction identifier. (device type: 3)
Cause. The requesting process tried to access an audited file but had no current
transaction identifier, or a process called ACTIVATERECEIVETRANSID,
ENDTRANSACTION, or ABORTTRANSACTION but the message specified by the
message-tag has no associated transaction identifier.
Recovery. See the NonStop TM/MP Operations and Recovery Guide.
76 (%114) Transaction is in the process of ending.

(device type: 3 or none)
Cause. The transaction was ending so it could not be aborted or resumed.

78 (%116) Transaction identifier is invalid or

obsolete. (device type: 3 or none)
Cause. The transaction identifier was invalid or obsolete, or it is no longer available

because the Transaction Management Facility (TMF) subsystem removed it from the
system.

2- 30
79 (%117) A transaction attempted to update or delete

a record which it has not previously locked.
(device type: 3)
Cause. The transaction failed to lock a record before attempting to change or delete it.
80 (%120) Invalid operation on audited file or

non-audited disk volume. (device type: 3)
Cause. An invalid operation was attempted on an audited file or a nonaudited disk

volume.
81 (%121) Operation is not valid for a transaction

which still has nowait I/Os outstanding on a
disk or process file. (device type: 3 or
none)
Cause. The operation was invalid because the transaction had one or more
outstanding nowait I/O operations on a disk or process file. Error 81 is received by
ENDTRANSACTION or REPLY.
82 (%122) TMF not running on this system or on the

remote system. (device type: 0, 3, or
none)
Cause. BEGINTRANSACTION failed for one of the following reasons:

• The Transaction Management Facility (TMF) subsystem was not running on this
system.
• An I/O operation to an audited disk or a process file on a remote system was part
of a TMF transaction, but the TMF subsystem was not running on the remote
system.

2- 31
Recovery. Ensure that the TMF subsystem is running on all systems involved in the
transaction, then retry the command.
83 (%123) Attempt to begin more concurrent

transactions than can be handled.
(device type: none)
Cause. BEGINTRANSACTION failed because the process reached its maximum

number of concurrent transactions. The maximum is one transaction if the TFILE is
not open; otherwise, the maximum equals the TFILE depth.
84 (%124) TMF has not been configured on this system

or on the remote system. (device type: not
applicable)
Cause. BEGINTRANSACTION failed for one of the following reasons:

• The Transaction Management Facility (TMF) subsystem was not configured on this
system.
of a TMF transaction, but the TMF subsystem was not configured on the remote
system.
Recovery. Wait, then retry. Make sure the TMF subsystem is configured on all
systems involved in the transaction.
85 (%125) A device has not been started for TMF.
Cause. A device has not been started for the Transaction Management Facility (TMF)
subsystem.
Recovery. Ensure that devices are enabled for the TMF subsystem. For example, use
the command ENABLE VOLUMES through TMFCOM.

2- 32
86 (%126) BEGINTRANSACTION is disabled either by the

operator or because one or more TMF limits
have been reached.
(device type: not applicable)
Cause. BEGINTRANSACTION failed due to explicit action by an operator or because

one or more audit trails have reached maxfiles, the maximum number of audit files
permitted.
Recovery. Contact the operator. If operator intervention caused the failure, have the
operator correct the problem.
If one or more audit trails have reached maxfiles, dump audit trails until the
maximum is no longer exceeded.
87 (%127) Waiting on a READ request and did not get

it. (device type: 10 and 60)
Cause. A subdevice that was waiting on a read request received an I/O request that
was not a read request.
88 (%130) A CONTROL READ is pending so a second READ

is not valid. (device type: 10 and 60)
Cause. A two-step read (CONTROL 22) was queued at the subdevice, so a second
two-step read was invalid.
89 (%131) A remote device cannot accept text because

it has no buffer available.
(device type: 10)
Cause. The remote device cannot accept text because it does not have an available
buffer.
Recovery. Retry the write from the application process.

2- 33
90 (%132) The transaction was aborted by the system

because its parent process died, a server
using the transaction failed, or a message
to a server using the transaction was
cancelled. (device type: 3 or none)
Cause. The transaction aborted because either its BEGINTRANSACTION process

failed, a server using the transaction failed, or a message to a server using the
transaction was canceled.
Recovery. See the NonStop TM/MP Application Programmer’s Guide.
91 (%133) A TMF crash occurred during commitment of

the transaction; the transaction may or may
not have been committed. (device type: 3)
Cause. A serious internal error occurred on a system that runs the Transaction
Management Facility (TMF) subsystem.
Recovery. Contact your service provider.
92 (%134) Distributed transaction aborted by system

because the path to a remote system that was
part of the transaction was down.
Cause. The transaction aborted because the path to a remote system that participated
in the transaction was down.
requested operation. The transaction is aborted.
93 (%135) A transaction was aborted because it spanned

too many audit trail files.
Cause. The transaction was aborted because audit information filled 45% of the
master audit trail since the transaction started. The Transaction Management Facility
(TMF) subsystem reserves audit trail space in case the transaction must be completely
backed out. If this transaction were allowed to continue past the 45% threshold, the
audit trail could fill to the begin-transaction-disable level and new transactions could
not start.

2- 34
requested operation. The transaction is aborted.
Recovery. If this is a normal transaction that requires additional audit trail space to
complete, consider temporarily increasing the audit trail capacity by increasing the
number of audit trail files per volume, or by adding another active audit volume. Restart
the transaction. See the NonStop TM/MP Operations and Recovery Guide.
94 (%136) A transaction was aborted by operator

command. (device type: 3 or none)
Cause. An operator command aborted a transaction.

Recovery. For further information, see the NonStop TM/MP Operations and Recovery
Guide.
95 (%137) A transaction was aborted because of disk

process takeover by backup.
(device type: 3)
Cause. The system aborted the transaction because of a processor failure that
caused loss of access to a disk. This can be caused by:
• The disk volume itself down while there is a transaction outstanding against it.
• A volume cannot access its audit trail.
• The primary process of a participating disk process fails and the backup takes
over.
96 (%140) The transaction was aborted because it

exceeded the AUTOABORT timeout duration.
Cause. The call to the ENDTRANSACTION procedure failed because the transaction
was aborted by the Transaction Management Facility (TMF) autoabort function, which
automatically aborts transactions that run longer than a set amount of time.
Recovery. If this is a normal transaction that requires additional time to complete,
consider increasing the autoabort threshold. Do this only if you are sure that there is
enough space in the audit trail to accommodate the audit trail files that may be pinned

2- 35
while the transaction completes. Restart the transaction. For more information, see the
NonStop TM/MP Operations and Recovery Guide.
97 (%141) Transaction aborted by call to

ABORTTRANSACTION. (device type: 3 or none)
Cause. The transaction was aborted by a call to ABORTTRANSACTION.

98 (%142) Allocation of a Transaction Control Block

failed because the local table is full, or
the table on a remote system is full.
(device type: 0, 3, or none)
Cause. The transaction could not be started because:

• The local Transaction Management Facility (TMF) network active transmissions
table was full.
of a TMF transaction, but the TMF network active transactions table on the remote
system was full.
99 (%143) Process attempted to use features of a

microcode option that is not installed on
this system. (device type: any except 2)
Cause. A process tried to use features of a microcode option not installed in the
system.
If this error is returned after an operation involving a TAPECATALOG DEFINE, the
specific cause of the error is that the operation tried to use a CLASS TAPECATALOG
DEFINE when this class of DEFINE was not supported.
If the cause of the error is that the operation tried to use a CLASS TAPECATALOG
DEFINE, the tape request fails.
Recovery. Ensure that the system has the required microcode.
If the cause of the error is that the operation tried to use a CLASS TAPECATALOG
DEFINE, you must either use a CLASS TAPE DEFINE in the operation or enable

2- 36
TAPECATALOG DEFINEs. See the DSM/Tape Catalog User’s Guide for more
information.
100 (%144) Device is not ready or the controller is not

operational. (device type: any except 2)

• The device was not powered up or was not online.
• A tape drive was accessed while rewinding.
• A tape drive is at a load point but not online.
• A heavily loaded processor received a call to open a server process but could not
respond.
• The printer was out of paper.
Recovery. Make the device ready. In the case of a heavily loaded processor, repeat
the call to open the process.
101 (%145) The tape is write protected.

(device type: 4)
Cause. The system could not write to the mounted tape because the tape is not write
enabled.
Recovery. Verify that the correct tape is in use. The method of write-enabling a tape
varies depending on the specific hardware. You may need to insert a write ring or set
a write protect tab on the tape volume, or you may need to change a write setting on
the tape drive.
102 (%146) Printer paper out, bail open or end of

ribbon. (device type: 5)
Cause. A printer could not continue because it was out of paper or because the paper
bail was not in place.
Recovery. Load more paper, close the bail, or replace the ribbon as needed.

2- 37
103 (%147) Disk not ready due to power failure.

(device type: 3)
Cause. The disk device was not ready because the system power failed.
Recovery. Retry the operation.
104 (%150) No response from printer.

(device type: 5.4)
Cause. The printer did not return the requested status; either the printer power was off
or a hardware problem occurred.
Recovery. Power up the device or repair it.
105 (%151) Invalid printer VFU buffer.

(device type: 5.4)
Cause. The printer direct-access vertical-format unit buffer was invalid. This error can
occur for the following reasons:
• More than one stop was defined for channel 0 (top of form).
• No stops were defined for one or more channels.
• Bits <12:15> of each word were not zeros.
Recovery. Correct the programming error.
106 (%152) A buffered WRITE has failed; data in printer

buffer was lost. (device type: 5.4)
Cause. The write operation failed, and the data in the printer buffer was lost.
Recovery. The data must be generated again before it can be printed.

2- 38
110 (%156) Only BREAK access is permitted.

(device type: 6 or 61)
Cause. The calling process could not access the specified terminal because the user
pressed BREAK, and the process had specified break mode.
Effect. The procedure sets the error code and returns without transferring any data.
Recovery. Unless the calling process uses SETMODE to indicate that its operations
have break access, the terminal is inaccessible until the process processing the break
calls SETMODE function 12.
If the calling process did not enable break, retry the operation, delaying at least 10
seconds between retries.
If the calling process enabled break, check $RECEIVE for the system break message
and take appropriate action.
111 (%157) Operation aborted because of BREAK.

(device type: 6 or 61)
Cause. The file-system procedure aborted terminal access because the user pressed
BREAK before the current operation finished.
Effect. The procedure sets the error code and returns without transferring any data.
Data might have been lost.
Recovery. If this process did not enable break and the error occurred during a write
operation, retry the operation, delaying 10 seconds between retries.
If this process did not enable break and the error occurred during a read operation,
recovery is application dependent.
Note. If more than one process is accessing a terminal when break is used, only break access
is allowed after break is entered; subsequent retries are rejected with error 110 until normal
access is permitted.
If this process enabled break, check $RECEIVE for the system break message and
take appropriate action.

2- 39
112 (%160) READ or WRITEREAD preempted by operator

message (device type: 6) or too many user
console messages (device type: 1).
Cause. If the device is a terminal, an operator message has preempted the requested
READ or WRITEREAD operation.
If the device is the operator console, its internal buffer is full.
This error occurs only when an application process is using a terminal that is also
being used as the console device.
requested operation. Any data being entered when the preemption took place is lost.
Recovery. If the device is a terminal, send a message telling the terminal operator to
retype the last entry, then retry the read.
If the device is the operator console, wait, then retry.
113 (%161) DEFINE class or attributes are not valid for

the attempted function. (device type: 4)
Cause. A DEFINE was used in a manner that was not consistent with the class of the
DEFINE or was not compatible with the attribute values of the DEFINE.
If this error is returned after an operation involving a TAPE DEFINE, the specific cause
of the error may be that the operation tried to use a CLASS TAPE DEFINE when this
class of DEFINE was not supported.
If this error is returned after an operation involving a TAPE DEFINE or TAPECATALOG
DEFINE, the specific cause of the error may be one of the following:
• The DENSITY attribute was used with a specified DEVICE attribute that does not
support open reel tape volumes.
• The TAPEMODE attribute was used with a specified DEVICE attribute that does
not support cartridge tape volumes.
• Both the DENSITY and TAPEMODE attributes were specified.
Effect. The requested operation is ignored.
If the operation tried to use a CLASS TAPE DEFINE or CLASS TAPECATALOG
DEFINE, the tape request fails.
Recovery. Examine the DEFINE class and attributes as well as the procedure
parameters. See the Guardian Programmer’s Guide for an explanation of DEFINE
classes and a discussion of how to assign values to DEFINE attributes and how to use
DEFINEs in procedure calls. Change the DEFINE class or attributes or change the
procedure parameters as appropriate.

2- 40
If the cause of the error is that the operation tried to use a CLASS TAPE DEFINE when
this DEFINE class was not supported, you must either use a CLASS TAPECATALOG
DEFINE in the operation or enable TAPE DEFINEs. See the DSM/Tape Catalog User’s
Guide for more information.
If the error is due to an incompatibility among the DENSITY, TAPEMODE, and DEVICE
attributes of a TAPE DEFINE or TAPECATALOG DEFINE, you must change the
DEFINE in accordance with the following rules.
• Specify the DENSITY attribute only when the tape drive to be used supports open
reel tape volumes.
• Specify the TAPEMODE attribute only when the tape drive to be used supports
cartridge tape volumes.
• If a particular tape drive is to be used, specify the DEVICE attribute using a tape
drive that matches the DENSITY or TAPEMODE attribute.
• Never specify both the DENSITY and TAPEMODE attributes in the same DEFINE.
114 (%162) X25 Network problem - RESTART FAILURE.

(device type: 61)
Cause. The line will not come up.

Effect. X.25 aborts the request; an event message is sent to the EMS collector
process.
Recovery. Fatal error; no recovery is possible. Determine the error from a trace of the
code and try again. If the problem is still not apparent, submit the trace, log file,
CONFLIST, subunit, and line configuration to your service provider.
115 (%163) X25 Network problem - RESET FAILURE.

(device type: 61)
Cause. A reset attempt timed out.

Effect. X.25 aborts the request; an event message is sent to the EMS collector
process.
Recovery. Fatal error; no recovery is possible. Determine the error from a trace of the
code and try again. If the problem is still not apparent, submit the trace, log file,
CONFLIST, subunit, and line configuration to your service provider.

2- 41
File-System Errors
119 (%167) Error code value was too large to fit into
an 8-bit buffer; file-system error number
is greater than 255.
Cause. The file-system error number is greater than 255 for NEWPROCESS or
NEWPROCESSNOWAIT. SeeSection 5, NEWPROCESS AND
NEWPROCESSNOWAIT Errors, for an explanation of the information returned in the
errinfo parameter.
Effect. File-system error numbers greater than 255 can be used when a buffer larger
than 8 bits is available.
Recovery. Determine the actual file-system error and take the appropriate action.
Corrective action is application dependent. For NEWPROCESS and
NEWPROCESSNOWAIT, these file-system error numbers are returned in the
errinfo parameter.
120 (%170) Data parity error, or attempt to access a

tape whose density is higher than the switch
setting on the tape drive.

• A hardware data-parity error occurred and persisted through several retries of the
operation.
• The tape density is higher than the tape drive allows.
Effect. On a disk or tape READ or READUPDATE, the procedure returns as much
(invalid) data as possible. You can determine the number of characters returned by
checking the procedure count-read parameter if the procedure is a wait read or the
count-transferred parameter of AWAITIO[X] if it is a nowait read.
Recovery. If the device is a tape drive, clean the tape head and retry the operation. If
this action fails, try another tape.
Any other corrective action is device dependent and application dependent.
121 (%171) Data overrun error, hardware problem.

Cause. A hardware data-overrun error occurred and persisted through several retries
of the operation.
Recovery. Corrective action is device dependent. Contact your service provider.

2- 42
File-System Errors
122 (%172) Request aborted due to possible data loss

caused by reset of circuit, CLB sequence
error (device type: 6, 7, 9, 11, or 61); or
DP2 disk process takeover (device type: 3)

• A request aborted due to a possible loss of data caused by a circuit reset.
• There is a sequence error in the CIU-to-LIU bus (CLB). This is a data
communications error.
• There is a DP2 process takeover.
This error occurs only on unaudited, buffered files with a sync depth equal to 0. The
message is returned once for each open of the buffered file. This message also can
be returned on any call that supplies a file number except for CLOSE, which never
returns an error.
Recovery. Corrective action is device dependent and application dependent.
123 (%173) Subdevice is busy. (device type: 5, 6, or

10)
Cause. An operation was in progress on the specified subdevice.

Recovery. Corrective action is subdevice dependent and application dependent.
124 (%174) A line reset is in progress, loss of data is

possible. (device type: 6, 10, 51, 53, 59,
60, or 63)
Cause. A line reset was in progress for the subdevice or line, possibly causing a loss
of data.
Recovery. Corrective action is device dependent and application dependent. If a
data-communication process is involved, obtain a trace and save it for your service
provider.

2- 43
File-System Errors
130 (%202) Illegal disk address requested, or

formatting error occurred.
(device type: 3)
Cause. The requested address was too large for the disk space, or an error occurred
during formatting.
Recovery. Correct the coding error or reformat the disk.
131 (%203) Write-check error from disk; internal

circuitry fault. (device type: 3.0 or 3.1)
Cause. The disk hardware detected an internal circuitry fault.

Recovery. Notify your system operator.
132 (%204) Seek incomplete from disk; cylinder address

not reached after retry. (device type: 3.0
or 3.1)
Cause. The disk read-write heads did not reach the desired cylinder address after a
retry.
133 (%205) Access not ready on disk; cylinder address

not reached. (device type: 3.0 or 3.1)
Cause. The disk read-write heads did not reach the desired cylinder address.

2- 44
File-System Errors
134 (%206) Address compare error on disk.

(device type: 3)
Cause. A header search failure or header miscompare occurred on the disk. This
error indicates either a request for a bad address or, possibly, a head alignment or
formatting problem.
135 (%207) Write-protect violation with disk write.

(device type: 3)
Cause. An attempt was made to write to a write-protected disk.

Recovery. Reset the write-protect switch to allow writes.
136 (%210) Disk unit ownership error (dual-port disk).

(device type: 3)

• The driver tried to take ownership of the drive through the controller, and the
attempt failed.
• An incorrect controller address was passed to the disk process.
Effect. The disk process automatically retries.
Recovery. If the automatic retry is not successful, retry the operation yourself.
137 (%211) Controller buffer parity error.

Cause. A parity error occurred in the controller buffer.


2- 45
File-System Errors
138 (%212) Interrupt overrun; a device interrupted the

processor before the software could respond.
Cause. The device interrupted the processor before the software could respond.
Recovery. Wait, then try again. If retries do not succeed, notify your system operator
since a disk drive could go down.
139 (%213) Controller error; internal diagnostic

failure. (device type: any except 2)
Cause. The controller failed its internal diagnostics.

Recovery. Notify your system operator, or contact your service provider.
140 (%214) Modem error (communication link not yet

established, modem failure, momentary loss
of carrier, or disconnect) or FOX link to an
EXPAND line handler is down.
(device type: 6, 7, 10, 11, 12, 59, 60, 61,
63)
Cause. The specific meaning of this error depends on the subsystem that returned it.
Causes can include:
• The communications link is not yet established
• A modem failure occurred
• A momentary loss of carrier occurred
• The modem or link is disconnected
• The interprocessor bus monitor process ($IPB or for TorusNet configurations,
$IPBn, where n is a number in the range of 1 through 4) reported that the FOX link
or TorusNet vertical link to an Expand process is down
• A subunit or logical unit is not in the started condition
Refer to the appropriate subsystem manual for additional information.

2- 46
File-System Errors
Recovery. Corrective action is device dependent. If the problem is not apparent,

submit the trace, log file, CONFLIST, subunit, and line configuration to your service
provider.
148 (%224) Attempt to read unwritten data or to over-

write existing data on an optical disk.
(device type: 3.56)
Cause. There was an attempt to write to an optical disk where data was already
written, or there was an attempt to read data from a location on an optical disk where
there was no data.
Effect. The procedure sets the error code and returns.
150 (%226) End-of-tape marker detected.

(device type: 4)
Cause. The tape unit encountered the end-of-tape marker in the forward direction
during this operation, or it passed the marker on a previous operation and has not yet
passed it in reverse.
Effect. The last operation is completed normally, but very little tape is left.
Recovery. Informative message only; no corrective action is needed. If this error
occurs at the beginning of several tapes, contact your service provider regarding a
possible hardware problem.
151 (%227) Runaway tape detected, or attempt to access

a tape whose density is lower than the
switch setting on the tape drive.
(device type: 4)
Cause. The tape drive detected a runaway tape. This error usually occurs when the
system tries to read a blank tape or a tape written at a density lower than the tape drive
can read.
Recovery. Check the tape and replace it if it is faulty. If a read fails, change the
density switch setting or mount the tape on a tape drive that can read the tape at the
density at which it was written. If this error occurs at the beginning of several tapes,
contact your s regarding a possible hardware problem.

2- 47
File-System Errors
152 (%230) Unusual end (device type: 4 or 11)--tape

unit went offline or CP6100 file closed
while requests still in progress
(device type: 11).
Cause. The device controller returned an “unusual end” status. If the device is a tape
unit, the tape drive was taken offline manually during this operation.
Effect. Part of the requested operation might be complete.
Recovery. Take corrective action appropriate to the device. If this error occurs with
several tapes, contact your service provider regarding a possible hardware problem.
153 (%231) Tape drive power restored.

(device type: 4)
Cause. The tape-drive power failed and was restored.

Effect. The tape unit automatically returns to the ready state and to the beginning of
the tape.
Recovery. If two units are set to the same unit number when the power is restored,
the wrong unit might be placed online. To prevent this problem, ensure that all
application programs perform a rewind and unload operation when finished with a tape
unit.
154 (%232) BOT detected during backspace files or

backspace records. (device type: 4)
Cause. The tape unit encountered the beginning-of-tape (BOT) marker during a
backspace-files or backspace-records operation.
Effect. Tape motion stops.
155 (%233) Only nine-track magnetic tape allowed on

this system. (device type: 4)
Cause. A seven-track tape device was specified for an operation requiring a nine-track
tape (most HP subsystems require a nine-track tape).
Recovery. Retry, specifying a drive configured for a nine-track tape.

2- 48
File-System Errors
156 (%234) Tape command rejected, TIL or TLAM protocol

violation detected, or X25AM level-4 ITI
protocol violation. (device type: 4, 12,
or 56)

• The tape command was rejected.
• The procedure made a connect request (CONTROL 11) after the Tandem-to-IBM
Link (TIL) was already connected or after an internal link error occurred.
• The TLAM process has entered an unrecoverable state.
For a TLAM error, a trace or processor dump is required to determine the exact nature
of the problem.
Effect. The procedure aborts the call and disconnects the link. For a tape, no
operation is performed.
Recovery. For a rejected tape command, check that you indicated a valid command.
Inform your system manager that the tape controller is probably unloaded.
For a superfluous connect request, correct the coding error.
For an internal link error, contact your system operator.
157 (%235) I/O process internal system error.

Cause. An internal system error occurred.

Recovery. Contact your system operator.
158 (%236) Invalid function code requested for Hyper

Link. (device type: 26)
Cause. An operation specified an invalid Hyper Link function code.

Recovery. Contact your system operator or supply the correct function code.

2- 49
File-System Errors
159 (%237) Device mode wrong for Hyper Link I/O

request, or attempt to execute PUP PRIMARY
command while the tape drive is waiting for
a labeled tape to be mounted.
(device type: 26)

• The Hyper Link I/O process received a test request, but the I/O process was not in
the test mode of operation.
• An attempt was made to issue a PUP PRIMARY command (on D-series releases)
or an SCF PRIMARY command (on G-series releases), but the tape drive was
waiting for a labeled tape to be mounted.
Recovery. Request the test mode of operation prior to issuing test requests, or mount
a labeled tape and retry the operation.
160 (%240) Request is invalid for device state;

protocol error. (device type: 6, 7, 10,
11, 51, 60, or 63)
Cause. A protocol error occurred. An Expand process received a PUP PRIMARY

request (on D-series releases) or an SCF PRIMARY request (on G-series releases),
but the backup process was not running or was misconfigured, or the Expand process
could not transfer ownership to its backup for some reason.
Recovery. Corrective action is device dependent. If this problem persists and a data
communications process is involved, obtain a trace and submit it to your service
provider.
161 (%241) Impossible event occurred for line state.

(device type: 7, 10, 11, or 60)
Cause. An event occurred that was impossible for the current line state; this error
probably indicates a hardware problem.
provider.

2- 50
File-System Errors
162 (%242) Operation timed out. (device type: 6, 7,

10, 11, or 60)
Cause. The specified operation timed out after several retries.

provider.
163 (%243) EOT received (device type: 7.0, 7.1, 7.2,

7.3, or 7.8) or power at autocall unit is
off (device type: 7.56 or 11).
Cause. Either an end-of-transmission (EOT) message was received while the

procedure was waiting for a line bid or for a message, or the power at the auto-call unit
was off.
provider.
164 (%244) Disconnect received (device type: 7.0, 7.1,

10, 11, 61, or 63) or data line is occupied
(device type: 7.56 or 11).
Cause. Either a disconnect was received, a send disconnect call was issued while a
request was outstanding, or the data line was busy. This error message also occurs
when the specified number of retries for an Expand send message are used up.
Recovery. Corrective action is device dependent.

2- 51
File-System Errors
165 (%245) RVI received (device type: 7.0, 7.1, 7.2,

or 7.3) or data line not occupied after
setting call request (device type: 7.56 or
11).
Cause. A reverse interrupt (RVI) was received, or the data line was not occupied after
setting the call request.
provider.
166 (%246) ENQ received (device type: 7.0, 7.1, 7.3,

or 7.9) or auto call unit failed to set
present-next-digit (device type: 7.56 or
11).
Cause. An inquiry (ENQ) was received, or the auto-call unit failed to set “present next
digit.”
provider.
167 (%247) EOT received on line bid (device type: 7.0,

7.1, 7.3, or 7.8), or data-set-status not
set (device type: 7.56 or 11).
Cause. Either an end-of-transmission (EOT) message was received in response to a

line bid or selection, or “data set status” was not set after dialing all digits.
provider.

2- 52
File-System Errors
168 (%250) NAK received on line bid (device type: 7.0,

7.1, 7.3, or 7.8), or auto-call unit failed
to clear present-next-digit after digit-
present was set (device type: 7.56 or 11).
Cause. Either a negative acknowledgment (NAK) was received in response to a line

bid or selection, or the auto-call unit failed to clear “present next digit” after “digit
present” was set.
provider.
169 (%251) WACK received on line bid

(device type: 7.0, 7.1, or 7.3), auto-call
unit set abandon-call-and-retry
(device type: 7.56 or 11), or station
disabled or undefined (device type: 11).

• A wait for acknowledgment (WACK) was received in response to a line bid or
selection.
• The auto-call unit set “abandon call and retry.”
• The specified station was disabled or undefined.
provider.
170 (%252) No ID sequence received during circuit

assurance mode (device type: 7.0 or 7.1) or
invalid MCW entry number on write
(device type: 11.40).

• No ID sequence was received.
• A write request for a multipoint line had an invalid message control word (MCW)
entry number or an MCW entry number different from the currently active output
entry number.

2- 53
File-System Errors
171 (%253) No response received on bid/poll/select, or

reply invalid. (device type: 7, 10, 11,
58, 60, or 61)
Cause. The selected controller or device did not respond.

provider.
172 (%254) Reply not proper for protocol; invalid

control sequence or invalid data.
(device type: 6, 7, 9, 10, 11, or 60)
Cause. The selected device responded with an invalid control sequence or invalid
data.
provider.
173 (%255) Maximum allowable NAKs received

(transmission error) (device type: 6, 7,
10, or 60), invalid MCW on WRITE, or invalid
request ID (device type: 11).
Cause. The specific meaning for this error is device dependent.

Recovery. Corrective action is device dependent. For a 6520, 6524, or 6530 terminal,
check that the power is on. If this problem persists and a data communications
process is involved, obtain a trace and submit it to your service provider.
174 (%256) WACK received or CLB frame aborted.

(device type: 11)
Cause. Either a wait-for-acknowledgment (WACK) sequence was received as the text

acknowledgment or a link request occurred while the request was pending.

2- 54
File-System Errors
requested operation. Data might be lost.
provider.
175 (%257) Incorrect alternating ACK received

(device type: 7.0, 7.1, 7.2, or 7.3), or
command rejected (device type: 11).
Cause. Either an incorrect alternating acknowledgment (ACK) was received or a

command reject condition was generated.
provider.
176 (%260) Poll sequence ended with no responder.

(device type: 7.3, 7.8, 7.9, or 11.40)
Cause. The poll sequence ended, but no message was received in response.
Recovery. Corrective action is device or application dependent.
177 (%261) Text overrun (insufficient buffer space for

data transfer). (device type: 7, 10, 11,
or 60)
Cause. The data received on a read exceeds the amount allowed by the read count.
Recovery. Corrective action usually involves increasing the read count; refer to the
manual for the specific device for more information. If this problem persists and a data
provider.
178 (%262) No address list specified.

(device type: 7.2, 7.3, 7.8, 7.9, 11.40, or
61)
Cause. An address list was required for this operation, but none was specified.

2- 55
File-System Errors
179 (%263) Application buffer is incorrect

(device type: 10 or 61), control request
pending, or autopoll active
(device type: 11.40).
Cause. For an AM3270, TR3271, or X25AM data-communication line, an error was

encountered in the application buffer, typically a bad WCC on packet type.
For an EnvoyACP/XF line, the operation could not be performed because a control
request was pending or the auto poll feature was active.
For a SNAX line, there was a cryptography failure.
180 (%264) Unknown device status received.

(device type: 5.3, 6.6 through 6.10, or 10)
Cause. An invalid device status was received and could not be translated into a
usable error number.
Recovery. Use the Subsystem Control Facility (SCF) to determine what status was
received.
181 (%265) Subdevice expected status information but

received data instead. (device type: 10)
Cause. Data was sent to the subdevice, but the subdevice expected status
information.
182 (%266) SNALU access method outbound RU error; see

SNALU error code in status area of SNAX
header. (device type: 14.2)

2- 56
File-System Errors
Cause. SNAX was unable to transmit an outbound response unit (RU), and the
detailed SNAX application logical unit (SNALU) error code is contained in the status
area (2 bytes) of the SNAX header.
183 (%267) SNA session has ended. (device type: 14.2)
Cause. The file-system procedure call issued by the primary or secondary logical unit
could not be performed because the logical-unit-to-logical-unit session associated with
the specified device (filenum) no longer exists.
Recovery. Close, then reopen the process.
187 (%273) Operation returning with no useful data.
Cause. CONTROL 26 was previously issued to complete all outstanding requests

from this application.
Effect. All pending and active READ, WRITE, and WRITEREAD requests are
completed, and no data is transferred.
188 (%274) Damage to logical flow of events.

Effect. The state of the session is suspect.
189 (%275) Response not yet available.

Effect. The requesting program concludes that the actions initiated by a READ,
WRITE, or WRITEREAD request are in progress, and the response that would have
been returned is not yet available.

2- 57
File-System Errors
Recovery. Recovery is application dependent. A generic read operation will retrieve

the response when the response is available.
190 (%276) Device error; hardware problem.

Cause. A hardware error occurred and persisted through several retries of the
operation. A problem exists with either the device or its controller. The error might be
caused by a termination condition that was not expected.
provider.
191 (%277) Device power on (device type: 5), or

terminal reset (device type: 6).
Cause. The device power was switched off, perhaps due to power failure, then was
switched back on during this operation.
192 (%300) Device in diagnose mode; system operator

running diagnostics. (device type: 3, 4, 5,
or 6)
Cause. The system operator was running diagnostic programs on the specified
device.
Recovery. Wait, then try again.
193 (%301) Invalid or missing microcode file.

Cause. The operating system could not do one of the following:

• Locate the microcode files for a downloadable controller
• Read either of the microcode files because of disk file errors
• Download from the microcode files because they are not formatted properly

2- 58
File-System Errors
requested operation. The operator message MICROCODE LOADING FAILURE
(console message 100, EMS operator message 100, or DISK operator message 5035)
is displayed twice, once for the primary file and once for the backup.
Recovery. See the Operator Messages Manual for information on correcting this error.
194 (%302) Device use or mount request rejected by

operator. (device type: 4)
Cause. A MEDIACOM REJECT TAPEMOUNT command (on G-series systems) was

issued to reject a mount request or to request use of a drive for an unlabeled tape.
Effect. The open fails.
195 (%303) Operation requires use of $ZSVR but it is

not running; tape operation is not allowed.
(device type: 4)
Cause. On a system with tape label processing enabled, the named process $ZSVR
was not running when a tape operation was requested to open a tape file.
Effect. No tape operation is allowed.
Recovery. Start $ZSVR. For additional information, see the appropriate system
operator’s guide.
196 (%304) A tape label record is missing or incorrect.

(device type: 4)
Cause. The label record on the tape volume was incorrect or missing or a DEFINE
attribute value did not match the value for that attribute recorded in the tape label.
Effect. Access to the tape file is denied.
Recovery. Check that the correct tape was mounted. If a mismatch of DEFINE
attribute values occurred, modify the DEFINE attribute that did not match the tape
label. Use the MEDIACOM INFO TAPELABEL command (on G-series systems) to
dump the beginning of the volume group labels on the tape. Correct the label
accordingly. For additional information, see the appropriate system operator’s guide.
197 (%305) An SQL error has occurred.
Cause. A more detailed error explanation cannot be given because this system does
not support SQL objects.
If an Enscribe operation (such as PURGE, FUP ALTER, SETMODE, or FUP SECURE)
is attempted on an SQL sensitive file, this error is returned.
Effect. The operation fails.

2- 59
File-System Errors
Recovery. The operation can only be performed if your system uses the NonStop
SQL/MP product.
198 (%306) A DEFINE of the given name could not be

found. (device type: any)
Cause. The specified DEFINE name does not exist.

Effect. The request is ignored.
Recovery. Either correct the name or supply a DEFINE of the given name.
199 (%307) The disk file is SAFEGUARD protected.

(device type: 3)
Cause. An attempt was made to change a security attribute (security vector, license,
PROGID, and CLEARONPURGE) for a file protected by Safeguard.
Effect. No security attributes can be established for the file through the file system or
through FUP.
Recovery. Use SAFECOM to establish the security attributes.
200 (%310) The device is owned by an alternate port.

Cause. The logical ownership of the device was switched to the other processor to
which the device is configured.
Effect. The file system automatically retries the operation.
201 (%311) The current path to the device is down

(device type: any except 0 and 2), an
attempt was made to write to a non-existent
process (device type: 0), the
message-system request was incorrectly
formatted, or an error was found in the
message system interface.

• A request sent to a process or device on a remote system by way of the Expand
subsystem received this error because the Expand process on the remote system
could not deliver the request to the process or device.
• There was an incorrectly formatted message-system request.
• For a process file, either an attempt was made to write to a nonexistent process or
a pending WRITE or WRITEREAD was aborted because the server process read
the request using READUPDATE but died before it could reply.

2- 60
File-System Errors
• For a device, the current path to the device was down.

Effect. Either the operation never starts, the operation finishes but the path fails
before a reply can be made, or the processor fails.
For a process file opened with a sync depth of 0, the primary process fails if the
process file is a process pair.
Recovery. For a process file, retry the operation once to establish communication with
the backup process. If a second error 201 occurs, no backup exists (both paths are
down) and programmatic recovery is impossible.
For disk server processes, open these processes with a sync depth greater than 0.
Then take corrective action appropriate to the device and the application.
210 (%322) Device ownership changed during operation.

Cause. This error is associated with concurrent operations involving more than one
unit connected to a multiunit controller.
This error occurs when an operation is in progress in one unit on a multiunit controller
and an error is detected during an operation with another unit on the same controller.
(The other operation could have been on behalf of this or another application process.)
211 (%323) The processor performing the operation

failed during the operation.
(device type: any)
Cause. The processor module controlling the device associated with this file operation
failed (path error).
requested operation. The file operation itself stopped at some indeterminate point.
213 (%325) Channel data parity error (path error).

Cause. A controller or channel failure occurred (path error).

For a 3106 controller, a command could have arrived at the same instant that the
controller was busy processing a seek. However, if the problem is due to 3106
synchronization, this is an informative message only.
Effect. The file operation stopped at some indeterminate point.

2- 61
File-System Errors

For the 3106 controller, use the IOTRACE program to examine the I/O trace area of
the affected processor.
214 (%326) Channel timeout (path error).


215 (%327) I/O attempted to absent memory page

(hardware path error).
Cause. A hardware failure occurred (path error).

216 (%330) Memory breakpoint encountered during this

I/O operation. (device type: any except 2)
Cause. A memory-access breakpoint was encountered during this operation.

Recovery. If the breakpoint was not intentionally set, clear it using Debug. Note which
system call was used in the application and/or which system process is affected. If the
problem persists, contact your service provider.
217 (%331) Memory parity error during this I/O

(hardware path error).
Cause. A hardware failure occurred (path error).

218 (%332) Interrupt timeout occurred on a channel, or

a controller, modem, or the line between, or
lost the modem clock (path error).
Cause. A controller failure, channel failure, line disconnect between controller and
modem, or loss of modem clock occurred (path error).

2- 62
File-System Errors

219 (%333) Illegal device reconnect (path error).

Cause. An invalid device reconnect occurred (path error).

220 (%334) Protect violation; an I/O controller

attempted an illegal write.
Cause. An I/O controller tried to write to memory when it should not have done so.
221 (%335) Controller handshake violation (path error).


222 (%336) Bad channel status from EIO instruction

(path error). (device type: any except 2)

223 (%337) Bad channel status from IIO instruction

(path error). (device type: any except 2)

224 (%340) Controller error (fatal error).

(device type: any)

2- 63
File-System Errors
Cause. The controller has a fatal error that was recognized by its resident microcode.
Effect. All paths through the controller are down.
225 (%341) No unit assigned or multiple units assigned

to the same unit number (path error).
Cause. A path error occurred because no unit was assigned to the unit number or
because multiple units were assigned to the same unit number.
Recovery. Corrective action is device dependent and application dependent. If the
device is a disk, check the UNITS plugs on the drive connected to the controller.
230 (%346) Processor power failed, then restored.

Cause. The processor power failed and then was restored during this operation (path
error). At least one path, and possibly both paths, is operable.
231 (%347) Controller power failed, then restored.

Cause. The controller power failed and then was restored during this operation (path
error). At least one path, and possibly both paths, is operable.
Effect. The file operation stops at some indeterminate point.
Recovery. Corrective action is device dependent and application dependent. Check
the log for a channel reset (operator message 77) which can also cause a file-system
error 231 to occur (the power-on interrupt causes all controllers on the channel bus to
be cleared so they can continue processing).
232 (%350) Access is denied due to error in

communication with SMON.
Cause. There is a lack of Safeguard resource to support the request; this is an internal
error.
Effect. Access is not possible at this time.
Recovery. Wait, then try again. The number of retries is limited.
233 (%351) Error in call to SERVERCLASS_SEND_.

2- 64
File-System Errors
Cause. An error occurred in a call to SERVERCLASS_SEND_.

Effect. The call is not accepted.
Recovery. Correct the error in the call and try again.
240 (%360) Network line handler error; operation not

started. EXPAND performed an ownership
switch. (device type: 63)
Cause. The Expand process performed an ownership switch.

Effect. All traffic not yet sent to the remote system, and traffic that belongs to the
affected Expand process is terminated. The application should not see this error.
Recovery. The file system automatically retries. For additional information, see the
Expand Network Management Guide.
241 (%361) Network error; operation not started.

(device type: 63)
Cause. The network control process ($NCP) caused the Expand process to terminate
its unsent traffic for one of the following reasons:
• A connection to a remote system was established or reestablished over the same
path or a different path.
• The network control process came up (for example, after a crash or a takeover by
the backup).
• There was no longer a path to the remote system (console message 43, LDEV
ldev NET: CONNECTION LOST TO SYS nnn, is generated by the NCP).
Or, the Expand process encountered one of the following situations:
• The Layer 4 protocol detected a protocol error.
• An I/O power on occurred but all lines used by the Expand process were attached
to the controller for which the IOPON occurred.
• A processor power on occurred.
• The Expand process received a request from a system for which it is not the
current path.
• An Expand process received a request to forward a message, but all its lines are
down.
requested operation. The application should not see this error.
Recovery. The file system retries automatically. For additional information, see the
Expand Network Management Guide.

2- 65
File-System Errors
246 (%366) External cluster bypass error; operation

aborted. (device type: 63, subtype 3)
Cause. A message from one cluster to another bypassed the Expand process, went
directly to its destination, and terminated due to a disconnect or resynchronization of
the bus protocol between clusters.
Recovery. Retry the operation if possible. For additional information, see the Expand
Network Management Guide.
248 (%370) Network line handler error; operation

aborted. (device type: 63)
Cause. If this error was returned to the application, the Expand process performed an
ownership switch. The request, sent to its remote destination, is aborted because the
operating system does not know the state of the request.
If the error was returned in Expand or X25AM operator message 45, or in SNAX
operator message 23 or 24 (line not ready), the network process was unable to
establish level-2 communications or all level-2 retries were exhausted. Possible
reasons for this error are:
1. The other system was down
2. There was an incorrect NEXTSYS parameter in the configuration
3. A correctly configured Expand line was initially ready
4. There was garbled data or no data in or out
Effect. The procedure sets the error code and returns. The request to perform the
operation is sent but it might not have been performed.
Recovery. If the error is returned to the application, determine whether the operation
should be retried. If so, retry the request.
If the error is returned in an operator message, recovery is as follows for the
corresponding cause:
1. Select an alternate path if one exists. The state of the other system can be
determined by issuing an SCF INFO PROCESS $NCP, NETMAP command.
2. The Expand process is not usable until the correct NEXTSYS parameter is
supplied. Either correct the NEXTSYS parameter in the SYSGEN file and perform
another SYSGEN or use the Subsystem Control Facility (SCF) to alter the
NEXTSYS parameter and then bring the line up using SCF.

2- 66
File-System Errors
3. The line normally exhibits error 248 on the operator console when the line is
initially brought up. The message should be followed by another message
announcing that the line is up.
4. By performing successive SCF STATS commands (at one minute intervals) and
observing the U-FRAME counts, you can determine whether data is being
transmitted and received. The FCS error count indicates that garbled data was
received.
If the SCF STATS commands do not indicate both send and receive data, place the
local modem in analog loopback and observe the U-FRAME counts again. If both
send and receive counts are incrementing, the local controller and modem are
working properly. If both systems check out, first check the lines by using the
modem self test, then observe the U-FRAME counts with the remote modem in
digital loopback. If all tests indicate data is being transmitted and received, then a
trace at both ends of the line should indicate the cause of the not-ready condition.
If this problem persists, your service provider needs to run traces or perform other
corrective action. Refer to the Expand Network Management Guide for further
information.
249 (%371) Network error; operation aborted.

(device type: 63)
Cause. The network control process ($NCP) caused the Expand process to terminate
its traffic (already sent to its remote destination) for one of the following reasons:
• A connection to a remote system was established or reestablished over the same
path or a different path.
• The network control process came up (for example, after a crash or a takeover by
the backup).
• There was no longer a path to the remote system (operator message 43, LDEV
ldev NET: CONNECTION LOST TO SYS nnn, is generated by the network
control process).
Or, the Expand process encountered one of the following situations:
• The level-4 protocol detected a protocol error.
• An I/O power on occurred but all lines used by the Expand process were attached
to the controller for which the IOPON occurred.
• A processor power on occurred.
• The Expand process received a request from a system for which it is not the
current path.
• A Expand process received a request to forward a message, but all its lines were
down.

2- 67
File-System Errors
Effect. The procedure sets the error code and returns. The request to perform the
operation is sent but it might not have been performed.
Recovery. Corrective action is device dependent and application dependent. If this
problem persists, your service provider needs to run traces or perform other corrective
action. For additional information, see the Expand Network Management Guide.
250 (%372) All paths to the system are down.

(device type: 63)
Cause. The referenced system in the network is down, or it is not currently connected
to the system on which the requested process is running (path error).
Recovery. Corrective action is application dependent. If this problem persists, your
service provider needs to run traces or perform other corrective action.
251 (%373) Network protocol error (path error).

Cause. A network protocol error occurred (path error).

Recovery. Corrective action is application dependent. If sync depth is greater than 0,
keep trying until successful or until a different error is returned.
If this problem persists, your service provider needs to run traces or perform other
corrective action.
252 (%374) Required EXPAND class is not available.
Cause. The Expand class of transmission (such as FOX) is busy or is not available.
Effect. The message is not sent.
Recovery. Check for the availability of the required class and retransmit the message.
If this problem persists, your service provider needs to run traces or other corrective
action.
255 (%377) Net line handler flooded; too many

interrupts.
Cause. There were too many ownership errors. The network process was flooded.
Effect. All requests are aborted, and all lines go into the not-ready state. If all possible
combinations of ownership switches are unsuccessful, the lines go down.

2- 68
File-System Errors
Recovery. Check the operator messages to determine the cause of the ownership
errors. Correct the cause of the problem as required. If this problem persists, your
service provider needs to run traces or other corrective action.
300 (%454
through through Reserved for application
processes.
511 (except 538) %777)
These file-system error numbers are reserved for use by application processes.
538 (%) Extent size is greater than 65535 pages.
Cause. The CHECK^FILE procedure could not return the primary or secondary extent
value because the extent size is greater than 65,535 pages.
Effect. The operation is not executed.
Recovery. This is an informational message only; no corrective action is needed.
541 (%1035) A data structure version is incompatible

with the requested operation.
Cause. The version of a data structure does not support a requested operation.
Effect. The requested operation is not performed.
549 (%1045) Blockmode is not currently allowed on this

terminal.
Cause. The application attempted to perform a SETMODE 8 (with param1 greater

than or equal to 1), but block mode use was not allowed because the terminal was
being used as a console device.
Effect. The operation failed.
Recovery. If a block mode application needs to use the device, direct the console to
another terminal; the operator process will stop using the device and allow it to be used
in block mode.
550 (%1046) File operation attempted at illegal

position.
Cause. The current setting of the file’s positioning information is not valid for the
attempted operation. For example, an I/O was attempted under the large transfer
SETMODE while the current position was not a multiple of 2048; usually this condition

2- 69
File-System Errors
is caused when the last read before an end-of-file returned a number of bytes that is
not a multiple of 2048.
Recovery. Determine the reason for the incorrect position (for example, a bad
POSITION call or an unexpected read transfer count) and correct the problem.
551 (%1047) Duplicate exists for insertion-ordered

alternate key.
Cause. More than one insertion-ordered alternate key record exists.

Effect. Depending on the setting of a bit in the CREATE procedure or an item code in
the FILE_CREATELIST_ procedure, duplicate records with insertion-ordered alternate
keys are ordered by the value of the primary record key field or by the sequence in
which those records were inserted into the alternate key file.
560 (%1060) The function cannot be performed because one

of the processes involved cannot recognize a
PIN greater than 255.
Cause. The function cannot be performed because one of the processes involved
cannot run with a process identification number (PIN) greater than 255. A process
created by the D-series procedure PROCESS_LAUNCH_ or PROCESS_CREATE_
can request to be run at a high PIN (greater than 255), but a process created by the C-
series procedure NEWPROCESS or NEWPROCESSNOWAIT cannot run at a high
PIN.
Recovery. Any of the following can be performed to recover:
• One or more of the processes involved must be rerun with low PINs (PINs less
than 255).
• One or more of the processes involved must be changed to allow for high-PIN
requestors by setting HIGHPINREQUESTORS ON. A process that ran at a low
PIN continues to run at a low PIN but can now be opened by high-PIN processes.
Take this action only if the application can be opened by a high-PIN process.
• One or more of the processes involved must be changed to support high PINs.
561 (%1061) The item code in a list is not recognized.
Cause. The item code in a list was not recognized.

Effect. Either FILE_CREATELIST_ failed or FILE_GETINFOLIST_ or
FILE_GETINFOLISTBYNAME_ stopped at an item in error.

2- 70
File-System Errors
Recovery. The item code in error should be corrected.
563 (%1063) The size of an output buffer was too small.
Cause. The size of an output buffer was too small to hold the data. Error 563 is an
operating system error and should not be confused with error 177, which is a data
communications error.
Recovery. Increase the size of the buffer.
564 (%1064) The operation is not supported for this file

type.
Cause. The operation specified is not permitted on this file type.

Effect. The operation failed. Error 564 is returned rather than error 2 when an
unqualified failure occurs.
Recovery. Determine why an invalid operation was requested and correct the request.
565 (%1065) A malformed request was denied.
Cause. The server that received the message found the request malformed or not
appropriate under current conditions.
Recovery. The problem is probably on the requester side. Verify the conditions of the
requester; if they are satisfied, contact your service provider.
566 (%1066) This reply is malformed or not appropriate.
Cause. The requester found the reply from the server was malformed or not
appropriate under current conditions. The error is returned to the requester, not to the
reply procedure, because the problem is detected in the requester’s reply handling
(such as AWAITIO[X]).
Recovery. The problem is probably on the server side. Verify the conditions of the
server; if they are satisfied, contact your service provider.

2- 71
File-System Errors
567 (%1067) The define used is incompatible for use with

target system's TOS version.
Cause. A CLASS of DEFINE is being used in a context where it is not supported by

the operating system level.
Recovery. Increase the operating system level of the target system or use a
supported CLASS of DEFINE.
573 (%1075) The requested process handle cannot be

returned.
Cause. A process control operation was attempted using a process handle of a

system process (IOP).
Recovery. Only attempt the operation on user processes.
578 (%1102) The block size specified is too large.
Cause. The specified block size is too large.

Recovery. Reduce the block size specified in the file creation request to an
acceptable value. For format 1 files, the largest allowed value is 4096 bytes.
579 (%1103) The record size specified is too large for

the given block size, file type and format.
Cause. The specified record size is too large for the given block size, file type and
format.
Recovery. Increase the block size or reduce the record size specified in the file
creation request. The largest supported record size in a format 1 file is slightly less
than in a format 2 file for a given block size.
580 (%1104) An open failed because the file was oversize

and the opener did not specify use of 64-bit
primary keys.
Cause. An open failed because the file was a format 2 file and the opener did not
specify use of 64-bit primary keys. An attempt was made to open a relative file, an
entry-sequenced file, or an unstructured file with a potential size of over 4 gigabytes

2- 72
File-System Errors
without use of a special indicator specifying that the program was using 64-bit keys
rather than 32-bit keys.
Recovery. Update the application to use 64-bit keys. This involves changing the use
of 32-bit key values to 64-bit values, replacing calls to 32-bit procedures (such as
POSITION) with equivalent procedure calls (for example, FILE_SETPOSITION_), and
setting a 64-bit indicator in the FILE_OPEN_ procedure call. If the file does not contain
and does not need to contain 4 GB of data, an alternative is to reduce the maximum
file size (for example, reducing the maximum number of extents).
581 (%1105) An operation involving 32-bit primary keys

was attempted on an open which specified use
of 64-bit keys.
Cause. An operation involving 32-bit primary keys was attempted on an open which
specified use of 64-bit keys.
Recovery. Correct the application by replacing the 32-bit procedure call that caused
the error with an equivalent procedure call that accepts the use of 64-bit keys.
Alternatively, turn off the FILE_OPEN_ procedure indicator that specifies the use of 64-
bit keys. The second option will not allow the application to access non-key-
sequenced files that are 4 GB or larger.
582 (%1106) Alternate key information could not be

returned because it cannot be expressed in
the superseded format of the parameter.
Cause. Alternate key information could not be returned, because it cannot be

expressed in the superseded format of the parameter. For instance, if an alternate key
field has an offset greater than 4095, this error will be returned if the alternate key
information is requested using the FILEINQUIRE procedure, because the format from
this procedure allows only 12 bits for an offset.
Recovery. Update the application to use an interface supporting the full range of
alternate key information, for instance by using FILE_GETINFOLIST_ item 106.
583 (%1107) The extent size specified is too large or

the maxextents limit is too large.
Cause. The specified extent size is too large or a maxextents value bigger than 16
is specified for a non-key sequenced partitioned file. Usually this error occurs because
an extent size of more than 65,535 pages is specified for a format 1 file. The extent
size in question might be the primary or secondary extent size of a file, or it might be
any of the extent sizes specified in the partition description.

2- 73
File-System Errors

Recovery. For the extent size problem, reduce the extent size or arrange the
conversion of the file into format 2, which can handle larger extent sizes than format 1
files can. For the partitioned maxextents problem, avoid using a maxextents value
greater than 16.
584 (%1110) The operation could not be performed because

a software component does not support
format 2 disk files.
Cause. The operation could not be performed because a software component does
not support format 2 disk files. The component may be local to the caller or may be on
a remote system if a remote object is being manipulated.
Recovery. Check the software versions on the local system and, if applicable, the
remote system to determine which component does not provide support for format 2
files. Some subsystems, such as optical disk, might not have a version available which
can support format 2 files. There might also be down-level version disk volume
references in partition and alternate key descriptions. Upgrade the necessary
systems, or move the objects to a systems supporting format 2 files.
590 (%1116) The parameter value is invalid or

inconsistent with another.
Cause. A parameter value is not valid, or two parameter values are not compatible.
Error 590 is returned when no specific error applies to the problem.
Recovery. The value of the parameter must be an acceptable value. Correct the
problem and try again.
593 (%1121) The request was cancelled.
Cause. The request was abandoned.

Recovery. Informational message only; no corrective action is needed.
594 (%1122) A DSM/TC error was returned to $ZSVR. Refer

to the EMS log for detailed information.
Cause. DSM/TC returned a fatal error to $ZSVR.

Effect. The tape request failed.
Recovery. See the EMS log for details about the DSM/TC error. Then see the
DSM/Tape Catalog Messages Manual for recovery actions.

2- 74
File-System Errors
595 (%1123) A ZSSI error was returned to $ZSVR. Refer

to the EMS log for detailed information.
Cause. ZSSI returned a fatal error to $ZSVR.

Effect. The tape request failed.
Recovery. See the EMS log for details about the ZSSI error. Then see the DSM/Tape
Catalog Messages Manual for recovery actions.
597 (%1125) A required item is missing from an item

list.
Cause. An item code that should have been specified in an item list was not found.
Recovery. For each item in the item list, check that all of its required items are present
and are in the correct order.
632 (%1170) Not enough stack space to complete request.
Cause. A procedure is called but less than the required amount of data stack space is
available. Some inadequate stack situations can cause this error; others cause a trap
instead.
Effect. The last error is set to this error number, and the requested operation is not
completed.
Recovery. Increase the number of stack pages available or reduce the amount of
stack space used.
634 (%1172) A logical device number exceeded 16 bits.
Cause. A logical device number too large to fit in a 16-bit field (greater than 65535)
was used.
Effect. The requested operation is not completed.
Recovery. Reduce the number of bits in the logical device number or, if the error was
returned from a procedure call in a user application, convert the procedure call to use a
new equivalent procedure that supports 32-bit logical device numbers.

2- 75
File-System Errors
635 (%1173) A volume cannot be accessed because the

other side is locked.
Cause. This error is returned under either of the following conditions:

• An attempt was made to lock an optical disk volume, but either both drives were
not up or one volume was already locked.
• An attempt was made to access an optical disk volume, but the volume on the
other side of the cartridge was locked.
Effect. The optical disk volume either cannot be locked or cannot be accessed.
Recovery. Retry the locking operation after seeing that both optical disk drives are up
and both volumes are unlocked, or retry the access operation after unlocking the
volume on the other side of the cartridge.
638 (%1176) Process cannot be stopped until process

returns to stopmode 1.
Cause. A stop request passes the security checks but the process is running at
stopmode 2.
Effect. The stop request is queued until the process to be stopped reduces its
stopmode.
639 (%1177) Process cannot be stopped until process goes

to stopmode 0.
Cause. A stop request does not pass the security checks and the process is running
at stopmode 1 or 2.
Effect. The stop request is queued until the process to be stopped reduces its
stopmode to 0.
640 (%1200) DEBUGNOW is required instead of DEBUG for a

privileged process.
Cause. DEBUGNOW is required instead of DEBUG for a privileged process.

Effect. Debug mode will not be entered.
Recovery. Use DEBUGNOW instead of DEBUG
700 (%1274) FEWRONGRECOVSEQNUM - The sequence number of

the message received by the disk process
from a recovery process does not match.

2- 76
File-System Errors
Cause. DP2 received an incorrect message sequence number.

Recovery. Contact your HP service provider.
701 (%1275) FEDEVICEDOWNFORTRANSACTIONS - The disk

process received a message from a recovery
process that requires that the disk process
be in the UP state; it is currently in the
DOWN state.
Cause. Volume is in the DOWN state.

Effect. The operation fails. TMF generates an EMS event that provides information
regarding the error.
Recovery. The system administrator must change the volume to the UP state.
702 (%1276) FEUNSUPPORTEDOP - The disk process received

an erroneous message from a recovery
process. The message requested a physical
REDO and also requested that audit be
generated.
Cause. The request is not understood by DP2.

703 (%1277) FEAUDITTOONEW - The disk process encountered

a Creation Volume Sequence Number in an
audit record sent by a recovery process that
is more recent than the CRVSN of the File
Label.
Cause. The label stored by DP2 is older than the label described in the audit record.
704 (%1278) FEPVSNMISMATCH - The disk process

encountered a Previous Volume Sequence
Number in an audit record sent by a recovery
process that does not match the VSN of the
data block on disk.
Cause. The label stored by DP2 is older than the label described in the audit record.

2- 77
File-System Errors
705 (%1279) FENOTANADP - Generated by the disk process

when it receives a TMF Auditing Disk Process
request type message and the disk process is
not an ADP volume.
Cause. The disk process is not an Auditing Disk Process.

Effect. TMF does not work.
706 (%1280) FEINVALFORADP - Generated by the disk

process when it receives a request message
that is inappropriate for an Auditing Disk
Process.
Cause. The request is inappropriate for Auditing Disk Process.

707 (%1281) FEINVALIDDPNAMETIMESTAMP - Generated by the

disk process when the Disk Process Name
Stamp(DPNameTimeStamp)in the message sent by
a recovery process does not match the
current DPNameTimeStamp of the disk.
Cause. When the DPNameTimeStamp of the request does not match the
DPNameTimeStamp of the disk process.
708 (%1282) FEFILEUNDONEEDED - The disk process

encountered a File Label that had its
UndoNeeded flag set when a recovery request
specified that the UndoNeeded flag must not
be set.
Cause. The request specifies that the file is currently marked UndoNeeded but does
not have its UndoNeeded flag set.

2- 78
File-System Errors
709 (%1283) FEFILEREDONEEDED - The disk process

encountered a File Label that had its
RedoNeeded flag set when a recovery request
specified that the RedoNeeded flag must not
be set.
Cause. The request specifies that the file is currently marked RedoNeeded but does
not have its RedoNeeded flag set.
710 (%1284) FENULLOP - No further work needs to be done

to process the current portion of the
current request OPENTMF errors.
Cause. The current audit record does not need to be applied.

711 (%1285) FEINVALAUDITREC - The disk process received

a corrupt audit record in a message from a
recovery process.
Cause. The audit record is corrupted.

712 (%1310) FERMALREADYREGISTERED
Cause. An attempt was made to register a recoverable resource manager using the
TMF_REC_RM_CREATE_ procedure, but the name was already registered.
Effect. No change is made to the resource manager directory.
Recovery. Change the name to be registered and try again if needed.
713 (%1311) FERMOUTSTANDINGTRANS
Cause. An attempt was made to remove a recoverable resource manager from the
directory using the TM_REC_RM_REMOVE_ procedure, but the resource manager
still has unresolved transactions outstanding.
Effect. The resource manager directory is not changed.

2- 79
File-System Errors
Recovery. Resolve the transactions and then retry the remove operation. The
transactions can be resolved through the system management interface (TMF-
COM/TMFSERVE) or the resource manager can be opened and the transactions can
be resolved by communicating with the appropriate foreign transaction manager.
714 (%1312) FEINVALIDPROTOCOL
Cause. A call to TMF_EXPORT_ or TMF_IMPORT_ specified an invalid combination

of protocols and options for the transaction branch.
Effect. The operation, either export or import, did not succeed.
Repeat the operation with a valid combination of protocol and options.
715 (%1313) FEINVALIDTXHANDLE
Cause. An invalid transaction handle was specified. The handle could be completely
invalid, or the type of handle could be invalid for the operation specified, such as trying
to do a TMF_SETTXHANDLE_ using the transaction handle returned from
TMF_EXPORT_ for a non pre-prepare branch.
Recovery. Retry the operation with a valid transaction handle, if possible.
716 (%1314) FETXSUSPENDREJECTED
Cause. An attempt was made to suspend a transaction (TMF_SUSPEND_), but the

caller is not the beginner nor the importer nor the “resumer” of the transaction.
Effect. The transaction is not suspended for the process.
Recovery. No recovery is necessary. However, if the process expected the operation
to complete successfully, the problem should be investigated.
717 (%1315) FETXNOTSUSPENDED
Cause. An attempt was made to resume a transaction (TMF_RESUME_), but the

specified transaction is not currently suspended.
Effect. The transaction is not resumed by the calling process.
Recovery. Determine if the transaction should be suspended and why it is not and
repair the application as appropriate.
718 (%1316) FEINVALIDSIGNAL
Cause. An invalid signal value was found. Either an invalid value was specified to
TMF_WRITE_SIGNAL_ or an invalid buffer was specified to
TMF_INTERPRET_SIGNAL_.

2- 80
File-System Errors
Effect. The operation is not performed.

Recovery. Determine why the invalid signal value was used or why the TMF sub-
system thinks the buffer is invalid.
719 (%1317) FEDATASIZEEXCEEDED
Cause. An attempt was made to write branch or resource manager data, but the
amount of data specified would cause the allowable limits on the data to be exceeded.
Effect. In the case of branch data (TMF_WRITE_TX_DATA_) the transaction is
aborted. In the case of resource manager data (TMF_WRITE_RM_DATA_), the
operation fails and the resource manager data is not updated.
Recovery. Fix the application to not generate more than the allowable amount of
branch or resource manager data. Branch data is limited to 1.5K (1024) bytes per
branch, or a total of 12K (12288) bytes per transaction. Resource manager data is
limited to 1K (1024) bytes per resource manager.
721 (%1321) FEBEGINTXNOTCOMPLETED
Cause. The TMF_RESUME_ procedure was called before the BEGINTRANSACTION

processing for the specified transaction is complete. This could occur because the
BEGINTRANSACTION procedure does not wait for the processing to complete before
returning to the caller and the caller could subsequently call TMF_SUSPEND_ and
send the tid off to another process to resume before the processing is complete.
Effect. The TMF_RESUME_ fails.
Recovery. Delay for a few seconds and try the TMF_RESUME_ again.
722 (%1322) FENOMORERMCBS
Cause. An attempt was made to open a resource manager file, but the gateway
process’ local CPU has run out of available resource manager control blocks.
Effect. The resource manager open fails.
Recovery. The number of resource manager control blocks allocated per CPU can be
configured through the subsystem management interface. If this error is consistently
being returned, that value can be increased. Another possible recovery is to distribute
the gateway processes across other CPUs where the limit has not been reached.
723 (%1323) FENOMOREBCBS
Cause. An attempt was made to export or import a transaction branch, but the limit of
the number of branch control blocks per CPU has been exceeded.

2- 81
File-System Errors
Recovery. The number of branches per resource manager can be configured through
the subsystem management interface. If this error is consistently being returned, that
value should be increased. Another possible recovery is to add more resource
managers to the directory and distribute the load of transaction branches across them.
724 (%1324) FENOTNOWAITTFILE
Cause. An attempt is made to export a transaction branch with the TFILE not opened
in a nowaited manner.
Effect. The Export will fail.
Recovery. Open the TFILE in nowaited manner and then export the transaction
branch.
725 (%1325) FEIMPORTINVALOP
Cause. An attempt is made to call ENDTRANSACTION on an imported transaction

branch.
Effect. The ENDTRANSACTION call fails.
Recovery. As TMF did not start an imported transaction branch, TMF cannot issue
ENDTRANSACTION on this transaction.
727 (%1327) FETOOMANYRECRMS
Cause. An attempt was made to create a recoverable resource manager, but the
maximum allowable recoverable resource managers for the system is exceeded.
Recovery. Use the TMFCOM command ALTER BEGINTRANS to increase the total
allowable recoverable resource managers in a system. The maximum allowable value
is 16384.
728 (%1328) FESETTXHANDLEINVALOP
Cause. An attempt was made to set the current transaction of a process by invoking
TMF_SETTXHANDLE_ passing in the TxHandle obtained by exporting a transaction
branch to a volatile resource manager with pre-prepare option, but the exported
transaction branch is prepared.
Recovery. If the transaction is still active, and more work on behalf of that transaction
must be done by that volatile resource manager, then export another transaction
branch.
729 (%1331) FEBRANCHISPREPARED

2- 82
File-System Errors
Cause. An attempt was made to write branch data to a prepared exported or imported
transaction branch.
Recovery. The branch data of a prepared transaction branch cannot be updated.
730 (%1332) FEJOINSOUTSTANDING
Cause. There are outstanding TMF_JOIN_s for the process. This error occurs during
ENDTRANSACTION processing.
Recovery. Call TMF_SUSPEND_ for each outstanding join before calling
ENDTRANSACTION.
731 (%1333) FEALREADYJOINED
Cause. Transaction has already been joined by this process.

Recovery. A process can only have one join outstanding at any time.
732 (%1334) FEALREADYRESUMED
Cause. Transaction has already been resumed by this process.

Recovery. A process can only have one resume outstanding at any time.
733 (%1335) FEBRANCHISFAILED
Cause. The transaction branch has already failed. This error occurs when branch data
is updated while the transaction is being aborted.
Recovery. Correct the program logic.
734 (%1336) [FEFILENOTRECOVERABLE] DP2 cannot recover

the file because audit record cannot be
applied.
Cause. File recovery cannot be applied to this file because the SQL/MX software
version and object version do not match.
Effect. The file recovery operation fails.

2- 83
File-System Errors
735 (%1337) [FEZLTCOMMITHOLD] DP2 cannot guarantee zero

lost transactions and has activated commit
hold.
Cause. The remote mirror disk is not available, or COMMITHOLDMODE is ON for the
volume.
Recovery. TMF will activate CommitHold for the volume.
758 (%1366) Unable to allocate space from the control

block pool or trying to open too many files
or partitions on a volume.(device type: 3)
Cause. No space is available in the control block pool to allocate a new control block,
or a file or partition open request exceeds the maximum number of opens per volume.
Recovery. Wait for a few minutes, and then retry. Check the system for processes that
are requesting too many file or partition opens. If the problem persists, contact your
service provider.
759 (%1367) Unable to allocate space from audit

checkpoint pool. (device type:3)
Cause. No space is available in the audit checkpoint pool to allocate a new buffer.
Recovery. Wait for a few minutes, and then retry. If the problem persists, contact your
service provider.
760 (%1370) Unable to allocate space for a cache

buffer.(device type: 3)
Cause. No space is available for a new cache buffer.

Recovery. This is a Data Access Manager internal error. Contact your service
provider.
761 (%1371) Unable to allocate space from the work

pool.(device type: 3)

2- 84
File-System Errors
Cause. No space is available in the work pool to allocate a new buffer.

service provider.
762 (%1372) Unable to allocate space from the free space

table.(device type: 3)
Cause. No space is available in the disk free space table.

Recovery. Use the DCOM utility to recover free space on the disk. Then rebuild the
free space table by issuing the SCF command CONTROL DISK $disk-name,
REBUILDDFS. If the problem persists, contact your service provider.
763 (%1373) Unable to allocate space from the local

Cause. No space is available in the local pool to allocate a new buffer.

Recovery. This is a Data Access Manager internal error. Contact your service
provider.
764 (%1374) Unable to allocate space from the revive

Cause. No space is available in the revive pool to allocate a new buffer.

service provider.
765 (%1375) Unable to allocate space from the SQL data

area pool (MX buffer space).(device type: 3)
Cause. No space is available in the SQL data area pool to allocate a new buffer or
reuse an existing buffer. There are too many SQL/MX session requests for buffer
space.

2- 85
File-System Errors
Recovery. Reduce the number of concurrent SQL/MX sessions for the disk volume. If
the problem persists, consider increasing the SQLMXBuffer size by issuing the SCF
command ALTER DISK $disk-name, SQLMXBUFFER size in MB. If the problem
persists, contact your service provider.
Caution. Increasing the SQL/MX buffer space reduces the memory available for cache. This
can degrade performance.
766 (%1376) The maximum number of opens for a single

file or partition has been reached.(device
type: 3)
Cause. A file or partition open request exceeds the maximum number of opens for a
single file or partition on a volume.
Recovery. Wait for a few minutes, and then retry. Check the system for processes that
are requesting too many opens of a single file or partition. If the problem persists,
contact your service provider.
899 (%1603) FENOSWITCH - An attempt to switch CPUs with

a PUP or SCF command failed.
Cause. The primary or backup DP2 process is not running.

Recovery. Ensure that both the primary and backup DP2 processes are running.
1091 (%2103) The file or table cannot be purged until the

NOPURGEUNTIL date.
Cause. An attempt was made to purge a file for which the specified expiration date
has not been reached. The expiration date is set by using item code 57 of the
FILE_CREATELIST_ procedure or by using the NOPURGEUNTIL option of the FUP
ALTER command.
Effect. On a purge operation, if the NOPURGEUNTIL option is specified, the
expiration date (a four-word GMT timestamp) is checked against the current time. If the
timestamp is less than the current time, the file is not purged. All Enscribe files existing
before C10 are assumed to have a zero expiration date.
Recovery. The expiration date of a file can be set or changed with a call to the
C-series ALTER procedure or a call to the D-series FILE_ALTERLIST_ procedure.
1092 (%2104) The File Label will overflow (FELABTOOLONG).
Cause. The operation causes the file label to overflow.

2- 86
File-System Errors

Recovery. Recreate the file with larger primary and secondary extents. This will result
in a smaller file label.
The following command lists the attributes of a file (including the file label size):
fup info <filename>, detail
Use the following command to recreate the file with a larger primary and secondary
extent:
fup dup <filename>, <new-filename>, ext (pri-ext-size, sec-ext-
size)
1163 (%2213) Illegal operation attempted on a file having

a system reserved filename.
Cause. An attempt was made to refer to a file that has either a reserved name or
cannot be used in the current context. For example, this error is returned when a
Guardian internal name for an OSS file is given as the program file in the
PROCESS_CREATE_ procedure.
Recovery. Correct the error in the call and try again.
3502 (%6656) FEREQUESTALLOCATIONFAILURE
Cause. Enough storage was not available at the server to process the request
because of the demands of other activities at the time when the request was received.
Recovery. Wait till the server finishes processing other activities and retry the
operation. Contact HP support in case of repeated failure.
4000 (%7640 Open System Services Error

through through
4999 %11607)
Cause. A request was made of Open System Services that could not be completed.
Effect. The effect depends on the specific error returned.
Recovery. The recovery action depends on the specific error returned. See
Section 22, OSS Error Information for information on how to find the meaning of an
OSS error.

2- 87
File-System Errors
5001 (%11611) Request rejected: requestor executing on a

non-SMF system. (device type: 3.36, 25.0,
or 52.0)
Cause. NonStop Storage Management Foundation (SMF) received a request it could

not process. The requestor is most likely at a version level earlier than D40.00.
Recovery. The requestor’s system must be upgraded before the request can be
processed.
5002 (%11612) Creation of logical file rejected pending

deletion of volume from storage pool.
(device type: 3.36, 25.0, or 52.0)
Cause. A request to create a logical file was rejected because the target physical
volume was being deleted from a storage pool.
Recovery. Create the logical file on a physical volume still associated with a storage
pool.
5007 (%11617) SMF access violation: insufficient

SMF-privilege to access file.
(device type: 3.36, 25.0, or 52.0)
Cause. An attempt was made to purge or alter an SMF catalog file by some means
other than PUP or SMFIXUP.
Recovery. Use PUP or SMFIXUP to operate on SMF catalog files.
5010 (%11622) An error occurred during SMF lock manager

processing. (device type: 3.36, 25.0, or
52.0)
Cause. The NonStop Storage Management Foundation (SMF) internal lock manager
encountered an error (for example, a lock could not be obtained because of a problem
with a particular file).

2- 88
File-System Errors
Recovery. Appropriate recovery varies according to the specific cause of the error. In
some cases you can wait and retry the request. The event log should contain detailed
information that can help you determine the appropriate action.
5011 (%11623) An error occurred during access of the ANT

table. (device type: 3.36)
Cause. A file-system request (create, open, purge, rename, alter, or information

inquiry) was made that required a read and/or update of the NonStop Storage
Management Foundation (SMF) ANT file, which stores the logical file names and their
mappings. A file-system error occurred during the read or update of the ANT file.
Recovery. Appropriate recovery varies according to the file-system error that occurred
during access of the ANT file. Refer to the event log to determine the file-system error
and take appropriate action.
5012 (%11624) An error occurred during access of the

PENDOPS table. (device type: 3.36, 25.0,
or 52.0)
Cause. A file-system request (create, purge, rename, or alter) was made, or a file
recovery operation was occurring, that required access to the NonStop Storage
Management Foundation (SMF) PENDOPS (pending operations) file. A file-system
error occurred during access of the PENDOPS file.
Recovery. Appropriate recovery varies according to the file-system error that occurred
during access of the PENDOPS file. Refer to the event log to determine the file-system
error and take appropriate action.
5013 (%11625) The name range for a physical volume is

exhausted. (device type: 3.36)
Cause. A request was made to create a new NonStop Storage Management

Foundation (SMF) logical file, but there are no file names available on the physical
volume. In most cases, file names are not available because the virtual disk process
has lost communication with the pool process.
This error can occur either when you specify a physical volume or when you do not
specify a physical volume and this volume is the last one to be tried. In the latter case,
other volumes might have been tried and rejected either because they returned errors
or because they were unsuitable (for example, because they did not match the file’s
criteria for auditing or mirroring).
Recovery. If you specified a physical volume, retry the request either without
specifying a physical volume or specifying a different physical volume. If you did not

2- 89
File-System Errors
specify a physical volume, start the pool process (if it’s not running), wait, and retry the
request.
5014 (%11626) No physical volumes are available.

(device type: 3.36)
Cause. A request was made to create a new NonStop Storage Management

Foundation (SMF) logical file but there are no physical volumes available for one of the
following reasons:
• Either there are no physical volumes in the storage pool that the virtual disk
process is associated with, or none of the physical volumes in the pool is in the UP
state, or none of the physical volumes in the pool matches the pool’s criteria (for
example, a pool might require that physical volumes be mirrored, but currently all
the physical volume mirrors are in the DOWN state).
• None of the physical volumes in the storage pool matches the criteria specified for
the file. For example, the file might be audited but none of the physical volumes in
the pool are currently able to generate audit.
• You specified a physical volume on which to create the file, but that physical
volume either is not part of the pool that the virtual disk process is associated with
or it is subject to one of the problems described in the preceding bullets.
Recovery. Appropriate recovery varies according to the specific cause of the problem.
Refer to the event log to determine appropriate action. The recovery might be, for
example, to add physical volumes to the storage pool, bring volumes to the UP state,
or remove the physical volume specification from the original request.
5015 (%11627) The outcome for the request is unknown.

(device type: 3.36, 25.0, or 52.0)
Cause. During processing of a file-system request (create, purge, rename, or alter)

the NonStop Storage Management Foundation (SMF) subsystem lost communication
with the disk process. This could occur, for example, as a result of a processor failure.
Recovery. Take appropriate action to recover the disk process and then retry the
request.
5017 (%11631) Unable to read one of the SMF catalog

tables. (device type: 3.36, 25.0, or 52.0)
Cause. A process other than the NonStop Storage Management Foundation (SMF)
subsystem attempted to directly access an SMF catalog file and a file-system error
occurred.

2- 90
File-System Errors
Recovery. Refer to the event log for detailed information and take appropriate action.
5018 (%11632) An error occurred during message processing.

(device type: 3.36, 25.0, or 52.0)
Cause. A file-system request (create, purge, or alter) was made against a NonStop
Storage Management Foundation (SMF) object, and the reply from the disk process
contained an error. This often indicates a HP internal error.
Recovery. Refer to the event log for additional information and take action as
appropriate. Retry the request. If the problem persists, contact your service provider.
5028 (%11644) Unable to read a received message.

(device type: 3.36, 25.0, or 52.0)
Cause. A NonStop Storage Management Foundation (SMF) process received an

invalid message from any process.
Effect. The message is ignored and the requested operation is not performed.
5034 (%11652) Unable to start the thread manager.

(device type: 3.36, 25.0, or 52.0)
Cause. A request was made to a NonStop Storage Management Foundation (SMF)

disk process or pool process, which was unable to start the thread manager while
processing the request. This error is typically caused by a shortage of some resource,
usually memory.
appropriate. The recovery might be to wait and retry the request when the system is
less busy. If the problem persists or if it occurs frequently, contact your service
provider.
5035 (%11653) Unable to start a thread.

(device type: 3.36, 25.0, or 52.0)
Cause. A request was made to a NonStop Storage Management Foundation (SMF)

disk process or pool process, which was unable to start a thread while processing the
request. This error is typically caused by a shortage of some resource, usually
memory.

2- 91
File-System Errors
Recovery. Wait and retry the request when the system is less busy. If the problem
persists or if it occurs frequently, contact your service provider.
5043 (%11663) An error occurred during a read of a SMF

catalog file. (device type: 3.36, 25.0, or
52.0)
Cause. A file-system request (create, purge, rename, or alter) was made, and an error
occurred during a read of a NonStop Storage Management Foundation (SMF) catalog
file.
Recovery. Appropriate recovery varies according to the specific file-system error that
occurred during the read. Refer to the event log to determine the file-system error and
take appropriate action.
5048 (%11670) BEGINTRANSACTION error.

(device type: 3.36, 25.0, or 52.0)
Cause. The NonStop Storage Management Foundation (SMF) subsystem was trying
to begin a Transaction Management Facility (TMF) transaction during processing of the
operation when an error occurred on the call to the BEGINTRANSACTION procedure.
Effect. The requested operation failed and recovery processing is aborted. The SMF
subsystem periodically tries again to perform recovery processing.
Recovery. Appropriate recovery varies according to the specific error returned by the
BEGINTRANSACTION procedure. Refer to the event log and take action as
appropriate. The recovery might be to just retry the request. In some cases, no user
action is required.
5049 (%11671) A memory allocation failed.

(device type: 3.36, 25.0, or 52.0)
Cause. The NonStop Storage Management Foundation (SMF) subsystem was unable
to allocate memory due to a shortage of resources.
Recovery. Wait and retry the request. If the problem persists, bring the SMF process
to the DOWN state and then to the UP state before retrying the request again. If the
problem still persists, contact your service provider.
5050 (%11672) An unexpected error occurred during request

52.0)
Cause. This error message indicates a HP internal error.


2- 92
File-System Errors
5053 (%11675) The SMF lock manager reached its lock

threshold. (device type: 3.36, 25.0, or
52.0)
Cause. The maximum number of locks have been granted on a particular NonStop
Storage Management Foundation (SMF) object.
Recovery. Wait and retry the request. If the problem persists, bring the SMF process
to the DOWN state and then to the UP state before retrying the request again. If the
problem still persists, contact your service provider.
5064 (%11710) An error occurred during recovery

52.0)
Cause. An error occurred while the NonStop Storage Management Foundation (SMF)
subsystem was attempting to perform recovery processing.
subsystem periodically tries again to perform recovery processing.
Recovery. Appropriate recovery varies according to the specific error that occurred.
Refer to the event log for detailed information. In some cases, no user action is
required.
5065 (%11711) An error occurred during indeterminate

outcome processing. (device type: 3.36,
25.0, or 52.0)
Cause. A resource allocation failure occurred within the NonStop Storage

Management Foundation (SMF) subsystem during recovery processing after a 5015
error was returned.
Effect. The requested operation failed and recovery processing is aborted.
appropriate. Bring the SMF process to the DOWN state and then to the UP state.
Retry the request. If the problem persists, contact your service provider.

2- 93
File-System Errors
5072 (%11720) Recovery processing was unable to initialize

PENDOPS processing. (device type: 3.36,
25.0, or 52.0)
Cause. When a NonStop Storage Management Foundation (SMF) process attempted

to perform recovery processing after the failure of a create, purge, rename, or update
of an SMF object, an error occurred that prevented the SMF process from initializing
the PENDOPS (pending operations) file. This error can also be returned if there is a
lock failure during recovery processing or if the PENDOPS file is unavailable.
process enters the DOWN state.
Recovery. Refer to the event log for detailed information. In the case of a lock failure,
take appropriate action. Otherwise, wait and retry the operation by bringing the SMF
process to an UP state. If the problem persists, contact your service provider.
5073 (%11721) A PENDOPS semaphore failed; the process that

locked a SMF object could not be identified.
(device type: 3.36, 25.0, or 52.0)
Cause. When a NonStop Storage Management Foundation (SMF) process attempted

to perform recovery processing after the failure of a create, purge, rename, or update
of an SMF object, an error occurred in the PENDOPS (pending operations) file, which
prevented identification of the process that locked an SMF object.
process enters the DOWN state.
Recovery. Wait and retry the operation by bringing the SMF process to an UP state. If
the problem persists, contact your service provider.
5076 (%11724) An invalid operation code was found in the

PENDOPS table; the PENDOPS entry is invalid.
(device type: 3.36, 25.0, or 52.0)
Cause. When the NonStop Storage Management Foundation (SMF) subsystem

attempted to perform recovery processing after an SMF process failed, an invalid
operation code was found in the PENDOPS (pending operations) file.
Effect. The requested operation failed and recovery processing is aborted.

2- 94
File-System Errors
5313 (%12301) A virtual disk process encountered an error

when trying to perform the requested
operation. (device type: 3.36)
Cause. During processing of a file-system request (create, open, purge, rename, alter,
or information inquiry) on a NonStop Storage Management Foundation (SMF) logical
file, the virtual disk process encountered an error.
Recovery. Appropriate recovery varies according to the specific error. Refer the event
log to determine the specific file-system error and take appropriate action.

2- 95
File-System Errors Error Lists
Error Lists
If you are using the Subsystem Programmatic Interface (SPI) to send commands to a
subsystem, you might receive a file-system error list in a response. HP subsystems
return such an error list when, in performing your request, they call a file-system
procedure directly or indirectly and an error occurs on the call.
The contents of the error list depend on which procedure was called. The standard
SPI token ZSPI-TKN-PROC-ERR, which is present in every file-system error list,
identifies the procedure.
Each error list always includes the unconditional tokens listed under its description in
this subsection. In addition, each error list can include any of the conditional tokens
listed under its description and in Table 2-4.
If you are designing a subsystem that uses SPI, follow these guidelines when
constructing a file-system error list:
• Include all unconditional tokens listed in the error-list description.
• Make the ZFIL-TKN-XFILENAME token value null if you cannot obtain a valid file
name, such as when these errors occur:
• File-system error 16 (file not opened). This error indicates a coding error.
• File-system error 26 (no outstanding I/O requests) when it is returned by an
AWAITIO call in which file-number is equal to -1.
• File-system error 40 (operation timed out) when it is returned by an AWAITIO
call in which file-number is equal to -1.
• Optionally include any or none of the conditional tokens listed in Table 2-4. Note
that the tokens listed might not be appropriate for every file-system error. To obtain
values for the tokens, use FILE_GETINFOLIST_ or FILEINFO and pass the
procedure a valid file number, or use FILE_GETINFOLISTBYNAME_ or FILEINFO
and pass the procedure a valid file name.
This subsection does not discuss the mechanics of error-list construction. For
information about creating error lists, and for additional information about tokens and
token types, and for definitions of tokens whose names begin with ZSPI-, refer to the
SPI Programming Manual.
List (page 1 of 3)
Token Name Token Type Description
ZFIL-TKN-ALTKEYPARAMS ZSPI-TYP-BYTESTRING is the alternate key parameters.
ZFIL-TKN-BLOCKLENGTH ZSPI-TYP-UINT is the data block length.
ZFIL-TKN- ZSPI-TYP-INT is the current key length.
CURRENTKEYLENGTH

2- 96
List (page 2 of 3)
ZFIL-TKN- ZSPI-TYP-INT is the current key specifier.
CURRENTKEYSPEC
ZFIL-TKN- ZSPI-TYP-BYTESTRING is the current key value.
CURRENTKEYVALUE
ZFIL-TKN- ZSPI-TYP-INT is the current primary key length.
CURRENTPRIKEYLENGTH
ZFIL-TKN- ZSPI-TYP-INT2 is the current record pointer (relative
CURRENTRECPOINTER byte address).
ZFIL-TKN-DEVTYPE ZSPI-TYP-INT is the device type.
ZFIL-TKN-EOFPOINTER ZSPI-TYP-INT2 is the end-of-file location (relative
byte address).
ZFIL-TKN-ERRORDETAIL ZFIL-TYP-INT is the error-detail parameter
where applicable.
ZFIL-TKN-ERRORPARTITION ZSPI-TYP-INT is the partition in error.
ZFIL-TKN-EXTENTSIZE ZSPI-TYP-INT is the file extent size.
ZFIL-TKN-FILE- ZFIL-TYP-INT is the file-open-access mode.
OPEN-ACCESS
ZFIL-TKN-FILE- ZFIL-TYP-INT is the file-open-exclusion
OPEN-EXCLUSION mode.
ZFIL-TKN-FILE- ZFIL-TYP-INT is the file-open-nowait depth.
OPEN-NOWAIT
ZFIL-TKN-FILE- ZFIL-TYP-INT is the file-open-options flags.
OPEN-OPTIONS
ZFIL-TKN-FILECODE ZSPI-TYP-ENUM is the file code.
ZFIL-TKN-FILENUMBER ZSPI-TYP-INT is the file number.
ZFIL-TKN-FILETYPE ZSPI-TYP-ENUM is the file type.
ZFIL-TKN-KEYINERROR ZSPI-TYP-INT is the key specifier that is in error.
ZFIL-TKN-KEYPARAMS ZSPI-TYP-BYTESTRING is the key parameters.
ZFIL-TKN-LASTMODTIME ZSPI-TYP-TIMESTAMP is the last modified timestamp.
ZFIL-TKN-LDEV ZSPI-TYP-INT is the logical device number.
ZFIL-TKN-MAXEXTENTS ZSPI-TYP-INT is the maximum number of extents.
ZFIL-TKN-MAXSIZE ZSPI-TYP-INT2 is the maximum file size.
ZFIL-TKN- ZSPI-TYP-INT2 is the next record location (relative
NEXTRECPOINTER byte address).
ZFIL-TKN-NUMEXTENTS ZSPI-TYP-INT is the number of allocated extents.
ZFIL-TKN-NUMPARTITIONS ZSPI-TYP-INT is the number of partitions.
ZFIL-TKN-OBJECTFILE ZSPI-TYP-FNAME is the file name of the reporting
program.

2- 97
List (page 3 of 3)
ZFIL-TKN-OPENFLAGS ZSPI-TYP-UINT is the number of open flags.
ZFIL-TKN-OWNER ZSPI-TYP-BYTE-PAIR is the file owner ID or the CRAID
(process ID).
ZFIL-TKN- ZSPI-TYP-BYTESTRING is the partition parameters.
PARTITIONPARAMS
ZFIL-TKN-PARTITIONSIZE ZSPI-TYP-INT is the partition size.
ZFIL-TKN-RECORDLENGTH ZSPI-TYP-INT is the record length.
ZFIL-TKN- ZSPI-TYP-INT is the secondary extent size.
SECONDARYEXTENTSIZE
ZFIL-TKN-SECURITY ZSPI-TYP-ENUM is the file security.
ZFIL-TKN- ZSPI-TYP-INT is the subdevice number.
SUBDEVNUMBER
ZFIL-TKN- ZSPI-TYP-INT is the sync or receive depth.
SYNCRECEIVEDEPTH
ZFIL-TKN- ZSPI-TYP-INT is the unstructured buffer size.
UNSTRUCTUREDBUFSIZE

2- 98
File-System Errors 1: ZFIL-VAL-AWAITIO
1: ZFIL-VAL-AWAITIO
A call to the file-system procedure AWAITIO resulted in an unexpected error.
Unconditional Tokens
ZSPI-TKN-ERRLIST token-type ZSPI-TYP-LIST.

ZSPI-TKN-ERROR token-type ZSPI-TYP-ERROR.
ZSPI-TKN-PROC-ERR token-type ZSPI-TYP-ENUM.
ZFIL-TKN-XFILENAME token-type ZSPI-TYP-STRING.
ZSPI-TKN-ENDLIST token-type ZSPI-TYP-SSCTL.
Conditional Tokens
ZFIL-TKN-BUFFERADDR token-type ZSPI-TYP-UINT.

ZFIL-TKN-COUNTTRANSFERRED token-type ZSPI-TYP-INT.
ZFIL-TKN-FILENAME token-type ZSPI-TYP-FNAME.
ZFIL-TKN-TAG token-type ZSPI-TYP-INT2.
ZFIL-TKN-TIMEOUT token-type ZSPI-TYP-INT2.
ZSPI-TKN-ERROR is the standard SPI error token, whose value consists of the fields
Z-SSID and Z-ERROR. Z-SSID is the subsystem identifier ZFIL-VAL-SSID. Z-ERROR
is the file-system error code returned in the error parameter of FILEINFO.
ZSPI-TKN-PROC-ERR is the procedure in which the error occurred. Its value is
ZFIL-VAL-AWAITIO (1).
ZFIL-TKN-XFILENAME is the completely qualified file name (including node name) in
external format. The token ZFIL-TKN-XFILENAME can be null if certain errors occur.
For example, if file-system error 16 (file not opened) occurs, the token will be null.
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the I/O buffer.
ZFIL-TKN-COUNTTRANSFERRED is the length in bytes of transferred data.
ZFIL-TKN-FILENAME is the name of the file in internal network format. If the file
name cannot be represented in this format or if the file name is not known (file-system
error 16 occurred), this token will not appear.
ZFIL-TKN-TAG is the AWAITIO tag value.
ZFIL-TKN-TIMEOUT is the timeout value.

2- 99
File-System Errors 2: ZFIL-VAL-CHECKCLOSE
Effect
The AWAITIO operation fails.
Recovery
Follow the recovery procedure for the returned file-system error code as described
earlier in this section.
2: ZFIL-VAL-CHECKCLOSE
A call to the procedure FILE_CLOSE_CHKPT_ or CHECKCLOSE resulted in a
condition code less (CCL). Either the file number supplied in the call is invalid, or the
caller’s backup no longer exists.

Conditional Tokens
ZFIL-TKN-FILENUMBER token-type ZSPI-TYP-INT.

ZFIL-TKN-TAPEDISP token-type ZSPI-TYP-ENUM.
is the file-system error code returned in the error parameter of FILEINFO; it is always
zero, since CHECKCLOSE does not return an error code.
ZFIL-VAL-CHECKCLOSE (2).
Conditional Tokens
ZFIL-TKN-FILENUMBER is the number of the closed file.

2 -100
File-System Errors 3: ZFIL-VAL-CHECKMONITOR
ZFIL-TKN-TAPEDISP is the selected tape disposition.
Effect
The operation fails.
Recovery
If the file number is invalid, correct the file number. If the caller’s backup no longer
exists, this error is for information only; no corrective action is necessary.
3: ZFIL-VAL-CHECKMONITOR
The primary process paired with the process that called CHECKMONITOR stopped or
executed a call to CHECKSWITCH.

ZFIL-TKN-CHECKPOINTSTATUS token-type ZSPI-TYP-INT.
Conditional Tokens
ZFIL-TKN-OLDPRIMARY token-type ZSPI-TYP-CRTPID.
zero, since CHECKMONITOR does not return an error code.
ZSPI-TKN-PROC-ERR is the procedure in which the error occurred. Its value is ZFIL-
VAL-CHECKMONITOR (3).
ZFIL-TKN-CHECKPOINTSTATUS bits <8:15> indicate why the backup took over:
Value Meaning
0 Primary stopped

2 -101
File-System Errors 3: ZFIL-VAL-CHECKMONITOR
Value Meaning
1 Primary abnormally ended
2 Primary’s processor failed
3 Primary called CHECKSWITCH
Conditional Tokens
ZFIL-TKN-OLDPRIMARY is the old primary process ID.
Effect
Normal processing continues in the new primary.
Recovery
Informative message only; no corrective action is needed.

2 -102
File-System Errors 4: ZFIL-VAL-CHECKOPEN
4: ZFIL-VAL-CHECKOPEN
A call to the procedure CHECKOPEN returned a nonzero file-system error code.

Conditional Tokens
ZFIL-TKN-FILENUMBER token-type ZSPI-TYP-INT.

ZFIL-TKN-OPENFLAGS token-type ZSPI-TYP-UINT.
ZFIL-TKN-SYNCRECEIVEDEPTH token-type ZSPI-TYP-INT.
ZFIL-TKN-BLOCKBUFFER token-type ZSPI-TYP-INT.
ZFIL-TKN-BLOCKBUFFERLENGTH token-type ZSPI-TYP-UINT.
VAL-CHECKOPEN (4).
ZFIL-TKN-FILENAME is the file name (blank-filled if unknown).
Conditional Tokens
ZFIL-TKN-FILENUMBER is the number of the file opened in the primary process.
ZFIL-TKN-OPENFLAGS is the open flags used by the primary process.
ZFIL-TKN-SYNCRECEIVEDEPTH is the sync or receive depth used by the primary
process.
ZFIL-TKN-BLOCKBUFFER is the address of the sequential block buffer.
ZFIL-TKN-BLOCKBUFFERLENGTH is the length in bytes of the sequential block buffer.
Effect
The file might not be correctly opened in the primary or backup process of a process
pair.

2 -103
File-System Errors 5: ZFIL-VAL-CHECKPOINT
Recovery
Follow the recommended recovery procedure for the returned file-system error code.
5: ZFIL-VAL-CHECKPOINT
A call to the procedure CHECKPOINT returned a nonzero file-system error code.
Either an error was detected, or a backup takeover occurred.

Conditional Tokens
ZFIL-TKN-STACKBASE token-type ZSPI-TYP-UINT.

ZFIL-TKN-SREGISTER token-type ZSPI-TYP-UINT.
ZFIL-MAP-CHECKPOINTCELL
def ZFIL-DDL-CHECKPOINTCELL
version C00
FOR z-file-sync THROUGH z-file-number
end.
zero, since CHECKPOINT does not return an error code.
VAL-CHECKPOINT (5).

2 -104
File-System Errors 5: ZFIL-VAL-CHECKPOINT
ZFIL-TKN-CHECKPOINTSTATUS indicates the status of the call to CHECKPOINT.

This token returns a status word in the following form:
Value Meaning
<0:7> = 0 No error
<0:7> = 1 There is no backup or the primary process cannot communicate with the
backup; then <8:15> = file-system error number
<0:7> = 2 Takeover from the primary; then:
<8:15> =0 (primary stopped)
<8:15> =1 (primary abnormally ended)
<8:15> =2 (primary’s processor failed)
<8:15> =3 (primary called CHECKSWTICH)
Conditional Tokens
ZFIL-TKN-STACKBASE is the checkpoint base-of-stack.
ZFIL-TKN-SREGISTER is the current S register.
ZFIL-MAP-CHECKPOINTCELL is the data to checkpoint. This token can appear from
1 to 13 times.
Effect
If an error occurred, the operation fails.
If the backup process took over, the system continues normally.
Recovery
Check ZFIL-TKN-CHECKPOINTSTATUS for information about CHECKPOINT.
If the backup process took over, this is an informative message only; no corrective
action is necessary.

2 -105
File-System Errors 6: ZFIL-VAL-CHECKPOINTMANY
6: ZFIL-VAL-CHECKPOINTMANY
A call to the procedure CHECKPOINTMANY returned a nonzero file-system error
code. Either an error was detected or a backup takeover occurred.

Conditional Tokens

ZFIL-MAP-CHECKPOINTLIST
def ZFIL-DDL-CHECKPOINTLIST
version C00
FOR z-cellcount THROUGH z-checkpointcell
end.
zero, since CHECKPOINTMANY does not return an error code.
VAL-CHECKPOINTMANY (6).

2 -106
File-System Errors 6: ZFIL-VAL-CHECKPOINTMANY
ZFIL-TKN-CHECKPOINTSTATUS indicates the status of the call to

CHECKPOINTMANY. This token returns a status word in the following form:
Value Meaning
<0:7> = 0 No error
backup; then <8:15> = file-system error number
<0:7> = 2 Takeover from the primary; then
<8:15> = 0 (primary stopped)
<8:15> = 1 (primary abnormally failed)
<8:15> = 2 (primary’s processor failed)
<8:15> = 3 (primary called CHECKSWITCH)
<0:7> = 3 Invalid parameter; then
<8:15> = 1 (error in stack-base parameter)
<8:15> = n, n > 1 (error in word[n-2]; refer to the Guardian Procedure Calls
Reference Manual for more information
Conditional Tokens
ZFIL-MAP-CHECKPOINTLIST is the CHECKPOINT list passed to
CHECKPOINTMANY.
Effect
The operation fails. If the backup process took over, the system continues normally.
Recovery
If an error occurred, check ZFIL-VAL-CHECKPOINTSTATUS for information about
CHECKPOINTMANY.
If the backup process took over, this is an informative message only; no corrective
action is necessary.

2 -107
File-System Errors 7: ZFIL-VAL-CHECKSWITCH
7: ZFIL-VAL-CHECKSWITCH
A call to the procedure CHECKSWITCH returned a nonzero file-system error code.
Either an error was detected or a backup takeover occurred.

Conditional Tokens
is zero, since CHECKPOINT does not return an error code.
VAL-CHECKSWITCH (7).
ZFIL-TKN-CHECKPOINTSTATUS indicates the status of the call to CHECKSWITCH.
Value Meaning
<0:7> = 0 No error
backup; then
<8:15> = file-system error number
<8:15> = 0 primary stopped
<8:15> = 1 primary abnormally ended
<8:15> = 2 primary’s processor failed
<8:15> = 3 primary called CHECKSWITCH
Conditional Tokens

2 -108
File-System Errors 8: ZFIL-VAL-CONTROL
Effect
If an error occurred, the operation fails. If the backup process took over, the system
continues normally.
Recovery
Check ZFIL-VAL-CHECKPOINTSTATUS for information about CHECKSWITCH. If the
backup process took over, this is an informative message only; no corrective action is
necessary.
8: ZFIL-VAL-CONTROL
A call to the procedure CONTROL resulted in an unexpected error.

ZFIL-TKN-OPERATION token-type ZSPI-TYP-ENUM.
Conditional Tokens

ZFIL-TKN-PARAMETER token-type ZSPI-TYP-INT.
VAL-CONTROL (8).
ZFIL-TKN-OPERATION is the CONTROL operation code. Refer to the Guardian
Procedure Calls Reference Manual for information about CONTROL operation codes.

2 -109
File-System Errors 8: ZFIL-VAL-CONTROL
Conditional Tokens
ZFIL-TKN-PARAMETER is the CONTROL operation parameter. Refer to the Guardian
Procedure Calls Reference Manual for information about the CONTROL operation
parameter.
Effect
The attempted CONTROL operation fails.
Recovery

2- 110
File-System Errors 9: ZFIL-VAL-CREATE
9: ZFIL-VAL-CREATE
A call to the procedure CREATE resulted in an unexpected error.

Conditional Tokens
ZFIL-TKN-OPENDEFAULTS token-type ZSPI-TYP-UINT.
VAL-CREATE (9).
Conditional Tokens
ZFIL-TKN-OPENDEFAULTS is the file-label open defaults.
Effect
The attempted CREATE operation fails.
Recovery

2 -111
File-System Errors 10: ZFIL-VAL-KEYPOSITION
10: ZFIL-VAL-KEYPOSITION
A call to the procedure KEYPOSITION resulted in an unexpected error.

Conditional Tokens
ZFIL-TKN-COMPAREKEYLENGTH token-type ZSPI-TYP-BYTE-PAIR.

ZFIL-TKN-KEYVALUE token-type ZSPI-TYP-BYTESTRING.
ZFIL-TKN-KEYSPEC token-type ZSPI-TYP-INT.
ZFIL-TKN-POSITIONMODE token-type ZSPI-TYP-ENUM.
VAL-KEYPOSITION (10).
Conditional Tokens
ZFIL-TKN-COMPAREKEYLENGTH is the key length (<8:15>) and the compare length
(<0:7>).
ZFIL-TKN-KEYVALUE is the key value.
ZFIL-TKN-KEYSPEC is the key specifier.
ZFIL-TKN-POSITIONMODE is the positioning mode.

2- 112
File-System Errors 11: ZFIL-VAL-OPEN
Effect
The attempted KEYPOSITION operation fails.
Recovery
11: ZFIL-VAL-OPEN
A call to the OPEN procedure resulted in a condition code less (CCL) or a condition
code greater (CCG). If CCL (ZFIL-TKN-CONDITION = -1), the file is not opened. If
CCG (ZFIL-TKN-CONDITION = 1), the file is open, but the file system detected an
exceptional condition during the OPEN operation.

Conditional Tokens
ZFIL-TKN-CONDITION token-type ZSPI-TYP-INT.

ZFIL-TKN-PRIMARYFILE token-type ZSPI-TYP-INT.
ZFIL-TKN-PRIMARYPROCESS token-type ZSPI-TYP-CRTPID.
ZFIL-TKN-BLOCKBUFFERLENGTH token-type ZSPI-TYP-UINT.
VAL-OPEN (11).
Conditional Tokens
ZFIL-TKN-CONDITION is the condition code returned by OPEN (-1, 0, or 1).
ZFIL-TKN-PRIMARYFILE is the primary file number.
ZFIL-TKN-PRIMARYPROCESS is the primary process ID of the process for which the
backup process reports the error.

2- 113
File-System Errors 12: ZFIL-VAL-PURGE
ZFIL-TKN-BLOCKBUFFERLENGTH is the length of the requested sequential block

buffer.
Effect
The attempted OPEN operation fails.
Recovery
12: ZFIL-VAL-PURGE
A call to the PURGE procedure resulted in an unexpected error.

VAL-PURGE (12).
Effect
The attempted PURGE operation fails.
Recovery

2- 114
File-System Errors 13: ZFIL-VAL-POSITION
13: ZFIL-VAL-POSITION
A call to the POSITION procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-TKN-RECORDSPEC token-type ZSPI-TYP-INT2.
ZFIL-VAL-POSITION (13).
Conditional Tokens
ZFIL-TKN-RECORDSPEC is the record specifier (RBA).
Effect
The attempted POSITION operation fails.
Recovery

2- 115
File-System Errors 14: ZFIL-VAL-READ
14: ZFIL-VAL-READ
A call to the READ procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-TKN-TRANSFERCOUNT token-type ZSPI-TYP-INT.
ZFIL-VAL-READ (14).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the READ buffer.
ZFIL-TKN-COUNTTRANSFERRED is the length of the data read (in bytes).
ZFIL-TKN-TRANSFERCOUNT is the request length (in bytes).
Effect
The attempted read operation fails.

2- 116
File-System Errors 15: ZFIL-VAL-READLOCK
Recovery
15: ZFIL-VAL-READLOCK
A call to the READLOCK procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-READLOCK (15).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the READLOCK buffer.

2- 117
File-System Errors 16: ZFIL-VAL-READUPDATE
Effect
The attempted READLOCK operation fails.
Recovery
16: ZFIL-VAL-READUPDATE
A call to the READUPDATE procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-READUPDATE (16).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the READUPDATE buffer.

2- 118
File-System Errors 17: ZFIL-VAL-READUPDATELOCK
Effect
The attempted READUPDATE operation fails.
Recovery
17: ZFIL-VAL-READUPDATELOCK
A call to the READUPDATELOCK procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-READUPDATELOCK (17).

2- 119
File-System Errors 18: ZFIL-VAL-REPLY
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the READUPDATELOCK buffer.
Effect
The attempted READUPDATELOCK operation fails.
Recovery
18: ZFIL-VAL-REPLY
A call to the REPLY procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-TKN-RECEIVETAG token-type ZSPI-TYP-INT2.
ZFIL-TKN-REPLYCODE token-type ZSPI-TYP-INT.
ZFIL-VAL-REPLY (18).
2 -120
File-System Errors 19: ZFIL-VAL-SETMODE
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the REPLY buffer.
ZFIL-TKN-TRANSFERCOUNT is the length of REPLY (in bytes).
ZFIL-TKN-COUNTTRANSFERRED is the length of the data written (in bytes).
ZFIL-TKN-RECEIVETAG is the LASTRECEIVE or RECEIVEINFO tag value.
ZFIL-TKN-REPLYCODE is the error value returned with the reply.
Effect
The attempted REPLY operation fails.
Recovery
19: ZFIL-VAL-SETMODE
A call to the SETMODE procedure resulted in an unexpected error.

ZFIL-TKN-FUNCTION token-type ZSPI-TYP-ENUM.
Conditional Tokens

ZFIL-TKN-PARAM1 token-type ZSPI-TYP-INT.
ZFIL-TKN-LASTPARAMS token-type ZSPI-TYP-INT2.
ZFIL-VAL-SETMODE (19).
2 -121
File-System Errors 20: ZFIL-VAL-SETMODENOWAIT

ZFIL-TKN-FUNCTION is the SETMODE function. Refer to the Guardian Procedure
Calls Reference Manual for more information about the SETMODE functions.
Conditional Tokens
ZFIL-TKN-PARAM1 is the first function parameter.
ZFIL-TKN-PARAM2 is the second function parameter.
ZFIL-TKN-LASTPARAMS contains the previous parameter settings.
Effect
The attempted SETMODE operation fails.
Recovery
20: ZFIL-VAL-SETMODENOWAIT
A call to the SETMODENOWAIT procedure resulted in an unexpected error.
ZFIL-TKN-FUNCTION token-type ZSPI-TYP-ENUM.
Conditional Tokens

ZFIL-TKN-LASTPARAMS token-type ZSPI-TYP-INT2.

2 -122
File-System Errors 20: ZFIL-VAL-SETMODENOWAIT
ZFIL-VAL-SETMODENOWAIT (20).
ZFIL-TKN-FUNCTION is the SETMODENOWAIT function. Refer to the Guardian
Procedure Calls Reference Manual for more information about the SETMODENOWAIT
functions.
Conditional Tokens
ZFIL-TKN-PARAM1 is the first function parameter.
ZFIL-TKN-PARAM2 is the second function parameter.
ZFIL-TKN-LASTPARAMS contains the previous parameter settings.
Effect
The attempted SETMODENOWAIT operation fails.
Recovery

2 -123
File-System Errors 21: ZFIL-VAL-WRITE
21: ZFIL-VAL-WRITE
A call to the WRITE procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-WRITE (21).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the WRITE buffer.

2 -124
File-System Errors 22: ZFIL-VAL-WRITEREAD
Effect
The attempted WRITE operation fails.
Recovery
22: ZFIL-VAL-WRITEREAD
A call to the WRITEREAD procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-TKN-READCOUNT token-type ZSPI-TYP-INT.
ZFIL-TKN-WRITECOUNT token-type ZSPI-TYP-INT.
ZFIL-VAL-WRITEREAD (22).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the WRITEREAD buffer.

2 -125
File-System Errors 22: ZFIL-VAL-WRITEREAD
ZFIL-TKN-READCOUNT is the read length (in bytes).
ZFIL-TKN-WRITECOUNT is the write length (in bytes).
Effect
The attempted WRITEREAD operation fails.
Recovery

2 -126
File-System Errors 23: ZFIL-VAL-WRITEUPDATE
23: ZFIL-VAL-WRITEUPDATE
A call to the WRITEUPDATE procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-WRITEUPDATE (23).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the WRITEUPDATE buffer.

2 -127
File-System Errors 24: ZFIL-VAL-WRITEUPDATEUNLOCK
Effect
The attempted WRITEUPDATE operation fails.
Recovery
24: ZFIL-VAL-WRITEUPDATEUNLOCK
A call to the WRITEUPDATEUNLOCK procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-WRITEUPDATEUNLOCK (24).
Conditional Tokens
ZFIL-TKN-BUFFERADDR is the base address of the WRITEUPDATEUNLOCK buffer.

2 -128
File-System Errors 25: ZFIL-VAL-AWAITIOX
Effect
The attempted WRITEUPDATEUNLOCK operation fails.
Recovery
25: ZFIL-VAL-AWAITIOX
A call to the file-system procedure AWAITIOX resulted in an unexpected error.

Conditional Tokens
ZFIL-TKN-BUFFERADDRX token-type ZSPI-TYP-UINT.

ZFIL-TKN-TIMEOUT token-type ZSPI-TYP-INT2.
ZFIL-VAL-AWAITIOX (25).

2 -129
File-System Errors 26: ZFIL-VAL-CHECKPOINTMANYX
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the I/O buffer.
ZFIL-TKN-COUNTTRANSFERRED is the length of the transferred data (in bytes).
ZFIL-TKN-TAG is the AWAITIOX tag value.
ZFIL-TKN-TIMEOUT is the timeout value.
Effect
The AWAITIOX operation fails.
Recovery
26: ZFIL-VAL-CHECKPOINTMANYX
A call to the procedure CHECKPOINTMANYX returned a nonzero file-system error.
Either an error was detected or a backup takeover occurred.

Conditional Tokens

ZFIL-MAP-CHECKPOINTLIST
def ZFIL-DDL-CHECKPOINTLIST
version C00
FOR z-cellcount THROUGH z-checkpointcell
end.

2 -130
File-System Errors 26: ZFIL-VAL-CHECKPOINTMANYX
zero, since CHECKPOINTMANYX does not return an error code.
VAL-CHECKPOINTMANYX (26). ZFIL-TKN-CHECKPOINTSTATUS indicates the
status of the call to CHECKPOINTMANYX. This token returns a status word in the
following form:
Value Meaning
<0:7> = 0 No error
backup; then
<0:7> = 3 Invalid parameter; then
<8:15> = 1 error in stack-base parameter
<8:15> = n, n > 1 error in word[n-2]; refer to the Guardian Procedure Calls
Reference Manual for more information
Conditional Tokens
ZFIL-MAP-CHECKPOINTLIST is the CHECKPOINTX list passed to
CHECKPOINTMANYX.
Effect
If an error occurs, the operation fails. If the backup process takes over, the system
continues normally.
Recovery
Check ZFIL-TKN-CHECKPOINTSTATUS for information about CHECKPOINTMANYX.
If the backup process took over, this message is informative only; no corrective action
is necessary.
2 -131
File-System Errors 27: ZFIL-VAL-CHECKPOINTX
27: ZFIL-VAL-CHECKPOINTX
A call to the procedure CHECKPOINTX returned a nonzero file-system error. Either an
error was detected or a backup takeover occurred.

Conditional Tokens

ZFIL-MAP-CHECKPOINTCELL
def ZFIL-DDL-CHECKPOINTCELL
version C00
FOR z-file-sync THROUGH z-file-number
end.
zero, since CHECKPOINTX does not return an error code.
VAL-CHECKPOINTX (27).
ZFIL-TKN-CHECKPOINTSTATUS indicates the status of the call to CHECKPOINTX.
Value Meaning
<0:7> = 0 No error
backup; then

2 -132
File-System Errors 28: ZFIL-VAL-KEYPOSITIONX
Conditional Tokens
ZFIL-MAP-CHECKPOINTCELL is the data to checkpoint. This token can appear from
1 to 13 times.
Effect
If an error occurs, the operation fails. If the backup process takes over, the system
continues normally.
Recovery
Check ZFIL-TKN-CHECKPOINTSTATUS for information about CHECKPOINTX.
If the backup process took over, this message is informative only; no corrective action
is necessary.
28: ZFIL-VAL-KEYPOSITIONX
A call to the procedure KEYPOSITIONX resulted in an unexpected error.

Conditional Tokens
ZFIL-TKN-COMPAREKEYLENGTH token-type ZSPI-TYP-BYTE-PAIR.

ZFIL-TKN-KEYVALUE token-type ZSPI-TYP-BYTESTRING.
ZFIL-TKN-KEYSPEC token-type ZSPI-TYP-INT.
ZFIL-TKN-POSITIONMODE token-type ZSPI-TYP-ENUM.
ZFIL-VAL-KEYPOSITIONX (28).
2 -133
File-System Errors 29: ZFIL-VAL-READLOCKX

Conditional Tokens
ZFIL-TKN-COMPAREKEYLENGTH is the key length (<8:15>) and the compare length
(<0:7>).
ZFIL-TKN-KEYVALUE is the key value.
ZFIL-TKN-KEYSPEC is the key specifier.
ZFIL-TKN-POSITIONMODE is the positioning mode.
Effect
The attempted KEYPOSITIONX operation fails.
Recovery
29: ZFIL-VAL-READLOCKX
A call to the READLOCKX procedure resulted in an unexpected error.
Conditional Tokens


2 -134
File-System Errors 29: ZFIL-VAL-READLOCKX
ZFIL-VAL-READLOCKX (29).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the READLOCKX buffer.
Effect
The attempted READLOCKX operation fails.
Recovery

2 -135
File-System Errors 30: ZFIL-VAL-READUPDATELOCKX
30: ZFIL-VAL-READUPDATELOCKX
A call to the READUPDATELOCKX procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-READUPDATELOCKX (30).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the READUPDATELOCKX buffer.

2 -136
File-System Errors 31: ZFIL-VAL-READUPDATEX
Effect
The attempted READUPDATELOCKX operation fails.
Recovery
31: ZFIL-VAL-READUPDATEX
A call to the READUPDATEX procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-READUPDATEX (31).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the READUPDATEX buffer.

2 -137
File-System Errors 32: ZFIL-VAL-READX
Effect
The attempted READUPDATEX operation fails.
Recovery
32: ZFIL-VAL-READX
A call to the READX procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-VAL-READX (32).

2 -138
File-System Errors 33: ZFIL-VAL-REPLYX
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the READX buffer.
Effect
The attempted READX operation fails.
Recovery
33: ZFIL-VAL-REPLYX
A call to the REPLYX procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-TKN-RECEIVETAG token-type ZSPI-TYP-INT2.
ZFIL-TKN-REPLYCODE token-type ZSPI-TYP-INT.
ZFIL-VAL-REPLYX (33).
2 -139
File-System Errors 34: ZFIL-VAL-WRITEREADX
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the REPLYX buffer.
ZFIL-TKN-TRANSFERCOUNT is the length of the reply (in bytes).
ZFIL-TKN-RECEIVETAG is the LASTRECEIVE or RECEIVEINFO tag value.
ZFIL-TKN-REPLYCODE is the error value returned with the reply.
Effect
The attempted REPLYX operation fails.
Recovery
34: ZFIL-VAL-WRITEREADX
A call to the WRITEREADX procedure resulted in an unexpected error.

Conditional Tokens

ZFIL-TKN-READCOUNT token-type ZSPI-TYP-INT.
ZFIL-TKN-WRITECOUNT token-type ZSPI-TYP-INT.

2 -140
File-System Errors 35: ZFIL-VAL-WRITEUPDATEUNLOCKX

ZFIL-VAL-WRITEREADX (34).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the WRITEREADX buffer.
ZFIL-TKN-READCOUNT is the read length (in bytes).
ZFIL-TKN-WRITECOUNT is the write length (in bytes).
Effect
The attempted WRITEREADX operation fails.
Recovery
35: ZFIL-VAL-WRITEUPDATEUNLOCKX
A call to the WRITEUPDATEUNLOCKX procedure resulted in an unexpected error.

Conditional Tokens


2 -141
File-System Errors 35: ZFIL-VAL-WRITEUPDATEUNLOCKX
VAL-WRITEUPDATEUNLOCKX (35).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the WRITEUPDATEUNLOCKX
buffer.
Effect
The attempted WRITEUPDATEUNLOCKX operation fails.
Recovery

2 -142
File-System Errors 36: ZFIL-VAL-WRITEUPDATEX
36: ZFIL-VAL-WRITEUPDATEX
A call to the WRITEUPDATEX procedure resulted in an unexpected error.

Conditional Tokens

VAL-WRITEUPDATEX (36).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the WRITEUPDATEX buffer.

2 -143
File-System Errors 37: ZFIL-VAL-WRITEX
Effect
The attempted WRITEUPDATEX operation fails.
Recovery
37: ZFIL-VAL-WRITEX
A call to the WRITEX procedure resulted in an unexpected error.

Conditional Tokens

VAL-WRITEX (37).
Conditional Tokens
ZFIL-TKN-BUFFERADDRX is the base address of the WRITEX buffer.

2 -144
File-System Errors 38: ZFIL-VAL-CLOSE
Effect
The attempted WRITEX operation fails.
Recovery
38: ZFIL-VAL-CLOSE
A call to the file-system procedure FILE_CLOSE_ or CLOSE resulted in an unexpected
error.

Conditional Tokens
ZSPI-TKN-FILENAME token-type ZSPI-TYP-FNAME.
VAL-CLOSE (38).

2 -145
File-System Errors 39: ZFIL-VAL-DEVICEINFO2
Conditional Tokens
Effect
The CLOSE operation fails.
Recovery
39: ZFIL-VAL-DEVICEINFO2
A call to the file-system procedure FILE_GETINFOBYNAME_ or DEVICEINFO2
resulted in an unexpected error.

VAL-DEVICEINFO2 (39).
Effect
The DEVICEINFO2 operation fails.
Recovery

2 -146
File-System Errors 65: ZFIL-VAL-FILE-OPEN-CHKPT
65: ZFIL-VAL-FILE-OPEN-CHKPT
A call to FILE_OPEN_CHKPT_ returned a nonzero error as the function value.

ZFIL-TKN-STATUS token-type ZSPI-TYP-INT.
is the error code returned in the error parameter of FILE_OPEN_CHKPT_.
VAL-FILE-OPEN-CHKPT (65).
ZFIL-TKN-XFILENAME is the file name.
ZFIL-TKN-STATUS qualifies the source of the error.
Effect
Status is the reason for the non zero value of the error parameter of
FILE_OPEN_CHKPT_ and is the code returned in the status parameter to the
procedure. FILE_OPEN_CHKPT_ failed for the reason denoted by a non zero status
value with an error value as in the previously described error codes.
Recovery
66: ZFIL-VAL-FILE-CREATELIST
A call to FILE_CREATELIST_ returned a nonzero error as the function value.


2 -147
File-System Errors 67: ZFIL-VAL-FILE-OPEN
is the error code returned in the error parameter of FILE_CREATELIST_.
VAL-FILE-CREATELIST (66).
external format.
Effect
The operation failed.
Recovery
67: ZFIL-VAL-FILE-OPEN
A call to FILE_OPEN_ returned a nonzero error as the function value.

ZFIL-TKN-STATUS token-type ZSPI-TYP-INT.
is the error code returned in the error parameter of FILE_OPEN_.
VAL-FILE-OPEN (67).
ZFIL-TKN-XFILENAME is the file name.
ZFIL-TKN-STATUS qualifies the source of the error.

2 -148
File-System Errors 68: ZFIL-VAL-FILE-PURGE
Effect
Status is the file number returned by FILE_OPEN_. If an error occurred and ZFIL-
TKN-STATUS is -1, the open failed. If ZFIL-TKN-STATUS is not -1, the file was
opened with a warning.
Recovery
68: ZFIL-VAL-FILE-PURGE
A call to FILE_PURGE_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILE_PURGE_.
VAL-FILE-PURGE (68).
external format.
Effect
Recovery

2 -149
File-System Errors 69: ZFIL-VAL-FILE-CLOSE
69: ZFIL-VAL-FILE-CLOSE
A call to FILE_CLOSE_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILE_CLOSE_.
VAL-FILE-CLOSE (69).
Effect
Recovery
70: ZFIL-VAL-FILE-GETINFOBYNAME
A call to FILE_GETINFOBYNAME_ returned a nonzero error as the function value.


2 -150
File-System Errors 71: ZFIL-VAL-FILE-GETRECEIVEINFO
is the error code returned in the error parameter of FILE_GETINFOBYNAME_.
VAL-FILE-GETINFOBYNAME (70).
external format.
Effect
Recovery
71: ZFIL-VAL-FILE-GETRECEIVEINFO
A call to FILE_GETRECEIVEINFO_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILE_GETRECEIVEINFO_.
VAL-FILE-GETRECEIVEINFO (71).
Effect
Recovery

2 -151
File-System Errors 72: ZFIL-VAL-FILENAME-COMPARE
72: ZFIL-VAL-FILENAME-COMPARE
A call to FILENAME_COMPARE_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILENAME_COMPARE_.
VAL-FILENAME-COMPARE (72).
ZFIL-TKN-XFILENAME (first occurrence) is the name of the first file.
ZFIL-TKN-XFILENAME (second occurrence) is the name of the second file.
Effect
Recovery
73: ZFIL-VAL-FILE-GETOPENINFO
A call to FILE_GETOPENINFO_ returned a nonzero error as the function value.


2 -152
File-System Errors 74: ZFIL-VAL-DISK-REFRESH
is the error code returned in the error parameter of FILE_GETOPENINFO_.
VAL-FILE-GETOPENINFO (73).
external format
Effect
Recovery
74: ZFIL-VAL-DISK-REFRESH
A call to DISK_REFRESH_ returned a nonzero error as the function value.

is the error code returned in the error parameter of DISK_REFRESH_.
VAL-DISK-REFRESH (74).
external format.
Effect

2 -153
File-System Errors 75: ZFIL-VAL-FILE-RENAME
Recovery
75: ZFIL-VAL-FILE-RENAME
A call to FILE_RENAME_ returned a nonzero error as the function value.

Conditional Tokens
is the error code returned in the error parameter of FILE_RENAME_.
VAL-FILE-RENAME (75).
ZFIL-TKN-XFILENAME is the new file name.
Conditional Tokens
ZFIL-TKN-XFILENAME is the original file name.
Effect
The file is not renamed.
Recovery

2 -154
File-System Errors 76: ZFIL-VAL-FILENAME-FINDSTART
76: ZFIL-VAL-FILENAME-FINDSTART
A call to FILENAME_FINDSTART_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILENAME_FINDSTART_.
VAL-FILENAME-FINDSTART (76).
external format.
Effect
Recovery
77: ZFIL-VAL-FILENAME-FINDNEXT
A call to FILENAME_FINDNEXT_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILENAME_FINDNEXT_.
2 -155
File-System Errors 78: ZFIL-VAL-FILENAME-FINDFINISH
VAL-FILENAME-FINDNEXT (77).
Effect
Recovery
78: ZFIL-VAL-FILENAME-FINDFINISH
A call to FILENAME_FINDFINISH_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILENAME_FINDFINISH_.
VAL-FILENAME-FINDFINISH (78).
Effect
Recovery

2 -156
File-System Errors 80: ZFIL-VAL-FILE-CREATE
80: ZFIL-VAL-FILE-CREATE
A call to FILE_CREATE_ returned a nonzero error as the function value.

is the error code returned in the error parameter of FILE_CREATE_.
VAL-FILE-CREATE (80).
external format.
Effect
Recovery

2 -157
3 Sequential I/O Errors
The following error codes are produced by the sequential I/O (SIO) procedures. An
application process can use the following SIO procedures to sequentially access files:
CHECK^BREAK, CHECK^FILE, CLOSE^FILE, GIVE^BREAK, NOÊRROR,
OPEN^FILE, READ^FILE, SET^FILE, TAKE^BREAK, WAIT^FILE, and WRITE^FILE.
The error number is returned as the function result from the procedure call. Except for
errors 521, 532, and 533, all the error codes in this section indicate fatal coding errors.
Interactive users can obtain a short explanation of most SIO errors by entering the
following from the TACL prompt:
1> ERROR number
For further information about the sequential I/O (SIO) procedures, refer to the Guardian
Programmer’s Guide and the Guardian Procedure Calls Reference Manual.
Message Descriptions
This subsection lists each SIO error code and provides a description of each code.
512 SIOERRÎNVALIDPARAM: An invalid parameter was

specified.
Cause. The SIO procedure call contained an invalid parameter.

Recovery. Correct the parameter in error.
513 SIOERR^MISSINGFILENAME: A file name was missing.
Cause. The file control block (FCB) did not contain a file name.
Recovery. To correct this error, assign a file name before calling OPEN^FILE. You can
assign a file name by using the ALLOCATE^FCB macro, the ALLOCATE^FCB^D00
macro, the SET^FILE procedure, or the command interpreter ASSIGN command.

3 -1
Sequential I/O Errors Message Descriptions
514 SIOERR^DEVNOTSUPPORTED: The SIO procedures do not

support the specified device type.
Cause. SIO procedures do not support the specified device type. This error is returned
only by OPEN^FILE.
Recovery. Do not attempt an SIO operation on devices of this type.
515 SIOERRÎNVALIDACCESS: The specified access is

invalid.
Cause. This error results if:

• Meaningless access was specified (for example, a read from $0 or a line printer)
• Read-write access was specified for an EDIT file or in conjunction with blocking.
Recovery. Do not attempt the SIO operations listed above.
516 SIOERRÎNVALIDBUFADDR: The specified buffer address

is
invalid. A valid address is within 'G' [0:32767].
Cause. The buffer was not within 'G'[0:32767] of the user data area. This error is
returned by OPEN^FILE if the block buffer fails the test, or by READ^FILE and
WRITE^FILE if the data buffer fails the test.
Recovery. Move the buffer into the lower half of the user data area.
517 SIOERRÎNVALIDFILECODE: The specified file code in

the
SET^FILE call does not match the file code of the
file.
Cause. The file code specified in the ASSIGN command or in the

ASSIGN^FILECODE option of the SET^FILE call did not match the file code of the file.
This error is returned only by OPEN^FILE.

3 -2
Recovery. Change the file name or the file code in the ASSIGN command or in the
ASSIGN^FILECODE option of the SET^FILE call.
518 SIOERR^BUFTOOSMALL: The block buffer provided to

OPEN^FILE is too small.
Cause. The specified buffer was too small. This error is returned only by OPEN^FILE.
Recovery. For reading an EDIT file, allocate at least 144 bytes of buffer space.
For writing an EDIT file, allocate at least 1024 bytes of buffer space.
For blocking, allocate at least the same number of bytes for buffer space as the logical
record length.
519 SIOERRÎNVALIDBLKLENGTH: The block length specified

in the SET^FILE call does not match the block buffer
length in OPEN^FILE.
Cause. The ASSIGN block length or the ASSIGN^BLOCKLENGTH option of the

SET^FILE call was greater than the block buffer length in the OPEN^FILE call. This
error is only returned by OPEN^FILE.
Recovery. Correct the ASSIGN command or the ASSIGN^BLOCKLENGTH option, or
use a larger buffer.
520 SIOERRÎNVALIDRECLENGTH: The specified record length

was either too small or too large.
Cause. There are three possible causes:

• The specified record length was zero or greater than the max-record-length
specified in the OPEN^FILE call.
• The record length for the $RECEIVE file was less than 14.
• The record length was greater than 254 and the procedure specifies variable-
length records.
This error is returned only by OPEN^FILE.
Recovery. Correct the record length.

3 -3
521 SIOERRÎNVALIDEDITFILE: The specified file is not

a valid EDIT file.
Cause. The directory indicates that the EDIT file is damaged. This error is returned
only by OPEN^FILE.
Recovery. Refer to Section 15, EDITREAD and EDITREADINIT Errors, for corrective
action.
522 SIOERR^FILEALREADYOPEN: Either the SET^FILE or

CHECK^FILE operation is not valid on an open file
or OPEN^FILE was called for a file already open.
Cause. The program used SET^FILE or CHECK^FILE for a file that should be closed,
or it used OPEN^FILE for a file that is already open. This error is returned only by
SET^FILE, CHECK^FILE, and OPEN^FILE.
Recovery. Either close the file or correct the procedure call (for example, change
parameters to permit operation when the file is open).
523 SIOERRÊDITREADERR: An EDITREAD or EDITREADINIT

error occurred.
Cause. An EDITREAD or EDITREADINIT error occurred. This error is returned only

by OPEN^FILE and READ^FILE.
Recovery. Refer to Section 15, EDITREAD and EDITREADINIT Errors, for corrective
action.
SIOERR^FILENOTOPEN: The specified file was not open.
Cause. The specified file is closed. There is a check, read, set, write, or wait error.
Recovery. Either open the file or correct the procedure call (for example, change
parameters to permit operation when the file is closed).

3 -4
525 SIOERRÂCCESSVIOLATION: The requested operation was

inconsistent with the access mode.
Cause. The operation was inconsistent with the access mode specified at OPEN^FILE
time. For example, WRITE^FILE to a file opened for READACCESS only, or
READ^FILE from a file opened for WRITEACCESS only.
Recovery. Change the operation or access mode.
526 SIOERR^NOSTACKSPACE: The required operation failed

because of insufficient stack space.
Cause. A waited WRITE^FILE operation attempted to allocate a temporary buffer from

which to do the actual WRITE. For example, it needed to add padding beyond the end
of the user-supplied buffer, or the next physical transfer began at an odd byte.
However, the attempt failed. This error is returned only by WRITE^FILE.
Recovery. Increase the run-time memory size if it is less than 32K bytes; otherwise,
move one or more nonstring arrays to upper memory.
527 SIOERR^BLOCKINGREQD: The temporary buffer required

for a nowait WRITE^FILE operation was not provided.
Cause. A nowait WRITE^FILE operation required a temporary buffer but none was
provided.
Recovery. Supply a block buffer.
530 SIOERRÎNVALIDRECVWRITE: The program called

WRITE^FILE
for $RECEIVE before calling READ^FILE.
Cause. The program called WRITE^FILE for $RECEIVE without first calling
READ^FILE. This error is returned only by WRITE^FILE.
Recovery. Add the missing read.

3 -5
531 SIOERR^CANTOPENRECV: A call to CHECK^BREAK could not

open $RECEIVE or $RECEIVE was opened without calling
OPEN^FILE.
Cause. A call to CHECK^BREAK could not open $RECEIVE for break monitoring. The
user opened $RECEIVE without calling the OPEN^FILE procedure. This error is
returned only by CHECK^BREAK.
Recovery. Change the program so that a call to OPEN^FILE opens $RECEIVE.
532 SIOERRÎORESTARTED: A nowait I/O operation has

been restarted.
Cause. A nowait I/O was specified, and the first call to READ^FILE or WRITE^FILE
started the first physical I/O. The first call to WAIT^FILE waits for the first physical
operation to finish, starts the second, and returns this error. Subsequent calls to
WAIT^FILE repeat this pattern until the logical I/O operation is finished and either or
an actual error code is returned. READ^FILE, WRITE^FILE, and WAIT^FILE return this
error.
Effect. The nowait I/O operation is restarted.
Recovery. Call WAIT^FILE again to continue waiting.
533 SIOERRÎNTERNAL: An internal SIO error occurred.
Cause. An internal error occurred. This error is returned by WAIT^FILE and

CLOSE^FILE.
534 SIOERR^CHECKSUMCOMM: A discrepancy was detected

between the common FCB checksum and the previous
checksum.
Cause. While performing a checksum on the common file control block (FCB), the
system encountered a discrepancy between this checksum and the previous
checksum. This error is returned by all procedures.
requested operation and the process abends. The FCB might be damaged.
Recovery. Check the program for pointer errors or other errors that might have
caused this error.

3 -6
535 SIOERR^CHECKSUM: A discrepancy was detected

between the file FCB checksum and the previous
checksum.
Cause. While performing a checksum on the file control block (FCB), the system
encountered a discrepancy between this checksum and the previous checksum. This
error is returned by all procedures.
requested operation and the process abends. The FCB might be damaged.
Recovery. Check the program for pointer errors or other errors that could have
returned this error.
536 SIOÔLDCOMMFCB: The FCB format allocated is not

valid.
Cause. The file FCB used to open $RECEIVE is in the new D-series format, but the
common FCB is in the old C-series or earlier format.
Effect. If the open flag, ABORTÔPENERR is set or defaults to TRUE, then the
process ABENDs; if ABORTÔPENERR is set to FALSE, then error 536 is returned to
the caller of OPEN^FILE. This is a fatal error.
Recovery. This error probably results from a programming error. Define the common
FCB using the compile-time DEFINE^CBS^D or the run-time SET^FILE
INIT^FILEFCB^D.
537 SIOERRÊDITLINEOFLOW: The highest possible EDIT line

number was exceeded.
Cause. The SIO procedure call attempted to write a line to an EDIT file with a line
number greater than 99999.999 (the largest allowable line number).
Effect. The line is not written. The error is treated as fatal.
Recovery. Write smaller files, use a file organization other than EDIT files, use SIO or
IOEdit to specify fractional line numbers, or use IOEdit to renumber the file. The
smallest possible line number increment that SIO can set is 0.001.
538 SIOERRÊXTSIZEÔVERFLOW: Extent size is greater than

65535 pages
Cause. The CHECK^FILE procedure could not return the primary or secondary extent
value because the extent size is greater than 65,535 pages.
Recovery. Do not attempt the SIO operation listed above.

3 -7
539 SIOERR^PAGEWRITEÔVERFLOW: The highest possible page

in EDIT file has been written.
Cause. While writing the data to the edit file, the SIO procedure call attempted to write
a page above the edit file limit.
Effect. Data cannot be written further in the edit file.
Recovery. Use multiple edit files instead of a single edit file to write data or use a
structured file, for example, Entry-sequenced.

3 -8
4 DEFINE Errors
This section contains errors that relate specifically to DEFINE attribute sets. DEFINE
errors can occur when you create, save, restore, delete, or otherwise manipulate
DEFINEs of any class using the DEFINE procedures.
Error Codes
This subsection lists each DEFINE error code and provides a description of each code.
0 (%000000) Successful completion.
Cause. The call completed successfully.

Effect. The operation is successful.
2049 (%004001) A syntax error occurred in the DEFINE

name.
Cause. There was a syntax error in the DEFINE name.

Effect. The procedure returns the error code and does not perform the requested
operation.
Recovery. Correct the syntax error, then reissue the request.
2050 (%004002) The DEFINE name already exists.
Cause. The DEFINE name being added already exists.

Effect. The procedure returns the error code and does not add the DEFINE name.
Recovery. Correct or change the DEFINE name, then reissue the request.
2051 (%004003) The DEFINE name does not exist.
Cause. The specified DEFINE could not be found.

operation.
Recovery. Check that you have specified the correct DEFINE name, then reissue the
request.

4 -1
DEFINE Errors Error Codes
2052 (%004004) Unable to obtain file-system buffer

space.
Cause. Either file-system buffer space was not available or there is not enough room
in the process file segment (PFS). The PFS is an extended data segment associated
with the process that, among other things, contains DEFINE information. For example,
if too many files are open or if many nowait I/O operations are outstanding, there might
not be enough space in the PFS.
Recovery. Close some files, wait for nowait I/O to finish, then try again. For G-series
releases earlier than G06, restart with a larger PFS if appropriate.
2053 (%004005) Unable to obtain physical memory.
Cause. The physical memory available was not enough to perform the requested
operation.
operation.
Recovery. Wait, then try again. If the problem persists, check the system for
processes that use too much memory.
2054 (%004006) There was a bounds error in a

parameter.
Cause. A bounds error occurred in a parameter of the procedure call.

operation.
Recovery. This is a coding error; corrective action is application-dependent.
2055 (%004007) An attribute is not allowed for

the current CLASS.
Cause. One of the attributes provided is not allowed for the current DEFINE class.
operation.
Recovery. Correct the attribute for the current class, then retry the operation.

4 -2
2056 (%004010) An attribute is missing.
Cause. An attribute was missing from the procedure call.

operation.
Recovery. Add the attribute, then retry the operation.
2057 (%004011) An attribute required for the

current DEFINE CLASS is missing
from the working set.
Cause. An attribute that is required for this DEFINE class is missing from the working
set.
Effect. If the operation is DEFINESAVE, the working set is saved. If the operation is
DEFINERESTORE, the saved DEFINE is restored to the working attribute set. If the
operation is DEFINEADD, the DEFINE is not added.
Recovery. Add the attribute for the current class, then retry the operation.
2058 (%004012) The working set is inconsistent for

the current CLASS of DEFINEs.
Cause. The working set was inconsistent (two or more attributes have conflicting
values) for the current DEFINE class.
If the consistency check failed for a SORT DEFINE, the consistency check number
returned indicates the cause.
Cod
e Meaning
001 Use SCRATCHON or NOSCRATCHON, but not both.
If the consistency check failed for a TAPE DEFINE, the consistency check number
returned indicates the cause:
Cod
e Meaning
001 Use RETENTION or EXPIRATION, not both.
002 USE IN and USE EXTEND require LABELS ANSI, LABELS IBM, or LABELS
IBMBACKUP. In addition, if REELS is specified, the value must equal the number of
volumes listed for VOLUME.
003 VOLUME is required with LABELS ANSI, LABELS IBM, or LABELS IBMBACKUP.
004 The EBCDIC attribute cannot be used with LABELS ANSI.
005 If RECFORM F is specified, BLOCKLEN must be a multiple of RECLEN.
006 Use DEVICE or SYSTEM, but not both.

4 -3
Cod
e Meaning
007 DEVICE is required for LABELS BYPASS or LABELS OMITTED. The BLOCKLEN,
EBCDIC, EXPIRATION, FILEID, FILESECT, FILESEQ, GEN, OWNER, RECFORM,
RECLEN, REELS, RETENTION, SYSTEM, USE, VERSION, and VOLUME
attributes cannot be used with LABELS BYPASS or LABELS OMITTED.
008 VOLUME SCRATCH cannot be specified with USE IN or USE EXTEND.
009 FILEID must be specified with LABELS IBM or LABELS IBMBACKUP.
010 RECLEN must be greater than 0 when BLOCKLEN is omitted.
Note. The following CLASS DEFINEs do not return error 2058: CATALOG, DEFAULTS, MAP,
SEARCH, and SUBSORT.

4 -4
If the consistency check failed for a TAPECATALOG DEFINE, the consistency check
number returned indicates the cause.
Cod
e Meaning
001 Use RETENTION or EXPIRATION, but not both.
005 If RECFORM F is specified, BLOCKLEN must be a multiple of RECLEN.
006 Use DEVICE or AVRSYSTEM, but not both.
101 USE cannot be used with LABELS OMITTED.
102 USE is required.
103 FILEID is required.
104 CATALOG OFF is required when LABELS BYPASS or LABELS OMITTED is used.
105 COMMENT, OWNER, RETENTION, TAPEMODE, VERSION NEW, and VOLUME
SCRATCH cannot be used with USE IN.
106 VOLUME is required with USE EXTEND.
107 DEVICE is required when LABELS BYPASS or LABELS OMITTED is used.
108 VOLUME is required when LABELS ANSI, LABELS BACKUP, LABELS IBM, or
LABELS IBMBACKUP is used with USE IN and CATALOG OFF.
109 The character string for OWNER exceeds the maximum length for LABELS IBM or
LABELS IBMBACKUP.
110 BLOCKLEN, RECFORM, and RECLEN cannot be used with LABELS BACKUP,
LABELS BYPASS, or LABELS OMITTED.
112 EBCDIC can be used only with LABELS IBM or LABELS IBMBACKUP.
113 AVRSYSTEM, COMMENT, FILECAT, FILEID, GEN, OWNER, POOL, RETENTION,
USE, VERSION, and VOLCAT cannot be used with LABELS BYPASS.
114 AVRSYSTEM, OWNER, and VOLUME cannot be used with LABELS OMITTED.
115 USE EXTEND cannot be used with LABELS BACKUP or LABELS IBMBACKUP.
116 GEN must be an absolute value when CATALOG OFF is used.
117 GEN cannot be specified as a relative generation when USE OUT is used.
118 GEN cannot be +1 when USE IN is used.
DEFINERESTORE, the saved DEFINE is restored to the working attribute set.
Recovery. Correct the inconsistency, then retry the operation.

4 -5
2059 (%004013) The working set for DEFINEs is invalid.
Cause. The working set is invalid.

DEFINERESTORE, the saved DEFINE is restored to the working attribute set.
Recovery. Correct the problem, then retry the operation.
2060 (%004014) No more DEFINEs are allowed.
Cause. The DEFINE name specified in the DEFINENEXTNAME call is the last
DEFINE. The define-name parameter value remains unchanged.
operation.
2061 (%004015) No more attributes are allowed for

the DEFINE.
Cause. A call was made to DEFINEREADATTR, but no other attributes exist for the
DEFINE after the last one returned. The last call for the DEFINE was successful.
operation.
2062 (%004016) The attribute is invalid for the

DEFINE.
Cause. There was a syntax error in an attribute name.

operation.
Recovery. Correct the syntax, then reissue the operation.
2063 (%004017) The name of the default subvolume has a

syntax error in the SET DEFINE command.
Cause. There was a syntax error in a default name.

operation.

4 -6
Recovery. Correct the name in the default parameter, then reissue the operation.
2064 (%004020) Resetting a required DEFINE attribute

is not allowed.
Cause. An attempt was made to reset a required DEFINE attribute.

operation.
Recovery. You cannot reset a required DEFINE attribute.
2066 (%004022) A required parameter is missing.
Cause. A required parameter was not supplied.

operation.
Recovery. Add the missing parameter, then reissue the request.
2067 (%004023) An invalid value was supplied for an

attribute.
Cause. An illegal value was supplied for an attribute.

operation.
Recovery. Correct the value, then reissue the request.
2068 (%004024) The CLASS name identifies a

nonexistent CLASS.
Cause. The DEFINE class name identified a nonexistent class.

Recovery. Correct the class name, then reissue the request.
2069 (%004025) Attempt to add a DEFINE that does not

fall under current DEFMODE setting.
Cause. The process's DEFINE mode (DEFMODE) did not allow the DEFINE to be
added.
Effect. The procedure returns the error code and does not add the DEFINE.

4 -7
Recovery. Correct the DEFMODE setting for the desired operation.
2073 (%004031) Replacing the =_DEFAULTS DEFINE with a

DEFINE having the same name but a
class other than DEFAULTS is not
allowed.
Cause. The =_DEFAULTS DEFINE cannot be replaced with a DEFINE having the
same name but a class other than DEFAULTS.
Effect. The procedure returns the error code and does not replace the =_DEFAULTS
DEFINE.
Recovery. Specify the DEFINE with a DEFAULTS class, then retry the operation.
2074 (%004032) Deleting the DEFINE is not allowed.
Cause. Some DEFINEs cannot be deleted. An attempt to delete the =_DEFAULTS

DEFINE, for example, results in this message.
Effect. The DEFINE is not deleted.
Recovery. You cannot delete the =_DEFAULTS DEFINE.
2075 (%004033) A DEFINE option is invalid.
Cause. DEFINERESTORE option.<0:13> must be 0 or

DEFINESAVE option.<0:14> must be 0.
Effect. The option not selected must be set to 0 even though it is not used. Until it is
set to 0, other options cannot be selected.
Recovery. Set the unused option to 0.
2076 (%004034) The buffer is too small for the saved

DEFINE.
Cause. The buffer was too small to contain the DEFINE.

Effect. The DEFINE is not saved. The buffer size required to save the DEFINE is
returned in the length parameter.
Recovery. Repeat the DEFINESAVE operation with a buffer at least as large as the
length returned from the failed attempt.
2077 (%004035) An extended address parameter referred

to an invalid segment.
Cause. The DEFINERESTORE buffer or define-name is in an invalid segment.

Effect. The DEFINERESTORE operation is not performed.

4 -8
Recovery. Correct the address and retry.
2078 (%004036) The DEFINERESTORE buffer

does not contain a valid saved DEFINE.
Cause. DEFINERESTORE buffer does not contain a valid saved DEFINE created
by DEFINESAVE.
Effect. The DEFINE is not added to the current set or to the working set.
Recovery. Recovery is not possible.
2079 (%004037) Cannot save the working set because the

name is =_DEFAULTS and the working set
is not class DEFAULTS.
Cause. The working set could not be saved using DEFINESAVE because the
define-name is =_DEFAULTS and the working set is not of class DEFAULTS.
Effect. The working set is not saved.
Recovery. To save the named working set, change the class of the working set to
class DEFAULTS.
If the DEFAULTS DEFINE is saved, the DEFINERESTORE replace option must be
used to restore it as an active DEFINE because the =_DEFAULTS DEFINE is always
present.
2081 (%004041) Bad internal format.
Cause. Some internal error occurs during DEFINE propagation.

Effect. Process creation fails.
Recovery. Retry the operation. Contact HP support in case of repeated failure.

4 -9
5
NEWPROCESS AND
NEWPROCESSNOWAIT Errors
The following error codes and error lists are produced by the process-control
procedures NEWPROCESS and NEWPROCESSNOWAIT. These procedures can only
create processes with process identification numbers (PINs) in the low range of 0
through 254.
The NEWPROCESS and NEWPROCESSNOWAIT procedures return errors in a
format that is different from other procedures. In many cases, the error message must
be decoded. The error returned in the error parameter of the NEWPROCESS
procedure is explained in the Guardian Procedure Calls Reference Manual. The error
returned in the rightmost half (bits <8:15>) contains either an error subcode or a
file-system error number. Refer to Section 2, File-System Errors, for an explanation of
the file-system error number. When file-system error number 119 is returned in bits
<0:7> or <8:15>, look in the errinfo parameter for an explanation of the error.
On D-series and later releases, you can use the PROCESS_LAUNCH_ or
PROCESS_CREATE_ procedures. With these procedures, you can create processes
with PINs in the low range (0 through 254) or in the high range (256 through 65535).
Because you might want your processes to communicate with processes created with
the NEWPROCESS or NEWPROCESSNOWAIT procedure, you might want to specify
low PINs for processes that could require such communication.
For information about PROCESS_LAUNCH_ and PROCESS_CREATE_ procedure
errors, see Section 6, Process Creation Errors.
For further information about creating processes, refer to the discussions of these
procedures in the Guardian Programmer’s Guide and the Guardian Procedure Calls
Reference Manual.
Error Codes
A NEWPROCESS error code is returned to the calling procedure as two 8-bit fields or,
if returned by the command interpreter or other interactive program, the message is
displayed on the terminal as a 6-digit octal number. The leftmost half of the error word
(bits <0:7>) contains the NEWPROCESS error number; the rightmost half (bits <8:15>)
contains either an error subcode or a file-system error number.
0 (%000000) NO ERROR
Cause. The call was completed successfully.

Effect. None.
Recovery. None.

5 -1
NEWPROCESS AND NEWPROCESSNOWAIT Error Codes
Errors
1 (%000400) UNDEFINED EXTERNAL(S)
Cause. The process being started contains a call to an external procedure that is not
in the operating system code area, the user library (if applicable), or the application
code area.
Process creation occurs, and a message is printed on the home terminal. For example:
PID: \SYS10.4,94 \SYS10.$XL.SVOL.TEST (TNS)
External References Not Resolved to Any User/System Library:
Prg: \SYS10.$XL.SVOL.TEST -> MY_PROC (PROC)
Undefined externals
where MY_PROC is the name of the undefined external procedure in the program file
\SYS10.$XL.SVOL.TEST.
Effect. When the operating system finds a call to an undefined external procedure, it
replaces the original call with a call to Debug. If the operating system tries to invoke the
undefined external during program execution, the program goes into the debug state.
Recovery. Either correct the coding error, use linker (Binder or nld utility) to add the
procedure to the code area or user library, or refer the call to a procedure that is
already present in the application program.
2 (%001000) NO PCB AVAILABLE
Cause. All entries in the configured process control block (PCB) table for the
processor were in use or the process required a low PIN but none were available.
Effect. No process is created.
Recovery. Try a different processor or wait and then retry the existing call.
3 (%001400 + %nnn) FILE-SYSTEM ERROR ON program-file
Cause. The system monitor encountered a file-system error while accessing

program-file during process creation.
Effect. The call returns the error number of the file-system error (%nnn) in
error.<8:15>. No process is created.
Recovery. Refer to Section 2, File-System Errors, for corrective action for error
number %nnn.

5 -2
Errors
4 (%002000) UNABLE TO ALLOCATE MAP
Cause. Not enough space was available in the processor’s MAPPOOL to permit the
system monitor to generate the code and data-map copies required by the new
process (a configuration problem).
5 (%002400 + %nnn) FILE-SYSTEM ERROR ON SWAP FILE
Cause. An error occurred during the creation or opening of the swap file.
Effect. File-system error %nnn is returned in error.<8:15>. No process is created.
number %nnn.
6 (%003000 + %nnn) ILLEGAL FILE FORMAT
Cause. When the system monitor checked on whether this file is a program, the
program or library file failed the test. (The system monitor includes checks for file
codes 100 and 700.)
Bits <8:15> (%nnn) contain an error subcode that indicates the invalid file format. Bit
15 is set to 0 if a program file is in error and to 1 if a library file is in error. Error
subcodes are listed in Table 5-1.

5 -3
Errors
Table 5-1. NEWPROCESS Error 6 Error Subcodes (page 1 of 5)

Decima
l Octal Meaning
2 %002 The program file is not a disk file.
3 %003 The library file is not a disk file.
4 %004 The program file does not have file code 100 or 700.
5 %005 The library file does not have file code 100 or 700.
6 %006 The program file does not have the correct file structure.
7 %007 The library file does not have the correct file structure.
8 %010 The program file requires a later version of the operating system.
9 %011 The library file requires a later version of the operating system.
10 %012 The program file does not have a main procedure.
13 %015 The library file has a main procedure.
14 %016 Program file has a stack definition of zero pages.
16 %020 The program file has an invalid procedure entry-point (PEP) table.
17 %021 The library file has an invalid procedure entry-point (PEP) table.
18 %022 The initial extended segment information in the program file is
inconsistent.
20 %024 The program file resident size is greater than the code area
length.
21 %025 The library file resident size is greater than the code area length.
22 %026 The program file was not prepared by the nld utility or the Binder
program.
23 %027 The library file was not prepared by the nld utility or the Binder
program.
24 %030 The program file has undefined data blocks.
25 %031 The library file has undefined data blocks.
26 %032 The program file has data blocks with unresolved references.
27 %033 The library file has data blocks with unresolved references.
28 %034 The program file has too many (more than 32) TNS code
segments.
29 %035 The library file has too many (more than 32) TNS code segments.
30 %036 TNS/R code length in the program file is invalid.
31 %037 TNS/R code length in the library file is invalid.
32 %040 TNS/R code address in the program file is invalid.
33 %041 TNS/R code address in the library file is invalid.
34 %042 TNS/R data length in the program file is invalid.
35 %043 TNS/R data length in the library file is invalid.

5 -4
Errors

Decima
l Octal Meaning
36 %044 TNS/R data address in the program file is invalid.
37 %045 TNS/R data address in the library file is invalid.
38 %046 Program file has too many TNS/R code segments.
39 %047 Library file has too many TNS/R code segments.
40 %050 Program file has invalid TNS/R resident areas.
41 %051 Library file has invalid TNS/R resident areas.
42 %052 Accelerator header in program file is invalid.
43 %053 Accelerator header in library file is invalid.
44 %054 UC (user code) was accelerated with the wrong virtual address.
45 %055 UL (user library) was accelerated with the wrong virtual address.
46 %056 Program file has entry in TNS/R fixup list with invalid external
entry-point (XEP) index value or invalid code address value.
47 %057 Library file has entry in TNS/R fixup list with invalid external entry-
point (XEP) index value or invalid code address value.
48 %060 Accelerated program file has external procedure identifier list
(EPIL), internal procedure identifier list (IPIL), or external entry-
point (XEP) table with incorrect format.
49 %061 Accelerated library file has external procedure identifier list (EPIL),
internal procedure identifier list (IPIL), or external entry-point
(XEP) table with incorrect format.
50 %062 UC (user code) was accelerated using the wrong Accelerator
option (UC, UL, SC, or SL).
51 %063 UL (user library) was accelerated using the wrong Accelerator
option (UC, UL, SC, or SL).
52 %064 Program file was accelerated with incompatible version of the
Accelerator.
53 %065 Library file was accelerated with incompatible version of the
Accelerator.
54 %066 Program file has invalid callable gateway (GW) table.
55 %067 Library file has invalid callable gateway (GW) table.
56 %070 Wrong processor type is target in program file.
57 %071 Wrong processor type is target in library file.
58 %072 Program file has inconsistent TNS/R fixup list information.
59 %073 Library file has inconsistent TNS/R fixup list information.
60 %074 An internal structure of the program file contains an error.
61 %075 An internal structure of the library file contains an error.

5 -5
Errors

Decima
l Octal Meaning
64 %100 An internal structure of the program file has an entry point value
of 0.
68 %104 The list of unresolved procedure names in the program file
contains an error.
69 %105 The list of unresolved procedure names in the library file contains
an error.
70 %106 The fixup computed an invalid file offset to the code area of the
program file.
71 %107 The fixup computed an invalid file offset to the code area of the
library file.
72 %110 The program file has an invalid fixup item.
73 %111 The library file has an invalid fixup item.
76 %114 The program file has an instruction at a call site that is not the type
expected for its fixup item.
77 %115 The library file has an instruction at a call site that is not the type
expected for its fixup item.
78 %116 The header of a TNS/R native program file is not in correct format.
79 %117 The header of a TNS/R native library file is not in correct format.
80 %120 The code in the program file starts at the wrong virtual address.
81 %121 The code in the library file starts at the wrong virtual address.
82 %122 The program file has too much data for the main stack.
84 %124 The code area of the program file is too large.
85 %125 The code area of the library file is too large.
86 %126 The program file has a gateway (GW) table but no callable
procedures.
87 %127 The library file has a gateway (GW) table but no callable
procedures.
89 %131 The file codes of the program file and library file do not match.
90 %132 The program file being started can run only in the Guardian
environment and it is being started in the OSS environment, or
vice versa.

5 -6
Errors

Decima
l Octal Meaning
91 %133 The library file being started can run only in the Guardian
environment and it is being started in the OSS environment, or
vice versa.
92 %134 The program and the library conflict on global data mapping. This
error is reported on the program. The user library selected is not
compatible with the user library specified when the Binder
program originally created the program object file.
94 %136 The program needs to import data from the library and the library
is not exporting any data. This error is reported on the program.
96 %140 The program file uses a library and is switching to a new library,
but the program was accelerated by a version of the Accelerator
that cannot handle library data relocation at fixup time.
98 %142 The program file has no code spaces.
99 %143 The library file has no code spaces.
100 %144 The program file is not executable. Either it was not linked with the
nld utility, it was not linked correctly, or it has been corrupted.
101 %145 The library file is not executable. Either it was not linked with the
nld utility, it was not linked correctly, or it has been corrupted.
102 %146 The program file is not executable. Either it was linked with an
incompatible version of the nld utility or it has been corrupted.
103 %147 The library file is not executable. Either it was linked with an
incompatible version of the nld utility or it has been corrupted.
104 %150 The program file is not executable because it has more than one
Tandem information header. An error occurred during the linking
of the program file.
105 %151 The library file is not executable because it has more than one
Tandem information header. An error occurred during the linking
of the library file.
106 %152 The program file is not executable because it has more than one
REGINFO information header. An error occurred during the
linking of the program file.
107 %153 The library file is not executable because it has more than one
REGINFO information header. An error occurred during the
linking of the library file.
108 %154 The program file is not executable because it does not have a
GINFO information header. An error occurred during the linking of
the program file.
109 %155 The library file is not executable because it does not have a
GINFO information header. An error occurred during the linking of
the library file.

5 -7
Errors

Decima
l Octal Meaning
110 %156 The program file is not executable because it does not have either
a Tandem information header, a REGINFO information header, or
a text header. An error occurred during the linking of the program
file.
111 %157 The library file is not executable because it does not have either a
Tandem information header, a REGINFO information header, or a
text header. An error occurred during the linking of the library file.
112 %160 The program file specifies too many shared run-time libraries
(SRLs).
113 %161 The library file specifies too many shared run-time libraries
(SRLs).
114 %162 The program file specifies duplicate shared run-time libraries
(SRLs).
115 %163 The library file specifies duplicate shared run-time libraries (SRLs).
117 %165 The shared run-time library (SRL) does not export any
procedures.
121 %171 The shared run-time library (SRL) does not have a file code of
700.
122 %172 The list of unresolved flags in the program file contains an error.
124 %174 An attempt was made to spawn a shell script on a remote node.

Recovery. Take corrective action as indicated by the subcode. For example, if
subcode 2, 3, 4, or 5 is returned, use the FUP INFO command or the FILEINFO
procedure to check the file code.
7 (%003400) UNLICENSED PRIVILEGED PROGRAM
Cause. The program file contains procedures having CALLABLE or PRIV attributes,
but the program file is not licensed to execute in privileged mode and the super ID was
not running the program.
Recovery. Have the super ID license the program.

5 -8
Errors
8 (%004000 + %nnn) PROCESS NAME ERROR
Cause. The process name was invalid.

Effect. File-system error number %nnn is returned in error.<8:15>. No process is
created.
number %nnn.
9 (%004400) LIBRARY CONFLICT
Cause. The call specified a library file, but the program was either already running
with a different library or was not running with a library and a library was specified.
Recovery. All processes running a given program must use the same library. Always
specify the library file name to avoid the conflicts that can arise when the library file is
modified.
10 (%005000 + %nnn) UNABLE TO COMMUNICATE

WITH SYSTEM MONITOR PROCESS
Cause. The process could not communicate with the system monitor process,
possibly because the processor module where the program was to be run did not exist
or was inoperable. %nnn in error.<8:15> is the file-system error and indicates why
the NEWPROCESS error occurred.
Recovery. Select another processor, then try again.
11 (%005400 + %nnn) FILE-SYSTEM ERROR ON LIBRARY FILE
Cause. The system monitor encountered a file-system error while accessing library file
during process creation.
created.
number %nnn.
12 (%006000) PROGRAM FILE AND LIBRARY FILE ARE THE SAME
Cause. The program file and library file are the same file.

5 -9
Errors
Recovery. Select a different library file for the program.
13 (%006400 + %nnn) INVALID SEGMENT SIZE UNABLE

TO SET UP THE PAGE TABLES
Cause. The operating system could not set up an extended data segment for the
program.
created.
number %nnn.
14 (%007000 + %nnn) FILE-SYSTEM ERROR ON INITIAL

SETUP OF THE SWAP FILE
Cause. A file-system error occurred while the operating system was trying to set up
the swap file for a COBOL program.
created.
number %nnn.
15 (%007400 + %nnn) ILLEGAL HOME TERMINAL
Cause. The home terminal name for the new process does not exist or is not a legal
process or terminal name.
Effect. File-system error number%nnn is returned in error.<8:15>. No process is
created.
number %nnn.
16 (%010000 + %nnn) I/O ERROR ON HOME TERMINAL
Cause. An I/O error occurred at the home terminal. There are undefined externals in
the object file, and NEWPROCESS cannot OPEN or WRITE to the home terminal to
display the undefined-externals message.
Effect. The error number of file-system error %nnn is returned in error.<8:15>. No
process is created.
number %nnn.

5- 10
Errors
17 (%010400 + %nnn) DEFINE CONTEXT PROPAGATION

ERROR
Cause. An error occurred when existing DEFINEs were being propagated.

Bits <8:15> contain a subcode that indicates the cause of the error, as follows:
Subcod
e Meaning
0 Unable to convert a DEFINE name to network form.
2 Internal error.
3 Illegal DEFMODE supplied.

Recovery. Retry the call to NEWPROCESS. If errors recur, contact your service
provider.
18 (%011000) OBJECT FILE WITH AN ILLEGAL

DEVICE SUBTYPE
Cause. An attempt was made to run an object file with an invalid device subtype. The
device subtype is an attribute stored in each object file. The process created from an
object file is assigned the device subtype stored in that object file. Only named
processes are allowed nonzero device subtypes.
Device subtypes in the range 1 through 15 are reserved for processes that are:
• Created by the super ID
• Created from licensed object files
• Created from object files owned and provided by the super ID
Recovery. Verify that the call to NEWPROCESS contains the name parameter. If the
name parameter exists, either recompile or rebind the object file to change the device
subtype to an unrestricted value, or contact the super ID.

5 -11
Errors
19 (%011400) PROCESS DEVICE SUBTYPE SPECIFIED

IN BACKUP PROCESS ISN'T THE SAME
AS PRIMARY'S
Cause. The operating system tried to create a process, but the backup process device
subtype was not from the same program file as the primary process's device subtype.
This error is probably a programming error.
Recovery. Find out which object file was being passed to NEWPROCESS and correct
the error.
22 (%013000) PFS^SIZE IS OUT OF RANGE
Cause. The optional parameter pfs-size specified a value outside the range of
128K through 1024K bytes.
Recovery. Correct the specification to a value within the acceptable range. If
pfs-size is not specified or if 0 is specified, the default size of 256K bytes is used.
23 (%013400) CANNOT CREATE PFS
Cause. NEWPROCESS or NEWPROCESSNOWAIT cannot allocate the process file

segment (PFS).
Recovery. If insufficient PFS space was the cause of this error, either try to free up
more PFS space or wait for more PFS space to become available.
24 (%014000 + %nnn) UNKNOWN ERROR
Cause. An error was received when an attempt was made to start a process on a
remote system. Your system does not know how to interpret the error.
Effect. The call returns the PROCESS_CREATE_ error number %nnn, or 119 if the
actual error number is too large, in errinfo.<8:15>. No process is created.
Recovery. Refer to Section 6, Process Creation Errors, for corrective action for the
PROCESS_CREATE_ error received in bits <8:15> of the errinfo parameter, and
then perform a recovery action based on this error. If the problem persists and the
PROCESS_CREATE_ error number does not give you enough information to diagnose
the problem, contact your service provider.
25 (%014400) UNABLE TO ALLOCATE A PRIV STACK

FOR THE PROCESS

5- 12
Errors
Cause. There was no segment available for the priv stack of a TNS/R native process.
Recovery. Try a different processor or wait until the system load decreases and then
retry the existing call.
26 (%015000) UNABLE TO LOCK THE PRIV STACK

FOR THE PROCESS
Cause. There was not enough physical memory free to lock the priv stack of a TNS/R
native process.
27 (%015400) UNABLE TO ALLOCATE A MAIN STACK

FOR THE PROCESS
Cause. There was no segment available for the main stack of a TNS/R native
process.
28 (%016000) UNABLE TO LOCK THE MAINSTACK OF A TNS/R

NATIVE I/O PROCESS THAT IS CREATED BY
COUP
Cause. There was not enough physical memory free to lock the main stack for a
TNS/R native I/O process. This error is only returned to some privileged callers on
D-series releases.
29 (%016400 + %nnn) SECURITY INHERITANCE FAILURE
Cause. An error occurred during an attempt to obtain or propagate security identity

information.
Effect. The call returns the file-system error number %nnn in errinfo.<8:15>. No
process is created.
Recovery. Refer to Section 2, File-System Errors, for corrective action for the
file-system error returned in errinfo.<8:15>. If the file-system error indicates an

5- 13
Errors
invalid operation or an invalid parameter, the problem might be caused by a version

mismatch between the NonStop operating system and Standard Security. Correct your
system configuration if this is the case; otherwise, the problem is likely to be an internal
error that should be reported to your service provider.
30 (%017000) UNABLE TO ALLOCATE NATIVE GLOBALS
Cause. The system was unable to allocate the global data segment of a native
process.
31 (%017400) UNABLE TO LOCK NATIVE GLOBALS
Cause. The system was unable to lock the global data segments of a native I/O
process. This error is only returned to some privileged callers.
32 (%020000) MAIN STACK MAXIMUM VALUE TOO LARGE
Cause. The main stack maximum value is too large.

Recovery. Use the nld utility to specify a new value that is within the valid range.
33 (%020400) HEAP MAXIMUM VALUE TOO LARGE
Cause. The heap maximum value is too large.

34 (%021000) SPACE GUARANTEE VALUE TOO LARGE
Cause. The space guarantee value is too large.

35 (%021400 + %nnn) DUPLICATE SRL

5- 14
Errors
Cause. The process creation request specifies duplicate shared run-time libraries
(SRLs).
Effect. %nnn in error.<8:15> contains the SRL numbers of the duplicate SRLs in the
form xxyy (where xx is the first SRL and yy is the duplicate SRL). No process is
created.
Recovery. Use the original object files and the nld utility to re-create the program file.
If the error recurs, contact your service provider.
36 (%022000 + %nnn) UNKNOWN SRL SPECIFIED BY PROGRAM
Cause. The system was unable to find a shared run-time library (SRL) specified by the
program file.
Effect. No process is created. %nnn in error.<8:15> contains the SRL number of
the SRL that could not be found.
37 (%022400 + %nnn) UNKNOWN SRL SPECIFIED BY ANOTHER

SRL
Cause. The system was unable to find a shared run-time library (SRL) specified by
another SRL.
Effect. %nnn in error.<8:15> contains the SRL numbers of the two SRLs in the form
xxyy (where xx is the SRL that specifies the yy SRL). No process is created.
38 (%023000 + %nnn) TOO MANY SRLS
Cause. The process creation request specifies too many shared run-time libraries
(SRLs).
Effect. %nnn in error.<8:15> contains the maximum number of SRLs that can be
specified. No process is created.
Recovery. Use the original object files and the nld utility to re-create the program file,
using no more than 32 SRLs, the maximum number of SRLs permitted.
39 (%023400 + %nnn) PROGRAM FILE REQUIRES FIXUP
Cause. The program file requires fixups to a shared run-time library (SRL) that is
unavailable because it is running.
Effect. %nnn in error.<8:15> contains the SRL number of the running SRL. No
process is created.

5- 15
Errors
40 (%024000 + %nnn) SRL REQUIRES FIXUPS
Cause. A shared run-time library (SRL) requires fixups to another SRL that is
unavailable because it is running.
Effect. %nnn in error.<8:15> contains the SRL numbers of the two SRLs in the form
xxyy (where xx is the SRL that requires the fixup to the running yy SRL). No process
is created.
41 (%024400 + %nnn) PROGRAM FILE SECURITY VIOLATION
Cause. A security violation occurred. The program file is not licensed but a shared
run-time library (SRL) is licensed and has instance data.
Effect. %nnn in error.<8:15> contains the SRL number of the licensed SRL. No
process is created.
Recovery. License the program file, if possible.
42 (%025000 + %nnn) SRL SECURITY VIOLATION
Cause. A security violation occurred. Either the program file or a shared run-time
library (SRL) is licensed, but another SRL is not licensed.
Effect. %nnn in error.<8:15> contains the SRL number of the unlicensed SRL. No
process is created.
Recovery. Either remove the licensing of the program, if possible, or use the nld
utility to specify another SRL.
43 (%025400 + %nnn) PROGRAM REQUIRES SYMBOL FROM SRL
Cause. The program file requires a symbol from a shared run-time library (SRL) but
the SRL is not exporting it.
Effect. %nnn in error.<8:15> contains the SRL number of the SRL that does not
export the required symbol. No process is created.
Recovery. Use the nld utility to specify another SRL that contains the desired data
block.

5- 16
Errors
47 (%027400) CANNOT GUARANTEE SWAP SPACE
Cause. Requested swap space for the TNS/R native process cannot be guaranteed.
Recovery. Either use the nld utility to shrink the guaranteed swap space
specification, use NSKCOM to add additional swap space for the target processor, or
run on another processor.
48 (%030000 + %nnn) MISMATCH ON NUMBER OF SRLS
Cause. The number of shared run-time libraries (SRLs) specified by an SRL is

incorrect.
Effect. %nnn in error.<8:15> contains the SRL number of the SRL that caused the
error. No process is created.
49 (%030400 + %nnn) UNDEFINED EXTERNALS IN SRL
Cause. A shared run-time library (SRL) has undefined externals.

Effect. %nnn in error.<8:15> contains the SRL number of the SRL that has
undefined externals. No process is created.
Recovery. Check that the SRL versions match and obtain matching versions if they
don’t. If the problem persists, contact your service provider.
50 (%031000) INCORRECT NUMBER OF SRLS IN PROGRAM FILE
Cause. The number of shared run-time libraries (SRLs) specified for the program file
is incorrect.

5- 17
Errors
51 (%031400) INCORRECT NUMBER OF SRLS IN LIBRARY
Cause. The number of shared run-time libraries (SRLs) specified for the library file is
incorrect.
Effect. No process is created, and a message is printed on the home terminal. For
example:
PID: \SAT.0,36 \SAT.$DATA.MDTUNRES.TC03 (ELF)
SRL-Client References Symbols Not Found In Nominated SRL:
Prg: \SAT.$DATA.MDTUNRES.TC03E -> FRED_PROC (PROC) -> SRL:
\$DATA.MDTUNRES.SRL1
*ERROR* PROCESS_CREATE_ Error: 51,0
52 (%032000) SRL MUST BE LICENSED
Cause. A security violation occurred. A shared run-time library (SRL) must be licensed
to be used by callable or privileged code.
Recovery. Rebuild the program file. If the error recurs, contact your service provider.
53 (%032001) UNABLE TO OBTAIN GLOBAL VIRTUAL SPACE
Cause. Either there is insufficient virtual memory or there is insufficient lockable

memory in the processor.
Recovery. Try again or select another processor. If the problem persists, run the
PEEK program in the processor where process creation failed and determine new
memory needs.
54 (%033000) SYMBOLIC REFERENCE TARGET/SOURCE TYPE

MISMATCH
Cause. An attempt was made to resolve a procedure address to a data area or a data
address to a procedure area.
example:
PID: \SAT.1,284 \SAT.$DATA.SRLVOL.NULCLIE2 (ELF)
Native UL Symbolic Reference Error: Target/Source Type Mismatch:
Prg: \SAT.$DATA.SRLVOL.NULCLIE2 -> X_INT (DATA)
*ERROR* PROCESS_CREATE_ Error: 54
Recovery. Use the original files and the nld utility to recreate the program file.

5- 18
Errors
55 (%033001) EXTERNAL DATA REFERENCE NOT RESOLVED TO

ANY USER/SYSTEM LIBRARY
Cause. An anonymous data symbol reference is not located in a UL, Native UL, or
system library.
example:
Recovery. Either correct the coding error, use nld utility to add the data to the data
area or user library, or refer the call to the data that is already present in the application
program.
119 (%073400) ERROR NUMBER RETURNED IS TOO LARGE
Cause. The error returned in error.<0:7> is too large to fit into one byte.
Recovery. Instead of the error parameter, specify the errinfo parameter, which is
a two-word parameter, to obtain complete error information.
3xx SRL HAS INVALID FILE FORMAT
Cause. The call contained an invalid file format on shared run-time library (SRL)
number xx. This error is returned in the errinfo parameter, the second word of
which contains the error subcode. See Table 6-2, Error Subcodes for Process
Creation Errors 12, 13, 70, 76, 84, and 3xx, on page 6-7 for possible subcode values.
5xx FILE-SYSTEM ERROR ON SHARED RUN-TIME LIBRARY (SRL)

NUMBER XX.
Cause. There was a file-system error on shared run-time library (SRL) number xx.
This error is returned in the errinfo parameter, the second word of which contains
the error subcode.
number errinfo parameter. If the problem persists, contact your service provider.

5- 19
NEWPROCESS AND NEWPROCESSNOWAIT Error Lists
Errors
Error Lists
subsystem, you might receive a NEWPROCESS or NEWPROCESSNOWAIT error list
in a response. HP subsystems return such an error list when, in performing your
request, they call the NEWPROCESS or NEWPROCESSNOWAIT procedure directly
or indirectly and an error occurs on the call.
The contents of the error list depend on which procedure was called. The standard SPI
token ZSPI-TKN-PROC-ERR, which is present in every NEWPROCESS and
NEWPROCESSNOWAIT error list, identifies the procedure: its value is either
ZGRD-VAL-NEWPROCESS (3) or ZGRD-VAL-NEWPROCESSNOWAIT (4).
this subsection. In addition, each error list could include any of the conditional tokens
listed under its description.
constructing a NEWPROCESS or NEWPROCESSNOWAIT error list:
• Include all unconditional tokens listed in the error list description.
• Optionally include any or none of the conditional tokens listed in each error-list
description.
information about creating error lists, additional information about tokens token types,
and definitions of tokens whose names begin with ZSPI-, refer to the SPI Programming
Manual.

5- 20
NEWPROCESS AND NEWPROCESSNOWAIT 3: ZGRD-VAL-NEWPROCESS
Errors
3: ZGRD-VAL-NEWPROCESS
There was an error during a call to NEWPROCESS.

Conditional Tokens
ZGRD-TKN-OBJECTFILE token-type ZSPI-TYP-FNAME.

ZGRD-TKN-LIBRARYFILE token-type ZSPI-TYP-FNAME.
ZGRD-TKN-SWAPFILE token-type ZSPI-TYP-FNAME.
ZGRD-TKN-PRIORITY token-type ZSPI-TYP-UINT.
ZGRD-TKN-MEMORYPAGES token-type ZSPI-TYP-BYTE.
ZGRD-TKN-CPU token-type ZSPI-TYP-BYTE.
ZGRD-TKN-CRTPID token-type ZSPI-TYP-CRTPID.
ZGRD-TKN-PROCESSNAME token-type ZSPI-TYP-CRTPID.
ZGRD-TKN-HOMETERM token-type ZSPI-TYP-FNAME.
ZGRD-TKN-INSPECTFLAG token-type ZSPI-TYP-ENAME.
Z-SSID and Z-ERROR. Z-SSID is the subsystem identifier ZGRD-VAL-SSID.
Z-ERROR is the 16-bit error code returned in the error parameter of
NEWPROCESS. This error code indicates the outcome of the process-creation
attempt.
ZSPI-TKN-PROC-ERR is the procedure code. Its value is ZGRD-VAL-NEWPROCESS
(3).
Conditional Tokens
ZGRD-TKN-OBJECTFILE is the file name of the reporting program in internal format.
ZGRD-TKN-LIBRARYFILE is the library file name in internal format.
ZGRD-TKN-SWAPFILE is the data swap-file name in internal format, which is passed
for informational purposes only. This swap file is not used. Processes swap to a file
that is managed by the Kernel-Managed Swap Facility.
ZGRD-TKN-PRIORITY is the priority of the new process.
ZGRD-TKN-MEMORYPAGES is the size of the data stack in pages.
ZGRD-TKN-CPU is the processor number for the new process.

5- 21
NEWPROCESS AND NEWPROCESSNOWAIT 4: ZGRD-VAL-NEWPROCESSNOWAIT
Errors
ZGRD-TKN-CRTPID is the process ID returned by NEWPROCESS. If

ZGRD-TKN-CRTPID is 0, no process was created.
ZGRD-TKN-PROCESSNAME is the new process name.
ZGRD-TKN-HOMETERM is the name of the home terminal for the new process.
ZGRD-TKN-INSPECTFLAG contains the selected debugging attributes.
Effect
The effect of the error depends on the NEWPROCESS error code returned.
Recovery
Follow the recovery procedure for the returned NEWPROCESS error code as
described earlier in this section.
4: ZGRD-VAL-NEWPROCESSNOWAIT
There was an error during a call to NEWPROCESSNOWAIT.

Conditional Tokens

ZGRD-TKN-LIBRARYFILE token-type ZSPI-TYP-FNAME.
ZGRD-TKN-SWAPFILE token-type ZSPI-TYP-FNAME.
ZGRD-TKN-MEMORYPAGES token-type ZSPI-TYP-BYTE.
ZGRD-TKN-CPU token-type ZSPI-TYP-BYTE.
ZGRD-TKN-PROCESSNAME token-type ZSPI-TYP-CRTPID.
ZGRD-TKN-HOMETERM token-type ZSPI-TYP-FNAME.
ZGRD-TKN-INSPECTFLAG token-type ZSPI-TYP-ENUM.
NEWPROCESSNOWAIT. This error code indicates the initial outcome of the
process-creation attempt.

5- 22
NEWPROCESS AND NEWPROCESSNOWAIT 4: ZGRD-VAL-NEWPROCESSNOWAIT
Errors

ZGRD-VAL-NEWPROCESSNOWAIT (4).
Conditional Tokens
ZGRD-TKN-OBJECTFILE is the file name of the reporting program in internal format.
ZGRD-TKN-LIBRARYFILE is the library file name in internal format.
ZGRD-TKN-SWAPFILE is the data swap-file name in internal format, which is passed
for informational purposes only. This swap file is not used. Processes swap to a file
that is managed by the Kernel-Managed Swap Facility.
ZGRD-TKN-MEMORYPAGES is the size of the data stack in pages.
ZGRD-TKN-CPU is the processor number for the new process.
ZGRD-TKN-PROCESSNAME is the new process name.
ZGRD-TKN-HOMETERM is the name of the home terminal for the new process.
ZGRD-TKN-INSPECTFLAG contains the selected debugging attributes.
Effect
Effect depends on the returned NEWPROCESSNOWAIT error code.
Recovery
Follow the recovery procedure for the returned NEWPROCESSNOWAIT error code as

5- 23
6 Process Creation Errors
The following error codes and error lists are produced by the PROCESS_LAUNCH_
and PROCESS_CREATE_ procedures, which create a Guardian process. Most of the
error codes are also returned by the PROCESS_SPAWN_ procedure, which creates
an Open System Services (OSS) process. See Section 9, PROCESS_SPAWN_ Open
System Services (OSS) Errors, for more information on error codes and error lists
produced by PROCESS_SPAWN_.
You can create a process with a number of different procedures. On C-series releases,
you typically used the NEWPROCESS or NEWPROCESSNOWAIT procedure. On
D-series and G-series releases, you probably use the PROCESS_LAUNCH_ or
PROCESS_CREATE_ procedure. With the PROCESS_LAUNCH_ and
PROCESS_CREATE_ procedures, you can create processes with process
identification numbers (PINs) in the low range (0 through 254) or in the high range (256
through 65535). In the event that you might want your processes to communicate with
processes created with the NEWPROCESS or NEWPROCESSNOWAIT procedure,
you might want to specify low PINs for processes that could require such
communication.
If PROCESS_LAUNCH_ or PROCESS_CREATE_ is called in a nowait manner, error
information is returned in system message -102 (nowait PROCESS_LAUNCH_ or
PROCESS_CREATE_ completion). See Section 20, System Messages, for details.
For further information about starting processes and about the PROCESS_LAUNCH_
and PROCESS_CREATE_ procedures, refer to the Guardian Programmer’s Guide and
the Guardian Procedure Calls Reference Manual.
Error Codes
The PROCESS_LAUNCH_ and PROCESS_CREATE_ error codes are the values
returned to the calling process in the error parameter of PROCESS_LAUNCH_ and
PROCESS_CREATE_. For many classes of errors, additional information is returned in
the error-detail parameter.
The PROCESS_SPAWN_ Guardian error codes are the values returned to the calling
process in the ZSYS-DDL-PROCESSRESULTS.Z-TPCERROR field of the
process-results parameter of PROCESS_SPAWN_. For many classes of errors,
additional information is returned in the
ZSYS-DDL-PROCESSRESULTS.Z-TPCDETAIL field of the process-results
parameter.
In descriptions of errors that can be returned by PROCESS_LAUNCH_,
PROCESS_CREATE_, and PROCESS_SPAWN_, the term “error detail” refers to the
information returned in either the error-detail parameter of PROCESS_LAUNCH_
and PROCESS_CREATE_ or the ZSYS-DDL-PROCESSRESULTS.Z-TPCDETAIL field
of the process-results parameter of PROCESS_SPAWN_.

6 -1
Process Creation Errors Error Codes
See Section 9, PROCESS_SPAWN_ Open System Services (OSS) Errors, for

PROCESS_SPAWN_ error lists and for OSS error codes returned in the
ZSYS-DDL-PROCESSRESULTS.Z-ERRNO field of the process-results
parameter to PROCESS_SPAWN_.
0 NO ERROR
Cause. The call was completed successfully.

Effect. The process was created, or creation was initiated if the procedure was called
in a nowait manner. In the latter case, creation results are returned in a user-level
system message.
1 A FILE SYSTEM ERROR WAS ENCOUNTERED ON THE PROGRAM FILE
Cause. A file-system error occurred on the program file during process creation. The
error-detail contains a file-system error number.
file-system error returned in the error-detail information.
2 PARAMETER ERROR
Cause. The call contained an invalid combination of options. For the

PROCESS_LAUNCH_ and PROCESS_SPAWN_ procedures, the error-detail
information indicates the first parameter to be found in error. See Table 6-1 for
possible values. For the PROCESS_CREATE_ procedure, the error-detail information
contains the number of first parameter found to be in error, where 1 designates the
leftmost parameter.
Recovery. Check the call to make sure the options are correct and in the proper order.

6 -2
3 BOUNDS VIOLATION
Cause. There was a bounds violation on a reference parameter. For the

PROCESS_LAUNCH_ and PROCESS_SPAWN_ procedures, the error-detail
information indicates the first parameter to be found in error. See Table 6-1 for
possible values. For the PROCESS_CREATE_ procedure, the error-detail information
contains the number of first parameter found to be in error, where 1 designates the
leftmost parameter.
Recovery. Pass a correct reference address.

Errors 2 and 3 (page 1 of 2)
error- PROCESS_LAUNCH_ Structure PROCESS_SPAWN_ Structure Field
detail Field or Parameter in Error or Parameter in Error
1 Z^PROGRAM^NAME oss-program-file
2 Z^LIBRARY^NAME Z^LIBRARYNAME
3 Z^SWAPFILE^NAME Z^SWAPFILENAME
4 ZÊXTSWAPFILE^NAME ZÊXTSWAPFILENAME
5 Z^PRIORITY Z^PRIORITY
6 Z^CPU Z^CPU
9 Z^NAMEÔPTIONS Z^NAMEOPTIONS
10 Z^PROCESS^NAME Z^PROCESSNAME
14 Z^HOMETERM^NAME Z^HOMETERM
15 Z^MEMORY^PAGES Z^MEMORYPAGES
16 Z^JOBID Z^JOBID
17 Z^CREATEÔPTIONS Z^CREATEOPTIONS
18 Z^DEFINES^NAME Z^DEFINES
19 Z^DEBUGÔPTIONS Z^DEBUGOPTIONS
20 Z^PFS^SIZE Z^PFSSIZE
21 The program image file is in error. The program image file is in error. This
This error subcode is returned if error subcode is returned if
PROCESS_LAUNCH_ is called from PROCESS_SPAWN_ is called from
COUP; it applies only to some COUP; it applies only to some
privileged callers on D-series releases. privileged callers on D-series releases.
22 param-list Not returned by PROCESS_SPAWN_
23 error-detail Z^TPCDETAIL
24 output-list Not returned by PROCESS_SPAWN_

6 -3

Errors 2 and 3 (page 2 of 2)
error- PROCESS_LAUNCH_ Structure PROCESS_SPAWN_ Structure Field
detail Field or Parameter in Error or Parameter in Error
25 output-list-len Not returned by PROCESS_SPAWN_
50 Not returned by process-extension
PROCESS_LAUNCH_
51 Not returned by ZÔSSOPTIONS
PROCESS_LAUNCH_
52 Not returned by argv
PROCESS_LAUNCH_
53 Not returned by envp
PROCESS_LAUNCH_
54 Not returned by envp contains an invalid address.
PROCESS_LAUNCH_
56 Not returned by inheritance
PROCESS_LAUNCH_
57 Not returned by An internal error
PROCESS_LAUNCH_
58 Not returned by fdinfo
PROCESS_LAUNCH_
59 Not returned by path
PROCESS_LAUNCH_
60 Not returned by inheritance-length
PROCESS_LAUNCH_
4 A FILE SYSTEM ERROR WAS ENCOUNTERED ON THE USER LIBRARY

FILE
Cause. A file-system error occurred on the user library file during process creation.
The error-detail contains a file-system error number.
5 ERROR WHILE ACCESSING SWAP FILE
Cause. An error occurred during the creation or opening of the swap file. The
error-detail information contains the error number of the file-system error that occurred.
Effect. The procedure sets the error code and returns without creating the process.

6 -4
6 ERROR WHILE ACCESSING EXTENDED SWAP FILE
Cause. An error occurred during the creation, opening, or initial setup of the extended
swap file. The error-detail information contains the error number of the file-system error
that occurred.
7 FILE SYSTEM ERROR ON PFS
Cause. An error occurred during the creation of the process file segment (PFS). The
8 ILLEGAL HOME TERMINAL
Cause. The home terminal name for the new process does not exist or is not a legal
process or terminal name. The error-detail information contains the error number of the
file-system error that occurred.
9 I/O ERROR ON HOME TERMINAL
Cause. An I/O error occurred at the home terminal. Undefined externals in the
program file prevent the operating system from opening or writing to the home terminal
to display the undefined-externals message. The error-detail information contains the
error number of the file-system error that occurred.
10 UNABLE TO COMMUNICATE WITH SYSTEM MONITOR PROCESS

6 -5
possibly because the processor module where the program was to be run did not exist
or was inoperable. The error-detail information contains the error number of the
file-system error that occurred.
11 PROCESS NAME ERROR
Cause. The process name was invalid. The error-detail information contains the error
number of the file-system error that occurred.
12 ILLEGAL PROGRAM FILE FORMAT
Cause. An error was detected in the file format of the Program file. The error-detail
information contains an error subcode that indicates the invalid file format error. These
errors are listed in Table 6-2.
subcode 1 or 2 is returned, use the FUP INFO command to check the file code. In
many cases, the object file was built improperly or corrupted. It should be relinked and
perhaps recompiled.
13 ILLEGAL USER-LIBRARY FILE FORMAT
Cause. An error was detected in the file format of the user library file. The error-detail
information contains an error subcode that indicates the invalid file format error.
These error subcodes are listed in Table 6-2.
subcode 1 or 2 is returned, use the FUP INFO command to check the file code. In
many cases, the object file was built improperly or corrupted. It should be relinked and
perhaps recompiled.

6 -6
3xx (page 1 of 4)
Subcode Meaning
1 The file is not a disk file.
2 For the program file designated in the process creation request: The file is in
the Guardian file system and does not have file code 100 or 700, or it is in the
OSS file system and not recognizable as a TNS or ELF object file or a shell
script.
Otherwise: A TNS library file was expected, but the file is in the Guardian file
system and does not have file code 100, or the file is in the OSS file system
and not recognizable as a TNS object file.
3 The file does not have the correct file structure.
4 The file requires a later version of the operating system.
5 A program lacks an entry point, or an attempt was made to load a library as a
program. (An entry point is specified by a TAL or pTAL procedure having the
MAIN attribute, or by naming a native procedure in the –e option to the linker.)
6 An attempt was made to load a program as a library, or a TNS user library has
a MAIN procedure.
7 A TNS program file does not have data pages.
8 A native object file requires fixup to SRLs by the nld utility, or a TNS object file
was not prepared by the Binder program.
9 The file header INITSEGS is not consistent with its size.
10 The file resident size is greater than the code area length.
11 The file was not prepared by the nld utility or the Binder program.
12 The file has undefined data blocks.
13 The file has data blocks with unresolved references.
14 The file has too many TNS code segments.
15 Accelerated code length in the file is invalid.
16 Accelerated code address in the file is invalid.
17 Accelerated data length in the file is invalid.
18 Accelerated data address in the file is invalid.
19 The file has too many accelerated code segments.
20 The file has invalid resident areas in accelerated code.
21 Accelerator header in the file is invalid.
22 Either the UC (user code) option or the UL (user library) option was not used
when the file was accelerated.
23 File has entry in TNS/R fixup list with invalid external entry-point (XEP) index
value or invalid code address value.
24 Accelerated file has external procedure identifier list (EPIL), internal procedure
identifier list (IPIL), or external entry-point (XEP) table with incorrect format.

6 -7
3xx (page 2 of 4)
Subcode Meaning
25 UC (user code) or UL (user library) was accelerated using the wrong
Accelerator option (UC, UL, SC, or SL).
26 The file was accelerated with an incompatible version of the Accelerator.
27 The file has an invalid callable gateway (GW) table.
28 The program file contains processor-specific code that cannot be run on the
current processor.
29 Fixup of accelerated code was attempted in an object file that was not
accelerated.
30 An internal structure of the file contains an error.
32 An internal structure of the file has an entry point value of 0.
34 The list of unresolved procedure names contains an error.
35 The fixup computed an invalid file offset to the code area.
36 The file has an invalid fixup item.
38 The instruction at a call site is not the type expected for its fixup item.
39 Not used.
40 A virtual address specified in an ELF file is outside its allowed range. For
example, a text or data segment is specified at an address not valid for this type
of file.
41 Not used.
42 The code area or data area is too large.
43 The file either has a gateway (GW) table but no callable procedures or has
gateways that are not in the (GW) area.
44 The file codes of the program file and library file do not match. (Not used since
G06.12; see 5 and 6 instead.)
45 The file being started can run only in the Guardian environment and it is being
started in the OSS environment, or vice versa.
46 Either the TNS program or the TNS User library (but not both) expected the
library to contain global data.
47 Either the TNS program needs to import data from the TNS User library and the
library is not exporting any data, or the library needs global data space and the
program is not providing it.

6 -8
3xx (page 3 of 4)
Subcode Meaning
48 A TNS program file uses a TNS shared run-time library and is switching to a
new library, but the program was accelerated by an old version of the
Accelerator that does not support library data relocation at fixup time. Use a
version of the Accelerator provided with the D30.00 or later release of the
operating system.
49 A TNS object file has no code space.
50 The native object file is not loadable. Either it is a linkable file (such as unlinked
compiler output) or it is an incorrect type of loadable file (such as a DLL
encountered with a non-PIC program.)
51 The program or library file does not have a valid ELF header for execution on
this NonStop operating system. The file either is not targeted for this system, is
not an ELF file, or has been corrupted.
52 The file is not executable because it has more than one instance of a header
structure unique to HP NonStop Server systems. An error occurred during the
linking of the file.
53 An ELF file has a header specifying more than one instance of a segment that
should be unique. The file is corrupt or was not built by a valid linker.
54 The non-PIC ELF file is not loadable because it does not have a GINFO
header. An error occurred during the linking of the file, or it is corrupt.
55 An ELF file is lacking a required segment.
56 The file specifies too many shared run-time libraries (SRLs).
57 The file specifies duplicate shared run-time libraries (SRLs).
58 The shared run-time library (SRL) does not export any procedures.
60 An ELF library file was expected, but the file is in the Guardian file system and
does not have a file code of 700 or it is in the OSS file system and not
recognizable as an ELF file.
61 Two related structures in the ELF file have inconsistent lengths.
62 An attempt was made to spawn a shell script on a remote node.
63 An inconsistency exists in the set of public SRLs.
64 Some value (other than an address) specified in an ELF file is outside its range.
The file may be corrupted.
65 The current export-digest index specified in an ELF SRL file is greater than the
count of export digests in that file. The file is probably corrupted.
66 The count of export digests in an ELF SRL exceeds 256. The file is probably
corrupted.
67 A public SRL is marked to require a PIN less than 255; a PIN less than 255 is
not allowed.
68 One of the headers that is expected to be at the front of an ELF file did not fit
near enough to the front.

6 -9
3xx (page 4 of 4)
Subcode Meaning
69 This PIC ELF file is not supported on TNS/R systems: it is licensed, or it
contains callable functions.
70 The ELF file is too big (EOF > 231 bytes).
71 Some value in the TNS object file header is out of range; the file may be
corrupt.
72 The EF_TANDEM_INSTANCE_DATA value in the ELF header is not valid or
not consistent with the data program headers found; the file may be corrupt.
73 The p_flags in the ELF header for the resident text header are not as expected;
the file may be corrupt.
74 The loadfile has resident text, but no data constant segment and is not marked
data_resident. This is not supported.
75 The DLL has callable functions but also has unprotected data. This is not
supported.
76 An address to be stored into a relocation site does not fit in 32 bits.
77 The loadfile uses the 64 bit data model. It is not supported on this system.
78 The loadfile is an import library or implicit DLL, not a program, ordinary DLL, or
public DLL.

6- 10
14 UNDEFINED EXTERNALS
Cause. The process being started contains a call to an external procedure that is not
in the operating system code area, the user library (if applicable), or the application
code area.
Effect. Process creation occurs, and a message is printed on the home terminal. For
example:
PID: \SYS10.4,94 \SYS10.$XL.SVOL.TEST (TNS)
Prg: \SYS10.$XL.SVOL.TEST -> MY_PROC (PROC)
Undefined externals
where MY_PROC is the name of the undefined external procedure in the program file
\SYS10.$XL.SVOL.TEST.
When the operating system finds a call to an undefined external procedure, it replaces
the original call with a call to the PROCESS_DEBUG_ procedure. If the process tries
to invoke the undefined external during program execution, the program goes into the
debug state.
Recovery. Either correct the coding error, use Binder to add the procedure to the code
area or user library, or refer the call to a procedure that is already present in the
application program.
15 NO PCB AVAILABLE
Cause. All entries in the process control block (PCB) table for the processor were in
use or the process required a low process identification number (PIN) and none were
available.
16 UNABLE TO ALLOCATE MAP
Cause. Not enough space was available in the processor’s MAPPOOL to permit the
system monitor to generate the code and data-map copies required by the new
process (this might be a configuration problem).
Recovery. Too many processes might have been configured to run in the current
processor. Try a different processor or wait and then retry the existing call.

6- 11
17 UNLICENSED PRIVILEGED PROGRAM OR LIBRARY
Cause. The program file contains procedures having CALLABLE or PRIV attributes,
but the program file is not licensed to execute in privileged mode and the super ID was
not running the program.
Recovery. Have the super ID license or run the program under a locally validated
super ID.
18 LIBRARY CONFLICT
Cause. The call specified a library file, but the program was either already running
with a different library or was not running with a library.
This error is also generated if a user library file is specified when running an old TNS
program containing an “implicit user library.” (Prior to D30, a large TNS program file
could be created with 16 segments of user code and up to 16 additional segments
mapped as user library. Subsequently, the user-code limit and user-library limits were
raised to 32 segments each, and the binder stopped creating programs with an implicit
user library.)
Recovery. All non-PIC processes running a given program must use the same library.
Always specify the library file name to avoid the conflicts that can arise when the library
file is modified.
19 PROGRAM FILE AND LIBRARY FILE ARE THE SAME
Cause. The program file and library file are the same file.
Recovery. Select a different library file for the program.
20 PROGRAM FILE WITH AN ILLEGAL DEVICE SUBTYPE
Cause. An attempt was made to run a program file with an invalid device subtype. The
device subtype is an attribute stored in each program file. The process created from a
program file is assigned the device subtype stored in that program file. Only named
processes are allowed nonzero device subtypes.
Device subtypes in the range 1 through 15 are reserved for processes that are either:
• Created by the super ID
• Created from licensed program files

6- 12
• Created from program files owned and provided by the super ID

Subtypes 1 through 47 are reserved for HP use.
Recovery. Verify either that the call specifies a process name or that the program file
is marked as “must run named” by the Binder, the TAL RUNNAMED command, or the
C RUNNAMED pragma. If the call specifies a process name, either recompile or rebind
the program file to change the device subtype to an unrestricted value, rebind the
program file to run named, or contact the super ID.
21 PROCESS DEVICE SUBTYPE SPECIFIED IN BACKUP PROCESS

IS NOT THE SAME AS PRIMARY
Cause. The operating system tried to create a process, but the backup process device
subtype was not from the same program file as the primary process device subtype.
This error is probably a programming error.
Recovery. Find out which program file was being passed to PROCESS_LAUNCH_,
PROCESS_CREATE_, or PROCESS_SPAWN_, and correct the error.
22 BACKUP CREATION SPECIFIED, BUT CALLER IS UNNAMED
Cause. The operating system tried to create a backup process, but the caller was
unnamed.
Recovery. Run the caller as a named process, or rebind its program file as “must run
named” by using the Binder.
24 DEFINE CONTEXT PROPAGATION ERROR
Cause. An error occurred when existing DEFINEs were being propagated. The
error-detail information contains either a file-system error, a DEFINE error number, or
the error subcode 2, which indicates that an excessive number of DEFINEs were to be
propagated. File-system errors are described in Section 2, File-System Errors. DEFINE
errors are described in Section 4, DEFINE Errors.
Recovery. Retry the call. If errors recur, contact your service provider.
26 DYNAMIC IOP ERROR
Cause. This error is returned only to privileged callers, or unprivileged callers

attempting to use certain privileged features. On G series, it can be returned by an
attempt to create a D-series I/O Process.

6- 13

Recovery. Run software that is properly configured for your system.
27 ILLEGAL PFS SIZE
Cause. The size specified for the process file segment (PFS) is invalid.
Recovery. Specify a valid PFS size either in the RUN command or with the Binder or
nld utility. Then try again. (This error is not generated as of G06.12.)
29 UNABLE TO ALLOCATE A PRIV STACK FOR THE PROCESS
Cause. There was no segment available for the priv stack of a TNS/R native process.
30 UNABLE TO LOCK THE PRIV STACK FOR THE PROCESS
Cause. There was not enough physical memory free to lock the priv stack for a TNS/R
native process.
31 UNABLE TO ALLOCATE A MAIN STACK FOR THE PROCESS
Cause. There was no segment available for the main stack of a TNS/R native
process.
32 UNABLE TO LOCK THE MAIN STACK OF A NATIVE I/O PROCESS

THAT IS CREATED DYNAMICALLY
Cause. There was not enough physical memory free to lock the main stack for a
TNS/R native process. This error is only returned to some privileged callers.

6- 14
33 SECURITY INHERITANCE FAILURE
Cause. An error occurred during an attempt to obtain or propagate security identity

information. The error-detail information contains the number of the file-system error
that occurred.
file-system error returned in the error-detail information. If the file-system error
indicates an invalid operation or an invalid parameter, the problem might be caused by
a version mismatch between the NonStop operating system and Standard Security.
Correct your system configuration if this is the case. Otherwise, the problem is likely to
be an internal error that should be reported to your service provider.
35 INTERNAL PROCESS CREATION ERROR
Cause. An internal error occurred in the PROCESS_LAUNCH_,

PROCESS_CREATE_, or PROCESS_SPAWN_ procedure.
Recovery. Make a note of the error subcode returned in the error-detail information
and retry the call. Contact your service provider.
36 CHILD'S PFS ERROR
Cause. An error occurred in the specification of the process file segment (PFS). The
file-system error returned in the error-detail information. For example, if file-system
error 31, “Unable to obtain file-system buffer space,” is returned, specify a larger PFS
size for the process.
37 UNABLE TO ALLOCATE GLOBAL DATA FOR THE PROCESS
Cause. Process memory-segment limit exceeded.The system was unable to allocate

the global data segment of a native process. If non-zero, error-detail indicates:
1. Insufficient swap space available from KMSF
2. Overlap of Vaddrs
3. Too many segments loaded

6- 15
Effect. No process is created. The name of the file for which the error occurred may
have been reported to the home terminal.
Recovery. Contact your local service provider.
38 UNABLE TO LOCK IOP GLOBAL DATA FOR THE PROCESS
Cause. The system was unable to lock the global data segments of a native I/O
process. This error is only returned to some privileged callers.
Recovery. Stop processes or wait until the system load decreases before attempting
to run this process.
40 THE MAIN STACK MAXIMUM VALUE, SPECIFIED EITHER BY THE

PROCEDURE CALL OR BY THE OBJECT FILE, IS TOO LARGE
Cause. The main stack maximum value is too large.

Recovery. Either change the procedure call (if applicable) or use the nld utility to
change the object file to specify a new value that is within the valid range.
41 THE HEAP MAXIMUM VALUE, SPECIFIED EITHER BY THE

Cause. The heap maximum value is too large.

42 THE SPACE GUARANTEE VALUE, SPECIFIED EITHER BY THE

Cause. The space guarantee value is too large.


6- 16
43 THE PROCESS CREATION REQUEST SPECIFIES TWO FILES THAT

CONTAIN THE SAME SHARED RUN-TIME LIBRARY (SRL) NAMES
Cause. The process creation request specifies duplicate shared run-time libraries
(SRLs); the error-detail information contains the numbers of the duplicate SRLs in the
form xxyy (where xx is the first SRL and yy is the duplicate SRL).
44 UNABLE TO FIND A SHARED RUN-TIME LIBRARY (SRL)

SPECIFIED
BY THE PROGRAM FILE
Cause. The system was unable to find a shared run-time library (SRL) specified by the
program file; the error-detail information contains the SRL number of the SRL that
could not be found.
45 UNABLE TO FIND A SHARED RUN-TIME LIBRARY (SRL)

SPECIFIED
BY ANOTHER SRL
Cause. The system was unable to find a shared run-time library (SRL) specified by
another SRL; the error-detail information contains the SRL numbers of the two SRLs in
the form xxyy (where xx is the SRL that specifies the yy SRL).
46 THE PROCESS CREATION REQUEST SPECIFIES TOO MANY SHARED

RUN-TIME LIBRARIES (SRLS)
Cause. The process creation request specifies too many shared run-time libraries
(SRLs); the error-detail information contains the maximum number of SRLs that can be
specified.
Recovery. Use the original object files and the nld utility to re-create the program file,
using no more than 32 SRLs, the maximum number of SRLs permitted.

6- 17
47 THE PROGRAM FILE REQUIRES FIXUPS TO A SHARED RUN-TIME

LIBRARY (SRL) BUT THE PROGRAM FILE IS CURRENTLY RUNNING
Cause. The program file requires fixups to a shared run-time library (SRL) that is
unavailable because it is running; the error-detail information contains the SRL number
of the running SRL.
48 A SHARED RUN-TIME LIBRARY (SRL) REQUIRES FIXUPS TO

ANOTHER SRL
Cause. A shared run-time library (SRL) requires fixups to another SRL that is
unavailable because it is running; the error-detail information contains the SRL
numbers of the two SRLs in the form xxyy (where xx is the SRL that requires the fixup
to the running yy SRL).
49 SECURITY VIOLATION. THE PROGRAM FILE IS NOT LICENSED

BUT
A SHARED RUN-TIME LIBRARY (SRL) CONTAINING INSTANCE
DATA
IS LICENSED
Cause. A security violation occurred. The program file is not licensed but a shared
run-time library (SRL) is licensed and has instance data; the error-detail information
contains the SRL number of the licensed SRL.
Recovery. License the program file, if possible.
50 SECURITY VIOLATION. EITHER THE PROGRAM FILE OR SHARED

RUN-TIME LIBRARY (SRL) IS LICENSED BUT A SHARED RUN-
TIME
LIBRARY (SRL) IS NOT LICENSED
Cause. A security violation occurred. Either the program file or a shared run-time
library (SRL) is licensed, but another SRL is not licensed; the error-detail information
contains the SRL number of the unlicensed SRL.

6- 18
Recovery. Either remove the licensing of the program, if possible, or use the nld
utility to specify another SRL.
51 THE PROGRAM FILE REQUIRES A SYMBOL FROM A SHARED RUN-

TIME LIBRARY (SRL) BUT THE SRL IS NOT EXPORTING IT
Cause. The program file requires a symbol from a shared run-time library (SRL) but
the SRL is not exporting it; the error-detail information contains the SRL number of the
SRL that does not export the required symbol.
Recovery. Use the nld utility to specify another SRL that contains the desired data
block.
52 THE SPECIFIED VERSION, Z^VERSION, OF THE

ZSYS^DDL^PLAUNCH^PARMS STRUCTURE IS NOT SUPPORTED
Cause. (This error is returned only by PROCESS_LAUNCH_ .) The version

Z^VERSION of the ZSYS^DDL^PLAUNCH^PARMS structure is not supported.
Recovery. Recode or recompile and rebuild the program.
53 THE SPECIFIED VERSION, Z^VERSION, OF THE

ZSYS^DDL^PLAUNCH^PARMS STRUCTURE IS INCOMPATIBLE WITH
THE SPECIFIED LENGTH,Z^LENGTH,OF THE STRUCTURE
Cause. (This error is returned only by PROCESS_LAUNCH_ .) The version

Z^VERSION of the ZSYS^DDL^PLAUNCH^PARMS structure is incompatible.
Cause. A privileged interface to the PROCESS_LAUNCH_ procedure is invalid.


6- 19
55 THE SPECIFIED SPACE GUARANTEE CANNOT BE ALLOCATED
Cause. The space guarantee specified in Z^SPACE^GUARANTEE

(PROCESS_LAUNCH_) or Z^SPACEGUARANTEE (PROCESS_SPAWN_) cannot be
allocated.
Recovery. Either use the nld utility to shrink the guaranteed swap space
specification, use NSKCOM to add additional swap space for the target processor, or
run on another processor.
Cause. An internal error occurred.

Recovery. Retry the call. If this error recurs, contact your service provider.
57 A SHARED RUN-TIME LIBRARY (SRL) HAS UNDEFINED EXTERNALS
Cause. A shared run-time library (SRL) has undefined externals; the error-detail
information contains the SRL number of the SRL that has undefined externals.
Recovery. Check that the SRL versions match and obtain matching versions if they
don’t; otherwise, contact your service provider.



6- 20
60 SECURITY VIOLATION; A SHARED RUN-TIME LIBRARY (SRL)

CONTAINING CALLABLE PROCEDURES MUST BE LICENSED TO BE
USED BY CALLABLE OR PRIVILEGED CODE
Cause. A security violation occurred; a shared run-time library (SRL) containing

callable procedures must be licensed to be used by callable or privileged code.
Recovery. Rebuild the program file. If the error recurs, contact your service provider.
61 UNABLE TO ALLOCATE MEMORY FROM SYSTEM POOL
Cause. Memory for a needed buffer was unavailable in the system pool in the
processor.
Recovery. Try again or select another processor. If the problem persists, run the
PEEK program in the processor where process creation failed and determine new
memory needs.
62 SYMBOLIC REFERENCE TARGET/SOURCE TYPE MISMATCH
Cause. An attempt was made to resolve a procedure address to a data area or a data
address to a procedure area.
example:
Native UL Symbolic Reference Error: Target/Source Type Mismatch:
Recovery. Use the original object files and the nld utility to recreate the program file.

6- 21
63 EXTERNAL DATA REFERENCES NOT RESOLVED TO ANY USER/SYSTEM

LIBRARY
Cause. An anonymous data symbol reference was not located in a UL, Native UL, or
system library.
example:
Recovery. Either correct the coding error, use nld utility to add the data to the data
area or user library, or refer the call to the data that is already present in the application
program.
64 UNABLE TO HONOR FLOATTYPE ATTRIBUTE
Cause. The system couldn't honor the floattype attribute requested by a process-
creation request. The error-detail information contains an error subcode that indicates
the nature of the error. These error subcodes are listed in Table 6-3.
Recovery. Take corrective action as indicated by the subcode. As of G06.20 this code
is reported only for a program file; earlier releases use it also for libraries. These error
subcodes are listed in Table 6-3.

6- 22
Table 6-3. Error Subcodes for Process Creation Error 64

Subcode Meaning
1 IEEE floating-point not supported by processor.
Run your program on a processor model that supports IEEE floating-point
operation. If the process does not operate on floating-point data, but was
mistakenly marked as using IEEE floating-point, use the linker to force the
floattype attribute to tandem_float. The floattype attribute can be specified to
the linker using the -set command at link time, or the -change command to
modify in an existing file. The linker is the nld utility for a non-PIC program, or
the ld utility for a PIC program.
2 Unrecognized floating-point specification in file.
The program or library file specified to create the process contains incorrect
or corrupt information about the floattype. Recompile the object file(s) using a
valid floattype.
3 Floattype attribute of user library conflicts with that of the program (pre-
G06.20 only).
Recompile and relink the object file(s) using the same floattype attribute for
all, or if the library does not operate on floating-point data, force its floattype
attribute to neutral_float. The floattype attribute can be specified to the nld
utility using the -set command at link time, or the -change command to modify
in an existing file.
65 REFERENCED SRL IS ALREADY IN USE
Cause. A shared run-time library (SRL) requires a symbol from another SRL but
cannot be referenced because that SRL is already in use.
Recovery. Use the nld utility to specify the another SRL that contains the correct
symbol.
66 UNABLE TO HONOR FLOATTYPE ATTRIBUTE OF A USER LIBRARY
Cause. The system couldn't honor the floattype attribute for the user library requested
by a process-creation request. The error-detail information contains an error subcode
that indicates the nature of the error. The error-detail indicates:
2: Unrecognized floating-point specification in file.
4: This library specified Tandem floating-point, which mismatches the program.
5: This library specified IEEE floating-point, which mismatches the program.
Effect. No process is created. If the user library is PIC, its name will have been
reported to the home terminal.
Recovery. Take corrective action as indicated by the subcode.
For subcode 2: The library file specified to create the process contains incorrect or

6- 23
corrupt information about the floattype. Recompile the object file(s) using a valid
floattype.
For subcode 4: Recompile and relink the object file(s) using the same floating-point
type for all, or if the library does not operate on floating-point data, force its floattype
attribute to neutral_float. The floattype attribute can be specified to the linker using the
-set command at link time, or the -change command to modify in an existing file. The
linker is the nld utility for a non-PIC user library (SRL), or the ld utility for a PIC user
library (DLL). Note that the library specified Tandem floating-point.
type for all, or if the library does not operate on floating-point data, force its floattype
attribute to neutral_float. The floattype attribute can be specified to the linker using the
-set command at link time, or the -change command to modify in an existing file. The
linker is the nld utility for a non-PIC user library (SRL), or the ld utility for a PIC user
library (DLL). Note that the library specified IEEE floating-point.
67 UNABLE TO HONOR FLOATTYPE ATTRIBUTE OF A DLL
Cause. The system couldn't honor the floattype attribute for a DLL. The error-detail
information contains an error subcode that indicates the nature of the error. The error-
detail indicates:
2: Unrecognized floating-point specification in file.

4: This library specified Tandem floating-point, which mismatches the program.
5: This library specified IEEE floating-point, which mismatches the program.
Effect. No process is created. The name of the DLL for which the error occurred will
Recovery. Take corrective action as indicated by the subcode.
For subcode 2: The DLL file specified to create the process contains incorrect or
corrupt information about the floattype. Recompile the object file(s) using a valid
floattype.
type for all, or if the DLL does not operate on floating-point data, force its floattype
attribute to neutral_float. The floattype attribute can be specified to the ld utility; use the
-set command at link time, or the -change command to modify in an existing file. Note
that the DLL specified Tandem floating-point.
type for all, or if the DLL does not operate on floating-point data, force its floattype
attribute to neutral_float. The floattype attribute can be specified to the ld utility; use the
-set command at link time, or the -change command to modify in an existing file. Note
that the DLL specified IEEE floating-point.

6- 24
68 A DEFINE = _RLD IS PRESENT BUT NOT VALID
Cause. error-detail indicates:

0 - The define is not of class SEARCH.
2055 - An attribute other than CLASS or SUBVOL0 is specified.
Otherwise - as reported by the DEFINEINFO function.
Recovery. Take action as indicated by the error subcode.
69 A FILE SYSTEM ERROR WAS ENCOUNTERED ON THE RUNTIME

LOADER LIBRARY (rld)
Cause. A file system error occurred on the runtime loader library while attempting the
process creation. The error-detail contains a file-system error number.
Recovery. Refer to Section 2, File-System Errors, for corrective action for the file-
system error returned in the error-detail information.
70 INVALID FILE FORMAT IN THE RUNTIME LOADER LIBRARY (rld)
Cause. An error was detected in the file format of the runtime loader library (RLD).
The error-detail information contains an error subcode that indicates the invalid file
format error. These errors are listed in Table 6-2.
Recovery. Take corrective action as indicated by the subcode. In many cases, the
object file was built improperly or corrupted; it should be relinked and perhaps
recompiled.
71 ERROR LOADING OR RUNNING THE RUNTIME LOADER (rld)
Cause. A failure occurred while loading or running the runtime loader (RLD). Error
details other than the following indicate that rld was not constructed or installed
correctly.
9- The process abended while rld was running.
10- The processed stopped while rld was running.
11- RLD is licensed.
12- rld returned an out-of-range error value to the operating system.
16- The export digest of the file does not match the export digest of the impImp file in
memory.
19- The user attempted to use RLD as a user library. This is not supported.
22- RLD began processing, but did not complete the update of a loadfile.

6- 25

Recovery. For subcode 11, unlicense RLD and reload the system. RLD should not
require being licensed. For subcode 16, verify that the copy of RLD you used is the
correct one. The set of public DLLs, including ZRLDDLL, should be installed and
preset for use with the implicit DLLs in this SYSnn. There is no recovery for subcode
error 19. For other subcodes, contact your service provider, reporting the error code
and the error detail.
72 THE RUNTIME LOADER (rld) REPORTED AN INTERNAL OR

UNRECOGNIZED ERROR.
Cause. The runtime loader (RLD) reported an internal or unrecognized error.

Recovery. Contact your service provider and report the error and also the error detail.
74 UNRESOLVED REFERENCE TO A FUNCTION
Cause. There was an unresolved reference to a function, so the process was not
created. Contrast with error 14, which is a warning. (For DLLs and their client
programs, unresolved function references are disallowed by default, but other options
can be specified at link or run time.)
Effect. No process is created. Information such as the name of the unresolved
function and the caller of it may be reported to the home terminal.
Recovery. Use the information reported on the home terminal, and attempt to fix the
problem. (Either fix the code in the caller or the callee, or get a version of code that
does not make the faulty call.)
75 A FILE SYSTEM ERROR WAS ENCOUNTERED ON A DLL
Cause. A file system error occurred on a DLL while attempting the process creation.
The error-detail contains a file-system error number.
Recovery. Refer to Section 2, File-System Errors, for corrective action for the file-
system error returned in the error-detail information.
76 INVALID FILE FORMAT IN A DLL
Cause. An error was detected in the file format of a DLL. The error-detail information
contains an error subcode that indicates the invalid file format error. These errors are
listed in Table 6-2.

6- 26
recompiled.
77 AN OBJECT FILE COULD NOT BE LOADED
Cause. error-detail contains the details:

1 - A DLL requires a PIN < 255 in a process with a higher PIN.
2 - A licensed or privileged object file does not use localized import.
3 - A public library requires fixed address space that is unavailable in this process.
4 - Insufficient address range is available to load the file.
5 - The loader lacks authority to load a licensed object file.
6 - The process has exceeded the maximum number of memory segments.
7 - A Globalized symbol required by a PIC program or DLL is defined in more than
one public library. (Globalized symbols are those identified by the native C++ compiler
as possibly having multiple definitions, of which the system must select one to be used
by all object modules loaded.)
8 - The C++ version of the specified library conflicts with one or more loadfiles loaded
for this process. Mixed C++ dialect versions are not allowed.
9- The loadfile is not licensed; license is required.
10- A DLL for this process is licensed or privileged and has unprotected data which
requires that all loadfiles in the process be licensed. There is at least one unlicensed
loadfile in the process.
11- A licensed DLL or a privileged program refers to an unlicensed DLL.
12- A process with a licensed but unprivileged program attempted to load an
unlicensed non-public DLL.
13- A licensed or privileged loadfile has globalized symbols.
14- The loadfile was specified as dataResident and is not licensed, has no callable
functions, and is not a program with a priv entry point.
15- This process can only be run by the local super ID.
16- RLD failed to pass to the operating system a function pointer necessary to process
the initialization functions, constructor callers, destructor callers, or termination
functions specified to the linker.
17- The specified loadfile was built with linker option -no_runtime_fixup, but it is not
preset to load with the symbol bindings available on this system or in this process.
18- The loadfile was built to use an Application Binary Interface version that is not
supported.
19- A process with OSS file privileges attempted to load an ordinary DLL with lesser
OSS file privileges.
Effect. No process is created. The name of the file involved may be reported on the
home terminal.
Recovery. This depends on the error detail.
1- Specify 'highpin off' when running the Program (or rebuild the DLL to not need a
low pin)
6- 27
2- If this file must be licensed, you must rebuild it specifying localized import rather
than globalized or semi-globalized
3 - Probably indicates a problem in the way the library was built. Contact your service
provider. Report the file name, error code, and the error detail.
4 - Probably indicates that either too many segments or too many large segments are
already loaded into the process. To get around this, you may have to try to run the
program with fewer libraries.
5 - If this file must be licensed, it cannot be loaded without “RLD authority”; you cannot
open it dynamically using dlopen().
6 - Probably indicates that either too many segments or too many large segments are
already loaded into the process. To get around this, you may have to try to run the
program with fewer libraries.
7 - If an inadvertent collision of symbol names occurred, the PIC source must be
changed, recompiled, and relinked using a unique symbol. Provide this information to
the supplier of the PIC program or DLL. If the same symbol is exported by two public
libraries, and it is proper for both of them to be loaded into the same process, report
the problem to your service provider. Avoid loading redundant or unnecessary public
SRLs; for example, ensure that all files loaded in the process depend upon the same
C++ runtime and/or tools.h++++ libraries.
8 - Don't load both a Version 3 C++ loadfile and any Version 1 or Version 2 C++
loadfile into the same process. If possible, relink a file that used a conflicting C++
version file so that it uses the same C++ version as other loadfiles loaded into the
process.
9-You must license this loadfile to continue using it in this manner.
10- Avoid this situation either by avoiding unprotected data or by licensing the other
loadfiles in the process when feasible and appropriate. Note that libraries with
unprotected data cannot be licensed and then used with an unlicensed program. A
loadfile’s data will be protected if it has no variable data, or if it is constructed using the
-instance_data data2hidden command to the linker. The former makes the library’s
global variables read-only at user privilege; the latter makes them invisible at user
privilege. Therefore, the variables are fully accessible only from privileged code. Note
that licensed DLLs must have localized import and cannot have globalized symbols.
11- License the referenced DLL when possible. Note that libraries with unprotected
data cannot be licensed and then used with an unlicensed program. This does not
apply to licensing most public DLLs such as the language-support libraries.
12- License the DLL when possible. Note that libraries with unprotected data cannot be
licensed and then used with an unlicensed program. This does not apply to licensing
most public DLLs such as the language-support libraries.
13- To license this loadfile, it cannot contain global symbols.
14- If the loadfile must be dataResident, license it. Otherwise relink it without specifying
that it be dataResident.
15- The process must be created by a person or process using the locally
authenticated super ID.
16- Contact your service provider. Report the error code and the error detail.
17-Relink the file if the environment changes, or don't set the option if the same file
must be used simultaneously in different environments. Contact your service provider.
Report the file name, error code, and the error detail.

6- 28
18- Recompile the source code with current compilers and relink the loadfile with the
current linker. Contact your service provider. Report the file name, error code, and the
error detail.
19- Set the minimum object file privileges on loading DLL.
78 AN UNSUPPORTED OPERATION WAS ATTEMPTED

1 - A PIC program attempted to load an SRL other than a public SRL.
2 - A PIC program or DLL was licensed.
3 - A User Library supplied for a PIC program was not a DLL.
4 - A public SRL requires another library that is not a public SRL.
5 - A public SRL is not a hybrid DLL-SRL.
The error-detail values 4 and 5 imply an incorrect installation of the public SRLs.
6 - The specified library uses Version 1 C++, which is not supported with a PIC
Program.
7- RLD cannot be on the liblist of any file loaded when a) the program has a priv entry
point, or b) the program file is licensed.
Effect. No process is created. The name of the file involved may be reported on the
home terminal.
Recovery. This depends on the error detail.
1 - Only attempt to use public SRLs and DLLs when running a PIC program. The home
terminal may indicate both the file that attempted to use the invalid file and the invalid
file. Replace one of these files as needed to follow the requirements.
2 - Licensed PIC Programs and DLLs are not supported on this system. The home
terminal may indicate which file was used. If it did not need to be licensed, unlicense it,
and try again.
3 - Only a DLL file can be used as a user library with a PIC program. If possible,
replace the user library with a valid DLL file, and try again.
4 and 5 - An incorrect installation of the public SRLs was done. Contact your service
provider.
6 - Do not include any loadfile that requires a Version 1 C++ loadfile when loading a
PIC Program. If possible, relink the loadfile that used the Version 1 loadfile so that it
uses a Version 2 or Version 3 C++ loadfile in place of the Version 1 C++ loadfile.
7- Fix the application so that RLD is not included on a liblist. Such a process cannot
call dlopen().
79 A RESOURCE LIMITATION WAS DETECTED BY THE RUNTIME LOADER

(rld)

1 - The RLD heap attempted to exceed available KMSF space.
2 - The RLD heap attempted to exceed its allocated address range.
3 - RLD limit of handles was exhausted. This error is reported for dynamic loading, not
at process creation.

6- 29
4 - The process limit for keys is exhausted. Keys are an operating-system resource
used by the loader.
Recovery. For detail 3:
Each call on dlopen() generates a new handle; avoid a large number of calls specifying
the same path argument.
For the others:
Contact your service provider, and report the error and the error detail.
80 ERROR LOADING A DROP-IN PROGRAM
Cause. A failure occurred while loading or running a program that must be “dropped
in” rather than run through RLD. Error details indicate that the drop-in program is not
constructed or installed correctly.
memory.
Recovery. For subcode 16 verify that the right copy of the drop-in program is being
used. It must have been preset for use with the implicit DLLs in the SYSnn. For other
subcodes, contact your service provider, reporting the error code and the error detail.
81 ERROR LOADING OR RUNNING THE TNS EMULATOR
Cause. A failure occurred while loading or running the TNS Emulator. Error details
other than the following indicate that the TNS Emulator is not constructed or installed
correctly.
memory.
Recovery. For subcode 16 verify that the right copy of the TNS emulator is being
used. The set of public DLLs, including ZTNSDLL, should be installed and preset for
use with the implicit DLLs in this SYSnn.For other subcodes, contact your service
provider, reporting the error code and the error detail.
82 A DEFINE=_TNS IS PRESENT BUT NOT VALID
Cause. A DEFINE is recognized by the systems, but it is not a valid DEFINE. Error
details are:
0 - The define is not class SEARCH
2055 - An attribute other than CLASS or SUBVOL0 is specified
All other details are as reported by the DEFININFO function.

6- 30

Recovery. Take corrective action as indicated by the error subcode.
83 A FILE SYSTEM ERROR WAS ENCOUNTERED ON THE TNS EMULATOR
Cause. A file system error occurred on the TNS Emulator while attempting process
creation. The error detail contains a file system error number.
Recovery. Refer to Section 2, File-System Errors for further details about the file-
system error returned in the error detail information.
84 INVALID FILE FORMAT IN THE TNS EMULATOR
Cause. An error is detected in the file format of the TNS Emulator. The error detail
information contains an error subcode that indicates the invalid file format. These
errors are listed in Table 6-2 on page 6-7.
Recovery. Take corrective action as indicated by the error subcode. In many cases,
the object file was improperly built or corrupted. Consider relinking and recompiling the
file.
99 ERROR IN PRELOADING A PUBLIC DLL
Cause. A failure occurred while attempting to preload a public DLL specified in the
zreg file. Error details are:
1- The export digest of the public DLL is not a match to the export digest found in the
specified zreg file.
2- The license value of the public DLL is not a match to the license value found in the
specified zreg file.
3- The public DLL is licensed and has unprotected data.
4- The public DLL is not preset.
5- The public DLL has a priv or callable Main procedure.
6- The public DLL does not support highpin.
7- The public DLL is not owned by super ID.
8- The public DLL has callable procedures and is not licensed.
9- A public DLL with this name has already been preloaded (duplicate name in zreg).
10- The text, data, or gateway of the public DLL overlaps that of another public DLL.
11- The export digest attribute for this public DLL is missing from the zreg file.
Effect. The public DLL is not preloaded into the public library table.

6- 31
Recovery. Contact your service provider reporting the error code and error detail if
you are using a zreg file and public DLL set that was created by PLINSTL and has not
been modified. If you have been attempting to adjust the public DLL set manually (not
recommended), fix or replace the loadfile based on the problem reported in the error
detail, or run PLINSTL to create an appropriate set of public DLLs.
104 UNABLE TO CREATE OSS PROCESS
Cause. (This error is returned only by PROCESS_SPAWN_.) An OSS process cannot

be created because there are insufficient resources.
Recovery. The recovery method depends on the error-detail code received:
Recovery. 1: Request TNS floating point in your creation request, or upgrade to a
release and processors that support IEEE floating point.
Recovery. 2: Change your process create request to one of the supported floating
point types.
Recovery. 3: Recompile the object file(s) with the same floating point type for all.
106 OSS START INTERPRETER ERROR
Cause. (This error is returned only by PROCESS_SPAWN_.) The

oss-program-file parameter is an interpreter shell script that cannot be started.
The error-detail information contains the error number of the file-system error that
occurred.
Effect. No interpreter script file is started.
Recovery. Refer to Section 2, File-System Errors, for corrective action for the error
returned in the error-detail information.
107 OSS STATIC VARIABLE ERROR
Cause. (This error is returned only by PROCESS_SPAWN_.) An error occurred during

the allocation of user data space for static variables used by the system library. The
Recovery. Refer to Section 2, File-System Errors, for corrective action for the error
returned in the error-detail information.
108 THE CALLING PROCESS IS NOT OSS
Cause. A Guardian process attempted to call an OSS-only process-creation function:

fork, exec..., tdm_fork, tdm_exec..., tdm_spawn...
6- 32

Recovery. Make the call from an OSS process.
110 OSS CURRENT WORKING DIRECTORY ERROR
Cause. (This error is returned only by PROCESS_SPAWN_.) The current working

directory for the new process could not be obtained. The current working directory is
specified in the ZSYS-DDL-FDINFO.Z-CWD field of the fdinfo parameter. If it is null,
then the caller’s current working directory is used. If the caller does not have a current
working directory, then the caller’s default volume is used. The error-detail information
contains either an OSS errno value or the error number of the file-system error that
occurred.
Recovery. If the error-detail information contains the OSS errno value ENOENT
(4002), then the specified current working directory does not exist or does not exist on
the current system. Make sure that the default volume exists and is local to the calling
process. For other error values, refer to Section 2, File-System Errors, for corrective
action for the error returned in the error-detail information.
111 OSS FILE DESCRIPTOR DUP ERROR
Cause. (This error is returned only by PROCESS_SPAWN_.) One of the file

descriptors specified to be duplicated with the OSS dup() function that was passed in
the fdinfo parameter could not be duplicated. The
parameter contains the index into the ZSYS-DDL-FDINFO.Z-FDENTRY structure to
identify which of the file descriptors failed to be duplicated. The
parameter contains an OSS dup() function errno value.
Recovery. Determine corrective action by referring to the Open System Services
System Calls Reference Manual for a description of the OSS dup() function errno
values.
112 OSS FILE DESCRIPTOR OPEN ERROR
Cause. (This error is returned only by PROCESS_SPAWN_.) One of the file

descriptors specified to be opened with the OSS open() function that was passed in
the fdinfo parameter could not be opened. The
parameter contains the index into the ZSYS-DDL-FDINFO.Z-FDENTRY structure to
identify which of the file descriptors failed to be opened. The
parameter contains an OSS open() function errno value.

6- 33

Recovery. Determine corrective action by referring to the Open System Services
System Calls Reference Manual for a description of the OSS open() function errno
values.
113 OSS FILE DESCRIPTOR OPEN TIMEOUT ERROR
Cause. (This error is returned only by PROCESS_SPAWN_.) The timeout value in the
ZSYS-DDL-FDINFO.Z-TIMEOUT field of the fdinfo parameter was reached before
the file descriptors specified in the fdinfo parameter could be opened.
Recovery. Recovery, if any, is application dependent. Recovery actions include but
are not limited to the following: select a longer timeout value, select a timeout value of
0 to allow the PROCESS_SPAWN_ call to return to the application without waiting for
files to be opened or duplicated, select a different file to be opened or duplicated, or
select a different open flag such as OSSOPEN^NONBLOCK to open the file for
nonblocked access.
114 CANNOT CREATE OSS PRIV PROCESS
Cause. (This error is returned only by PROCESS_SPAWN_.) The process cannot be

created because privileged OSS processes are not supported.
Recovery. Specify a program file that is not to be run as a privileged process when
calling the PROCESS_SPAWN_ procedure.
115 UNABLE TO ALLOCATE GLOBAL DATA OR HEAP FOR THE PROCESS

(PROCESS_SPAWN_ ONLY)
Cause. (This error is returned only by PROCESS_SPAWN_ .) The system was unable
to allocate global data or heap for the process.
Recovery. Run on another processor or try later.
116 UNABLE TO PROPAGATE SHARED RUN-TIME LIBRARY (SRL) DATA

(PROCESS_SPAWN_ ONLY).
Cause. (This error is returned only by PROCESS_SPAWN_ .) The system was unable
to propagate shared run-time library (SRL) data.

6- 34
Recovery. Run on another processor or try later.
3xx INVALID FILE FORMAT ON SHARED RUN-TIME LIBRARY (SRL)

NUMBER XX
Cause. An error was detected in the file format of shared run-time library(SRL)
number xx. The error-detail information contains an error subcode that indicates the
invalid file format error. See Table 6-2 on page 6-7 for possible values.
recompiled.
5xx FILE-SYSTEM ERROR ON SHARED RUN-TIME LIBRARY (SRL)

NUMBER XX
Cause. There was a file-system error on shared run-time library (SRL) number xx;
error-detail information contains a file-system error number.
file-system error returned in the error-detail information. If the problem persists, contact
your service provider.
1100-1499 INTERNAL ERROR WITHIN A MODULE OF THE OPERATING

SYSTEM
Cause. An internal error was detected within a module of the operating system.
Recovery. Report the internal error and any error detail that accompanied it to your
service provider.

6- 35
Process Creation Errors Error Lists
Error Lists
subsystem, you might receive an error list in a response. HP subsystems return such
an error list when, in performing your request, they call the PROCESS_LAUNCH_ or
PROCESS_CREATE_ procedure directly or indirectly and an error occurs on the call.
The standard SPI token ZSPI-TKN-PROC-ERR, which is present in every
PROCESS_LAUNCH_ and PROCESS_CREATE_ error list, identifies the procedure
that was called. Its value is ZGRD-VAL-PROCESS-CREATE (22) for
PROCESS_CREATE_ and ZGRD-VAL-PROCESS-LAUNCH (25) for
PROCESS_LAUNCH_.
constructing a PROCESS_LAUNCH_ or PROCESS_CREATE_ error list:
• Optionally include any or none of the conditional tokens listed in the error-list
description.
information about creating error lists, additional information about tokens and token
Programming Manual.

6- 36
Process Creation Errors 22: ZGRD-VAL-PROCESS-CREATE
22: ZGRD-VAL-PROCESS-CREATE
A call to PROCESS_CREATE_ failed due to an error.

Conditional Tokens
ZGRD-TKN-XOBJECTFILE token-type ZSPI-TYP-STRING.

ZGRD-TKN-XLIBRARYFILE token-type ZSPI-TYP-STRING.
ZGRD-TKN-XSWAPFILE token-type ZSPI-TYP-STRING.
ZGRD-TKN-EXTSWAPFILE token-type ZSPI-TYP-STRING.
ZGRD-TKN-XCPU token-type ZSPI-TYP-UINT.
ZGRD-TKN-PROCESSHANDLE token-type ZSPI-TYP-PHANDLE.
ZGRD-TKN-ERRORDETAIL token-type ZSPI-TYP-INT.
ZGRD-TKN-NAMEOPTION token-type ZSPI-TYP-ENUM.
ZGRD-TKN-XPROCESSNAME token-type ZSPI-TYP-STRING.
ZGRD-TKN-PROCESSDESCR token-type ZSPI-TYP-STRING.
ZGRD-TKN-XHOMETERM token-type ZSPI-TYP-STRING.
ZGRD-TKN-JOBID token-type ZSPI-TYP-INT.
ZGRD-TKN-XMEMORYPAGES token-type ZSPI-TYP-INT.
ZGRD-TKN-CREATEOPTIONS token-type ZSPI-TYP-UINT.
ZGRD-TKN-DEBUGOPTIONS token-type ZSPI-TYP-UINT.
ZGRD-TKN-PFS-SIZE token-type ZSPI-TYP-INT2.
PROCESS_CREATE_. This error code indicates the outcome of the process-creation
attempt.
ZSPI-TKN-PROC-ERR is the procedure code. Its value is ZGRD-VAL-PROCESS-
CREATE (22).
Conditional Tokens
ZGRD-TKN-XOBJECTFILE is the file name of the program file to be run.
ZGRD-TKN-XLIBRARYFILE is the name of the user library file to be used by the
process, if one was explicitly named.

6- 37
Process Creation Errors 22: ZGRD-VAL-PROCESS-CREATE
ZGRD-TKN-XSWAPFILE is the data swap-file name, which is passed for informational

purposes only. This swap file is not used. Processes swap to a file that is managed by
the Kernel-Managed Swap Facility.
ZGRD-TKN-EXTSWAPFILE is the name of the file to be used as the swap file for the
default extended data stack segment of the process. It is possible that swap space is
managed by the Kernel-Managed Swap Facility (KMSF), in which case, this token is
not applicable. Furthermore, this token is applicable to TNS processes only.
ZGRD-TKN-XCPU is the processor number for the new process.
ZGRD-TKN-PROCESSHANDLE is the process handle returned by
PROCESS_CREATE_. If it is null, no process was created.
ZGRD-TKN-ERRORDETAIL is the error-detail information returned by
PROCESS_CREATE_ for some classes of errors.
ZGRD-TKN-NAMEOPTION is the specified process name option.
ZGRD-TKN-XPROCESSNAME is the new process name.
ZGRD-TKN-PROCESSDESCR is the process descriptor returned by
PROCESS_CREATE_. If its length is 0, no process was created.
ZGRD-TKN-XHOMETERM is the name of the home terminal for the new process.
ZGRD-TKN-JOBID is the job ID to be assigned to the new process.
ZGRD-TKN-XMEMORYPAGES is the size of the data stack in pages. This token is
applicable to TNS processes only.
ZGRD-TKN-CREATEOPTIONS contains the specified process-creation options.
ZGRD-TKN-DEBUGOPTIONS contains the specified debugging options.
ZGRD-TKN-PFS-SIZE the size, in bytes, of the process file segment (PFS), if a
nondefault value was specified.
Effect
The effect of this error depends on the PROCESS_CREATE_ error code returned.
Recovery
Follow the recovery procedure for the returned PROCESS_CREATE_ error code as

6- 38
Process Creation Errors 25: ZGRD-VAL-PROCESS-LAUNCH
25: ZGRD-VAL-PROCESS-LAUNCH
A call to PROCESS_LAUNCH_ failed due to an error.

ZSPI-TKN-ERRORDETAIL token-type ZSPI-TYP-INT.
Conditional Tokens

ZGRD-TKN-XLIBRARYFILE token-type ZSPI-TYP-STRING.
ZGRD-TKN-XSWAPFILE token-type ZSPI-TYP-STRING.
ZGRD-TKN-EXTSWAPFILE token-type ZSPI-TYP-STRING.
ZGRD-TKN-XPROCESSNAME token-type ZSPI-TYP-STRING.
ZGRD-TKN-XHOMETERM token-type ZSPI-TYP-STRING.
ZGRD-TKN-PFS-SIZE token-type ZSPI-TYP-INT2.
ZGRD-TKN-MAINSTACK-MAX token-type ZSPI-TYP-INT2.
ZGRD-TKN-HEAP-MAX token-type ZSPI-TYP-INT2.
ZGRD-TKN-SPACE-GUARANTEE token-type ZSPI-TYP-INT2.
ZGRD-TKN-CREATE-OPT token-type ZSPI-TYP-INT2.
ZGRD-TKN-NAMEOPTION token-type ZSPI-TYP-ENUM.
ZGRD-TKN-DEBUGOPTIONS token-type ZSPI-TYP-UINT.
ZGRD-TKN-XCPU token-type ZSPI-TYP-UINT.
ZGRD-TKN-XMEMORYPAGES token-type ZSPI-TYP-INT.
ZGRD-TKN-JOBID token-type ZSPI-TYP-INT.
ZGRD-TKN-PROCESSHANDLE token-type ZSPI-TYP-PHANDLE.
ZGRD-TKN-PROCESSDESCR token-type ZSPI-TYP-STRING.
PROCESS_LAUNCH_. This error code indicates the outcome of the process-creation
attempt.
ZSPI-TKN-PROC-ERR is the procedure code. Its value is ZGRD-VAL-PROCESS-
LAUNCH (25).
ZSPI-TKN-ERRORDETAIL is the error-detail information returned by
PROCESS_LAUNCH_.

6- 39
Conditional Tokens
ZGRD-TKN-XOBJECTFILE is the file name of the program file to be run.
ZGRD-TKN-XLIBRARYFILE is the name of the user library file to be used by the
process, if one was explicitly named.
ZGRD-TKN-XSWAPFILE is the data swap-file name, which is passed for informational
purposes only. This swap file is not used. Processes swap to a file that is managed by
the Kernel-Managed Swap Facility.
ZGRD-TKN-EXTSWAPFILE is the name of the file to be used as the swap file for the
default extended data stack segment of the process. It is possible that swap space is
managed by the Kernel-Managed Swap Facility (KMSF), in which case, this token is
not applicable. Furthermore, this token is applicable to TNS processes only.
ZGRD-TKN-XPROCESSNAME is the new process name.
ZGRD-TKN-XHOMETERM is the name of the home terminal for the new process.
ZGRD-TKN-PFS-SIZE is the size, in bytes, of the process file segment (PFS), if a
nondefault value was specified.
ZGRD-TKN-MAINSTACK-MAX is the maximum size, in bytes, of the process main
stack, if a nondefault value was specified.
ZGRD-TKN-HEAP-MAX is the maximum size, in bytes, of the process heap, if a
nondefault value was specified. This token is applicable only to TNS/R native
processes.
ZGRD-TKN-SPACE-GUARANTEE is the maximum size, in bytes, of the amount of space
to be reserved with the Kernel-Managed Swap Facility (KMSF), if a nondefault value
was specified.
ZGRD-TKN-CREATE-OPT contains the specified process creation options.
ZGRD-TKN-NAMEOPTION is the specified process name option.
ZGRD-TKN-DEBUGOPTIONS contains the specified debugging options.
ZGRD-TKN-PRIORITY is the initial execution priority under which the new process is
to run, if a nondefault value was specified.
ZGRD-TKN-XCPU is the processor number for the new process, if a nondefault value
was specified.
ZGRD-TKN-XMEMORYPAGES is the minimum number of memory pages to be allocated
to the new process for user data, if a nondefault value was specified. This token is
applicable only to TNS processes.
ZGRD-TKN-JOBID is the job ID to be assigned to the new process.
ZGRD-TKN-PROCESSHANDLE is the process handle associated with the new process
that was returned by PROCESS_LAUNCH_. If it is null, no process was created.

6- 40
ZGRD-TKN-PROCESSDESCR is the process descriptor of the new process that was

returned by PROCESS_LAUNCH_. If its length is 0, no process was created.
Effect
The effect of this error depends on the PROCESS_LAUNCH_ error code returned.
Recovery
Follow the recovery procedure for the returned PROCESS_LAUNCH_ error code as

6- 41
7
PROCESS_GETINFOLIST_ Errors
The PROCESS_GETINFOLIST_ procedure is used to obtain detailed information
about a particular process or about processes within a processor that meet a specified
list of search criteria. The process of interest can be specified either by process handle
or by node name, processor number, and PIN.
The PROCESS_GETINFO_ procedure should be used when the caller requires only
selected information about a particular process. PROCESS_GETINFO_ returns many
of the same errors as PROCESS_GETINFOLIST_.
When PROCESS_GETINFOLIST_ returns information about one or more processes,
the error-detail parameter contains the number of attributes returned.
Error Codes
This subsection lists each PROCESS_GETINFOLIST_ procedure error code and
provides a description of each code.
0 NO ERROR
Cause. The information returned is for the processes specified; the error-detail
parameter contains the number of processes for which information has been returned
(which can be more than one process if in search mode).
Effect. None.
Recovery. None required.
1 A FILE SYSTEM ERROR OCCURRED
Cause. A file-system error occurred; the error-detail parameter contains the file-
system error number. File-system error 563 (buffer too small) is returned if the ret-
values-list buffer is too small to contain all of the requested information.
Effect. The procedure sets the error code and returns the error in the error-detail
parameter.
Recovery. Refer to Section 2, File-System Errors, for corrective action.

7 -1
PROCESS_GETINFOLIST_ Errors Error Codes
2 PARAMETER ERROR
Cause. The PROCESS_GETINFOLIST_ call contained an illegal combination of

options. The error-detail parameter contains the ordinal number of the first
(leftmost) parameter encountered whose option is in error.
Note that parameters are counted in TAL fashion, so nodename:length is lumped as

number 3 and processhandle is number 4. One reason this error can occur with detail
6 is if an attribute in ret-value-list specifies auxiliary data, which do not fit within the ret-
value-cnt.
Recovery. Correct the call to PROCESS_GETINFOLIST_ so that the options are
correct and in the proper order.
3 BOUNDS ERROR
Cause. There was a bounds violation on a reference parameter.

Effect. The procedure sets the error code; the error-detail parameter contains
the number of the parameter in error.
Recovery. Pass a correct reference address to PROCESS_GETINFOLIST_.
4 SPECIFIED PROCESS DOES NOT EXIST
Cause. The specified process does not exist or does not meet the search criteria.
Effect. The information returned is for higher-numbered processes; the error-
detail parameter contains the number of processes for which information has been
returned (which can be more than one process if in search mode).
Recovery. Refer to Section 2, File-System Errors, for corrective action.
5 UNABLE TO COMMUNICATE WITH PROCESSOR
possibly because the processor where the program was to be run did not exist or was
inoperable. The error-detail parameter contains the error number of the file-
system error that occurred.
6 UNABLE TO COMMUNICATE WITH NODE

7 -2
Cause. The process could not communicate with the node named, possibly because
the node did not exist or was inoperable. The error-detail parameter contains the
error number of the file-system error that occurred.
Recovery. Select another node, then try again.
7 NO MORE MATCHES
Cause. No more matches exist.

Effect. No more matches will be found; the error-detail parameter contains the
number of processes for which information has been returned (can be 0).
Recovery. None.
9 INVALID SEARCH CODE
Cause. The search attribute code was not valid; the error-detail parameter
contains the invalid code.
Effect. The search is not done.
Recovery. Correct the search attribute code, then try again.
10 INVALID SEARCH VALUE
Cause. The search value was not valid; the error-detail parameter contains the
associated attribute code.
Recovery. Correct the attribute value, then try again.
11 INVALID RETURN ATTRIBUTE CODE
Cause. The return attribute code was not valid; the error-detail parameter
contains the invalid code. This error can occur if the attribute index is unrecognized, or
if a non-zero auxiliary data size is specified for an attribute that does not require
auxiliary data.
Effect. The return is not done.
Recovery. Correct the return attribute code, then try again.
12 INVALID SEARCH OPTION
Cause. The search option was not valid.


7 -3
Recovery. Correct the search option, then try again.
13 - unused -
Cause. This number is not in current use.

Effect. N/A
Recovery. N/A
14 Invalid auxiliary data size specification in an attribute code.
Cause. The auxiliary data size specified was invalid. error-

detail contains the attribute code.
Effect. The procedure sets the error code and puts the attribute code in error-detail.
Recovery. Use the correct attribute code. The simplest way to do this is to use the
attribute identifier as defined in ZSYSC or ZSYSTAL. An attribute code contains two
fields: the low-order 12 bits hold the attribute index. The high-order 4 bits hold the
number of 16-bit words of auxiliary data that follow the attribute code in the ret-array-
list array. For example, attribute code ZSYS_VAL_PINF_LOADFILE_INFO is
121+4<<12, or 16505; the length value is 4 (this attribute requires eight bytes of
auxiliary data).
15 An iterative attribute was not the last attribute in ret-attr-list.
Cause. The iterative attribute specified in error-detail was not the last attribute in ret-
attr-list.
Recovery. Place the iterative attribute (and its auxiliary data, if any) last in the ret-attr-
list array.
16 Attribute not permitted in a search request.
Cause. The attribute specified in error-detail is not permitted in a search request.

Recovery. Do not specify this attribute code in et-attr-list when the srch-option is 1 or
2.
17 Attribute restricted to privileged callers.
Cause. error-detail contains the attribute code.

Recovery. Avoid using this attribute, or use it in a call from a function running in
priviliged mode.
7 -4
8
PROCESS_GETPAIRINFO_ Errors
The PROCESS_GETPAIRINFO_ procedure retrieves basic information about a
process pair. The caller can specify a process handle or process descriptor, or can
search through the named processes in a given system.
The PROCESS_GETINFO_ procedure should be used when the caller requires only
selected information about a particular process.
Error Codes
This subsection lists each PROCESS_GETPAIRINFO_ procedure error code and
0 NO ERROR
Cause. The information returned is for the processes specified.

Effect. The information is returned about the specified process pair.
2 PARAMETER ERROR
Cause. The PROCESS_GETPAIRINFO_ call contained an illegal combination of

options.
Recovery. Correct the call to PROCESS_GETPAIRINFO_ so that the options are
correct and in the proper order.
3 BOUNDS ERROR
Cause. There was a bounds violation on a reference parameter.

Effect. The procedure sets the error code.
Recovery. Pass a correct reference address to PROCESS_GETPAIRINFO_.

8 -1
PROCESS_GETPAIRINFO_ Errors Error Codes
4 SINGLE NAMED PROCESS
Cause. A process pair was requested but only a single named process was found.
Effect. Information is returned for the single named process, which can be the caller.
Recovery. None.
5 CALLER'S PAIR - CALLER IS CURRENT PRIMARY
Cause. The information returned is for the caller's pair, and the caller is the current
primary.
Effect. Information is successfully returned.
6 CALLER'S PAIR - CALLER IS CURRENT BACKUP
Cause. The information returned is for the caller's pair, and the caller is the current
backup.
Effect. Information is successfully returned.
7 PROCESS IS UNNAMED
Cause. The process is not a named process.

Effect. No information is returned because the process is unnamed. The process can
be the caller.
Recovery. No recovery action is required. Call PROCESS_GETINFO_ or
PROCESS_GETINFOLIST_ to obtain the process descriptor.
8 SEARCH IS COMPLETE
Cause. The search has finished.

Effect. No information is returned. The search finished and no more information was
found. Information is returned only if search by index was specified.

8 -2
PROCESS_GETPAIRINFO_ Errors Error Codes
9 PROCESS DOES NOT EXIST
Cause. The specified process does not exist, or the name refers to a device or volume
instead of a process.
Effect. No information is returned.
10 UNABLE TO COMMUNICATE WITH NODE
Cause. The process could not communicate with the node named, possibly because
the node did not exist or was inoperable.
Recovery. Select another node, then try again.
11 TARGET IS A PROCESS CONTROLLING A DEVICE OR VOLUME,

BUT BIT 15 WAS NOT SET
Cause. The target process was an IOP and bit 15 of the options parameter was not
set.
Effect. No information is returned.
Recovery. Change the application to accept information about IOPs by changing the
options parameter, or skip to the next processing step if the you do not want the
application to process information about IOPs.
13 LIMITED INFORMATION IS RETURNED FOR A PROCESS THAT IS

NOT STARTED
Cause. The target named process has not been started.

Effect. Limited information is returned. Returned process handles are null (-1 in each
word).

8 -3
9
PROCESS_SPAWN_ Open System
Services (OSS) Errors
The following error codes and error lists are produced by PROCESS_SPAWN_, the
process-control procedure that creates an Open System Services (OSS) process.
The process-results parameter of PROCESS_SPAWN_ returns the
ZSYS-DDL-PROCESSRESULTS structure, which contains errors detected by the OSS
environment and errors detected by the Guardian environment. All error fields should
be examined to determine the source of the error. Table 9-1 lists the error fields in the
structure.
Table 9-1. PROCESS_SPAWN_ Error Fields

Field Description
Z-ERRNO is an OSS errno value that is returned by the OSS environment.
Z-TPCERROR is a process creation error that is returned by the Guardian environment.
Z-TPCDETAIL is a value that provides further detail on some process creation errors
reported in Z-TPCERROR.
If PROCESS_SPAWN_ is called in a nowait manner, error information is returned in

system message -141 (nowait PROCESS_SPAWN_ completion). See Section 20,
System Messages, for details.
For further information about starting processes and about the PROCESS_SPAWN_
procedure, refer to the Guardian Procedure Calls Reference Manual.
Guardian Error Codes

The PROCESS_SPAWN_ Guardian error codes are the values returned to the calling
process in the ZSYS-DDL-PROCESSRESULTS.Z-TPCERROR field of the
process-results parameter.
For many classes of Guardian errors, additional information is returned in the
parameter.
Guardian process creation error codes returned by PROCESS_SPAWN_ are also
returned by PROCESS_CREATE_. These error codes are described in Section 6,
Process Creation Errors.

9 -1
PROCESS_SPAWN_ Open System Services (OSS) OSS Error Codes
Errors
OSS Error Codes

The PROCESS_SPAWN_ OSS error codes are the values returned to the calling
process in the ZSYS-DDL-PROCESSRESULTS.Z-ERRNO field of the
process-results parameter. Commonly returned OSS error codes are described in
this section. For more information on OSS error codes, see Section 22, OSS Error
Information.
0 NO ERROR
Cause. The PROCESS_SPAWN_ call was completed successfully. The

corresponding OSS errno value is ENOERR.
Effect. The process was created, or creation was initiated if PROCESS_SPAWN_ was
called in a nowait manner. In the latter case, creation results are returned in a user-
level system message.
4002 NO SUCH FILE OR DIRECTORY
Cause. Either the pathname is empty or a component of a specified pathname does

not exist. The corresponding OSS errno value is ENOENT.
Recovery. Create a file with the specified pathname or specify a pathname that exists.
4005 I/O ERROR
Cause. A physical input or output error occurred during an attempt to create or access
a file. The corresponding OSS errno value is EIO.
4007 ARGUMENT LIST TOO LONG
Cause. The number of bytes available for the new process’s combined argument
(argv) and environment (envp) lists has exceeded the system-imposed limit. This
limit, which includes the pointers and the null terminators on the strings, is available by
calling the sysconf(_SC_ARG_MAX) OSS function. The corresponding OSS errno
value is E2BIG.
Recovery. Specify a shorter argument list in the argv parameter and a shorter
environment list in the envp parameter.

9 -2
Errors
4008 EXEC FORMAT ERROR
Cause. The OSS program file specified by the oss-program-file parameter has
the appropriate permissions, but it is not in the format for executable files. The
corresponding OSS errno value is ENOEXEC.
Recovery. Either correct the oss-program-file parameter to refer to an
executable program file or alter the program file such that it can be executed. The
program file must be either a valid object file with a SYSTYPE of OSS or a valid script
file.
4009 BAD FILE DESCRIPTOR
Cause. A file descriptor specified in the fdinfo parameter is invalid. The

corresponding OSS errno value is EBADF.
Recovery. Correct the fdinfo parameter.
4011 NO MORE PROCESSES
Cause. System resources such as the process control block (PCB) space, MAPPOOL
space, stack space, or process file segment (PFS) space are inadequate. It is also
possible that the process name selected is already in use. The corresponding OSS
errno value is EAGAIN.
Recovery. Check the system for processes that are using too much memory,
terminate processes that are no longer needed, and call PROCESS_SPAWN_ again.
Check the program to see whether it uses too much buffer space, opens too many
files, or uses too many DEFINEs. If the problem is with the PFS, try running the
process with a larger PFS either by specifying a larger PFS size in the
ZSYS^DDL^PROCESSEXTENSION.Z^PFSSIZE field of the process-extension
parameter or by setting the size with the Binder. If the process name selected is
already in use, then choose another process name and call PROCESS_SPAWN_
again.
4012 INSUFFICIENT USER MEMORY
Cause. There is insufficient user memory to create the OSS process. The new
process image requires more memory than is allowed by the hardware or system-
imposed memory management constraints. The corresponding OSS errno value is
ENOMEM.

9 -3
Errors
Recovery. Terminate OSS processes that are no longer needed and call
PROCESS_SPAWN_ again.
4013 PERMISSION DENIED
Cause. Search permission is denied on the pathname or a component of a pathname

prefix. The corresponding OSS errno value is EACCES.
Recovery. Correct the parameter containing the erroneous pathname.
4014 BAD ADDRESS
Cause. A specified parameter has an invalid address. The corresponding OSS errno
value is EFAULT.
Recovery. Correct the erroneous parameter.
4020 NOT A DIRECTORY
Cause. A prefix within a specified pathname refers to a file that is not a directory. The
corresponding OSS errno value is ENOTDIR.
4022 INVALID FUNCTION ARGUMENT
Cause. One of the specified parameters is not valid or a required parameter is not
specified. The corresponding OSS errno value is EINVAL.
Recovery. Correct the erroneous parameter.
4126 CONNECTION TIMED OUT
Cause. The operation timed out. The timeout value specified in the fdinfo parameter
was reached before the file descriptors could be opened. The corresponding OSS
errno value is ETIMEDOUT.
Recovery. Call the PROCESS_SPAWN_ procedure again or increase the timeout
value.

9 -4
Errors
4131 FILE NAME TOO LONG
Cause. The specified pathname or a component of the pathname is longer than

PATH_MAX characters. (PATH_MAX is a symbolic constant that is defined in the OSS
limits.h header file.) The corresponding OSS errno value is ENAMETOOLONG.
4203 OSS NOT RUNNING OR NOT INSTALLED
Cause. Open System Services is not running. To create an OSS process, Open
System Services must be running. The corresponding OSS errno value is
EOSSNOTRUNNING.
Recovery. Run Open System Services.
4212 AN ERROR OCCURRED DURING INVOCATION OF A DEFINE

PROCEDURE
Cause. An error exists in a Guardian DEFINE. The corresponding OSS errno value
is EDEFINEERR.
Recovery. Recovery is application dependent. See the Guardian Programmer’s Guide
for an explanation of how to use DEFINEs in procedure calls.

9 -5
PROCESS_SPAWN_ Open System Services (OSS) Error Lists
Errors
Error Lists
subsystem, you might receive a PROCESS_SPAWN_ error list in a response. HP
subsystems return such an error list when, in performing your request, they call the
PROCESS_SPAWN_ procedure directly or indirectly and an error occurs on the call.
PROCESS_SPAWN_ error list, identifies the procedure. Its value is
ZGRD-VAL-PROCESS-SPAWN (24).
constructing a PROCESS_SPAWN_ error list:
description.
Programming Manual.

9 -6
PROCESS_SPAWN_ Open System Services (OSS) 24: ZGRD-VAL-PROCESS-SPAWN
Errors
24: ZGRD-VAL-PROCESS-SPAWN
There was an error during a call to PROCESS_SPAWN_.

ZSPI-TKN-PROCESS-RESULTS token-type ZSPI-TYP-BYTESTRING.
Conditional Tokens
ZGRD-TKN-PROGRAM-PATH token-type ZSPI-TYP-STRING.

ZGRD-TKN-LIBRARY-PATH token-type ZSPI-TYP-STRING.
ZGRD-TKN-SWAP-PATH token-type ZSPI-TYP-STRING.
ZGRD-TKN-EXTSWAP-PATH token-type ZSPI-TYP-STRING.
ZGRD-TKN-TERM-PATH token-type ZSPI-TYP-STRING.
ZGRD-TKN-CWD-PATH token-type ZSPI-TYP-STRING.
ZGRD-TKN-FD-PATH token-type ZSPI-TYP-STRING.
ZGRD-TKN-FDINFO token-type ZSPI-TYP-BYTESTRING.
ZGRD-TKN-ARG token-type ZSPI-TYP-STRING.
ZGRD-TKN-ENV token-type ZSPI-TYP-STRING.
ZGRD-TKN-PROCESS-EXTENSION token-type ZSPI-TYP-BYTESTRING.
ZGRD-TKN-PATH token-type ZSPI-TYP-STRING.
Z-ERROR is the OSS errno returned in the
parameter of PROCESS_SPAWN_.
ZGRD-VAL-PROCESS-SPAWN (24).
ZSPI-TKN-PROCESS-RESULTS is the value of the process-results parameter
(containing the structure ZSYS-DDL-PROCESSRESULTS) of PROCESS_SPAWN_.
Conditional Tokens
ZGRD-TKN-PROGRAM-PATH is the OSS pathname of the program file.
ZGRD-TKN-LIBRARY-PATH is the OSS pathname of the library file.
ZGRD-TKN-SWAP-PATH is the OSS pathname of the swap file.
ZGRD-TKN-EXTSWAP-PATH is the OSS pathname of the extended swap file.

9 -7
PROCESS_SPAWN_ Open System Services (OSS) 24: ZGRD-VAL-PROCESS-SPAWN
Errors
ZGRD-TKN-TERM-PATH is the OSS pathname of the home terminal for the new
process.
ZGRD-TKN-CWD-PATH is the OSS pathname of the current working directory for the
new process.
ZGRD-TKN-FD-PATH is the OSS pathname of a file descriptor to be opened by the
new process. The corresponding pointer to this OSS pathname is in the
ZSYS-DDL-FDINFO.Z-FDENTRY.Z-NAME field of the fdinfo parameter. The
fdinfo parameter is in the ZGRD-TKN-FDINFO token. If the pointer is null (0D), then
a zero-length OSS pathname must be specified. Because there can be multiple file
descriptors to be opened, there can be multiple occurrences of the Z-FDENTRY
substructure and of the corresponding ZGRD-TKN-FD-PATH token. Multiple
ZGRD-TKN-FD-PATH tokens must be provided in the same order as the pointers to
the OSS pathnames in the fdinfo parameter.
ZGRD-TKN-FDINFO is the value of the fdinfo parameter.
ZGRD-TKN-ARG is an argument string pointed to by the argv parameter. The argv
parameter is an array of pointers to strings. Because there can be multiple strings,
there can be multiple occurrences of the ZGRD-TKN-ARG token. Multiple
ZGRD-TKN-ARG tokens must be provided in the same order as elements of the argv
array.
ZGRD-TKN-ENV is the process environment string pointed to by the envp parameter.
The envp parameter is an array of pointers to strings. Because there can be multiple
strings, there can be multiple occurrences of the ZGRD-TKN-ENV token. The
ZGRD-TKN-ENV tokens do not need to be provided in the same order as the elements
of the envp array.
ZGRD-TKN-PROCESS-EXTENSION is the value of the process-extension
parameter.
ZGRD-TKN-PATH is the value of the path parameter.
Effect
The effect of this error depends on the values returned in the
ZSPI-TKN-PROCESS-RESULTS token.
Recovery
Follow the recovery procedure for the returned PROCESS_SPAWN_ errors as

9 -8
10 ALLOCATESEGMENT Errors
This section contains the error codes returned by the ALLOCATESEGMENT procedure
and the error lists associated with the ALLOCATESEGMENT procedure. Error codes
and error lists associated with the SEGMENT_ALLOCATE_ procedure are described in
Section 11, SEGMENT_ALLOCATE_ Errors. The error list for USESEGMENT is
described in Section 12, USESEGMENT Errors. The error list for SEGMENT_USE_ is
described in Section 13, SEGMENT_USE_ Errors.
Error Codes
The ALLOCATESEGMENT error codes are the values returned in the status
parameter to ALLOCATESEGMENT. ALLOCATESEGMENT error codes in the range 1
through 999 indicate a file-system error related to the creation or opening of a swap
file. Refer to Section 2, File-System Errors, for information about these codes.
0 NO ERROR
Cause. The operation was completed successfully.

Effect. None.
-1 ILLEGAL SEGMENT ID
Cause. The ALLOCATESEGMENT call either specified an invalid extended data

segment ID (not in the range 0 through 1023) or used the extended data segment ID of
a currently allocated extended data segment.
Effect. The procedure sets the error code and returns without allocating the extended
data segment.
Recovery. Specify a valid extended data segment ID, then retry the operation. User
processes can specify an extended data segment ID equal to 1023 or less; only
processes supplied by HP can specify an extended data segment ID greater than
1023.
-2 ILLEGAL SEGMENT SIZE
Cause. The ALLOCATESEGMENT call specified an invalid extended data segment

size.
data segment.
Recovery. Specify a valid extended data segment size, then retry the operation. The
range of valid values for a selectable segment is between 1 byte and 127.5

10- 1
ALLOCATESEGMENT Errors Error Codes
megabytes. (To allocate a shared flat segment, you must use the
SEGMENT_ALLOCATE_ procedure.)
-3 BOUNDS VIOLATION ON SWAP FILE PARAMETER
Cause. A bounds violation occurred on the ALLOCATESEGMENT swap-file name.

data segment.
Recovery. Pass a correct reference address or the swap-file name to
ALLOCATESEGMENT.
-4 ILLEGAL COMBINATION OF OPTIONS
Cause. The ALLOCATESEGMENT call contained an illegal combination of options.

data segment.
Recovery. Correct the call to ALLOCATESEGMENT so that the options are correct
and in the proper order.
-5 UNABLE TO ALLOCATE SEGMENT SPACE
Cause. ALLOCATESEGMENT could not allocate extended data segment space.

data segment.
Recovery. Retry the operation, or run the application on another processor.
-6 UNABLE TO ALLOCATE SEGMENT PAGE TABLE SPACE
Cause. ALLOCATESEGMENT could not allocate any segment page table space. The
segment page table stores one entry for each page allocated in the extended data
segment that it is related to.
data segment.

10- 2
ALLOCATESEGMENT Errors Error Codes
-7 SECURITY VIOLATION
Cause. A security violation occurred when the process tried to share extended data
segment space using ALLOCATESEGMENT.
data segment.
Recovery. Make sure the calling process access ID is one of the following:
• The same as that of the process whose PIN is specified in the
ALLOCATESEGMENT call
• The group manager for the access ID of the other process
• The super ID (255,255)
-8 PIN DOES NOT EXIST
Cause. The ALLOCATESEGMENT call specified an invalid process identification

number (PIN).
data segment.
Recovery. Specify a valid PIN, then retry the operation.
-9 NO SEGMENT ALLOCATED TO PIN
Cause. The process specified for segment sharing has not allocated an extended data
segment or the ID does not match the requestor.
data segment.
-10 TRYING TO SHARE SEGMENT WITH SELF
Cause. The process calling ALLOCATESEGMENT is trying to share an extended data

segment with itself.
data segment.
Recovery. Make sure the PIN specified in pin-and-flags is not the PIN of the
calling process.

10- 3
ALLOCATESEGMENT Errors Error Lists
-11 REQUESTED SEGMENT IS CURRENTLY BEING RESIZED
Cause. Either the requested segment cannot be accessed by

SEGMENT_ALLOCATE_ until the current resizing operation is complete or the
requested segment is a shared selectable segment but the allocated segment is a flat
segment.
Effect. The procedure sets the error code and returns without accessing the extended
data segment.
Recovery. Wait until the current resizing operation is complete, then retry the
operation. Make sure that the requested segment and the allocated segment are either
both flat segments or both selectable segments when sharing.
Error Lists
subsystem, you might receive an ALLOCATESEGMENT error list in a response.
Subsystems return such an error list when, in performing your request, they call the
ALLOCATESEGMENT procedure directly or indirectly, and an error occurs on the call.
token ZSPI-TKN-PROC-ERR, which is present in every ALLOCATESEGMENT error
list, identifies the procedure. Its value is ZGRD-VAL-ALLOCATESEGMENT (1).
constructing an ALLOCATESEGMENT error list:
• Optionally include any or none of the conditional tokens listed in each error-list
description.
Programming Manual.

10- 4
ALLOCATESEGMENT Errors 1: ZGRD-VAL-ALLOCATESEGMENT
1: ZGRD-VAL-ALLOCATESEGMENT
A call to ALLOCATESEGMENT returned a nonzero status as the function value.

Conditional Tokens

ZGRD-TKN-SEGMENTID token-type ZSPI-TYP-INT.
ZGRD-TKN-SEGMENTSIZE token-type ZSPI-TYP-INT2.
ZGRD-TKN-FILENAME token-type ZSPI-TYP-FNAME.
ZGRD-TKN-PINANDFLAGS token-type ZSPI-TYP-BYTE-PAIR.
Z-ERROR is the error code returned in the status parameter of
ALLOCATESEGMENT.
ZSPI-TKN-PROC-ERR is the procedure in which the error occurred. Its value is ZGRD-
VAL-ALLOCATESEGMENT (1).
Conditional Tokens
ZGRD-TKN-OBJECTFILE is the name of the object file containing the
ALLOCATESEGMENT procedure call in internal format.
ZGRD-TKN-SEGMENTID is the extended data segment ID requested.
ZGRD-TKN-SEGMENTSIZE is the extended data segment size in bytes.
ZGRD-TKN-FILENAME is the swap file name.
ZGRD-TKN-PINANDFLAGS contains the value of the pin-and-flags parameter:
Value Meaning
PINANDFLAGS.<0:7> Flag options
PINANDFLAGS.<8:15> PIN of process sharing the extended data segment

10- 5
ALLOCATESEGMENT Errors 1: ZGRD-VAL-ALLOCATESEGMENT
Effect
The extended data segment is not allocated.
Recovery
Follow the recovery procedure for the returned ALLOCATESEGMENT error status as

10- 6
11 SEGMENT_ALLOCATE_ Errors
This section lists and describes the error codes and error lists associated with the
SEGMENT_ALLOCATE_ procedure. The SEGMENT_ALLOCATE_ error codes are the
values returned in the error parameter to SEGMENT_ALLOCATE_. The same error
codes are returned by SEGMENT_ALLOCATE_CHKPT_.
Error codes and error lists associated with the ALLOCATESEGMENT procedure are
described in Section 10, ALLOCATESEGMENT Errors. The error list for
USESEGMENT is described in Section 12, USESEGMENT Errors. The error list for
SEGMENT_USE_ is described in Section 13, SEGMENT_USE_ Errors.
Error Codes
This subsection lists each SEGMENT_ALLOCATE_ procedure error code and provides
a description of each code.
0 NO ERROR
Cause. The operation was completed successfully.

Effect. None.
1 ERROR CREATING OR OPENING SEGMENT SWAP FILE
Cause. The system encountered an error while creating or opening the swap file. The
error-detail parameter contains the error number of the file-system error that
occurred.
data segment.
Recovery. Correct the call to SEGMENT_ALLOCATE_ to make sure the correct swap
file is specified. If there is no disk space available to create the swap file, either specify
a different volume or create space on the desired volume.

11 -1
SEGMENT_ALLOCATE_ Errors Error Codes
2 PARAMETER ERROR
Cause. The SEGMENT_ALLOCATE_ call contained an illegal combination of options.

Possible causes include:
• Omitting the required segment-id parameter
• Supplying a max-size value that is inconsistent with segment-size
• Supplying an invalid segment-type
• Setting one or more “reserved, must be 0” bits to a nonzero value
The error-detail parameter contains the ordinal number of the first (leftmost)
parameter encountered whose option is in error.
data segment.
Recovery. Correct the call to SEGMENT_ALLOCATE_ so that the options are correct
and in the proper order.
3 BOUNDS VIOLATION
Cause. A bounds violation occurred on a reference parameter. The reference

parameter address is checked for length and location. The error-detail parameter
contains the ordinal number of the first (leftmost) parameter encountered whose option
is in error.
Effect. The procedure sets the error code and returns without any further action.
Recovery. Pass a correct reference address to SEGMENT_ALLOCATE_.
4 ILLEGAL SEGMENT ID
Cause. The SEGMENT_ALLOCATE_ call either specified an invalid segment ID (not

in the range 0 through 1023) or used the segment ID of a currently allocated segment.
data segment.
Recovery. Specify a valid segment ID, then retry the operation. User processes can
specify a segment ID equal to 1023 or less; only processes supplied by HP can specify
a segment ID greater than 1023.

11 -2
5 ILLEGAL SEGMENT SIZE
Cause. The SEGMENT_ALLOCATE_ call specified an invalid segment size.

data segment.
Recovery. Specify a valid segment size, then retry the operation. Valid values for
segment-size depend on the type of segment to be allocated:
• For a flat segment, segment-size must be in the range of 1 byte to 128
megabytes.
• For a selectable segment, segment-size must be in the range of 1 byte to 127.5
megabytes.
6 UNABLE TO ALLOCATE SEGMENT SPACE
Cause. SEGMENT_ALLOCATE_ could not allocate segment space because not

enough contiguous memory was available.
data segment.
7 UNABLE TO ALLOCATE SEGMENT PAGE TABLE SPACE
Cause. SEGMENT_ALLOCATE_ could not allocate any segment page table space.
The segment page table stores one entry for each page allocated in the segment that it
is related to.
data segment.

11 -3
8 SECURITY VIOLATION
Cause. A security violation occurred when the process tried to share segment space
using SEGMENT_ALLOCATE_. This error indicates that the segment space could not
be shared.
data segment.
Recovery. Make sure the calling process access ID is one of the following:
• The same as the process whose process identification number (PIN) is specified in
the SEGMENT_ALLOCATE_ call.
• The group manager for the access ID of the other process
• The super ID (255,255)
9 PIN DOES NOT EXIST
Cause. The SEGMENT_ALLOCATE_ call specified an invalid PIN.

data segment.
10 NO SEGMENT ALLOCATED TO PIN
Cause. The process specified for extended data segment sharing has not allocated a
segment or IDs do not match.
data segment.
11 TRYING TO SHARE SEGMENT WITH SELF
Cause. The process calling SEGMENT_ALLOCATE_ is trying to share an extended

data segment with itself.
data segment.
Recovery. Make sure the PIN specified in pin is not the calling process’s PIN.

11 -4
12 REQUESTED SHARED SEGMENT IS INCOMPATIBLE
Cause. The requested segment is a shared selectable segment but the allocated
segment is a flat segment; or the requested segment is a shared flat segment but the
allocated segment is a selectable segment.
data segment.
Recovery. Make sure that the requested segment and the allocated segment are
either both flat segments or both selectable segments when sharing.
13 REQUESTED SEGMENT ALREADY ALLOCATED
Cause. The requested segment is already allocated for this process.

Recovery. Specify an unused segment ID, then retry the operation.
14 UNABLE TO ALLOCATE THE PST
Cause. The requested process segment table (PST) cannot be allocated. The
error-detail parameter contains the number of the file-system error that occurred.
data segment.
Recovery. Wait until more process file segment (PFS) space is available, or run the
application on another processor. Refer to Section 2, File-System Errors, for
information about the file-system error returned in error-detail.
15 REQUESTED ADDRESS RANGE ALREADY ALLOCATED
Cause. Part or all of the specified address range has already been allocated. This
error is returned if bit 15 of the alloc-options parameter is set to 1 and a flat segment
cannot be allocated. This error can also occur when bit 15 is not set, but either a flat
segment cannot be shared due to address-range overlap with another segment or a
flat segment cannot be allocated as there is no unallocated address range large
enough to hold the requested size. This error is returned only on TNS/R processors.
Effect. The extended data segment cannot be allocated.
Recovery. Correct the specification and try again.

11 -5
SEGMENT_ALLOCATE_ Errors Error Lists
Error Lists
subsystem, you might receive a SEGMENT_ALLOCATE_ error list in a response. HP
subsystems return such an error list when, in performing your request, they call the
SEGMENT_ALLOCATE_ procedure directly or indirectly and an error occurs on the
call.
SEGMENT_ALLOCATE_ error list, identifies the procedure. Its value is ZGRD-VAL-
SEGMENT-ALLOCATE (21).
constructing a SEGMENT_ALLOCATE_ error list:
description.
information about creating error lists, additional information about token and token
Programming Manual.

11 -6
SEGMENT_ALLOCATE_ Errors 21: ZGRD-VAL-SEGMENT-ALLOCATE
21: ZGRD-VAL-SEGMENT-ALLOCATE
A call to SEGMENT_ALLOCATE_ returned a nonzero error as the function value.

Conditional Tokens

ZGRD-TKN-SEGMENTSIZE token-type ZSPI-TYP-INT2.
ZGRD-TKN-XFILENAME token-type ZSPI-TYP-STRING.
ZGRD-TKN-PIN token-type ZSPI-TYP-UINT.
ZGRD-TKN-SEGMENTTYPE token-type ZSPI-TYP-ENUM.
ZGRD-TKN-MAXSEGSIZE token-type ZSPI-TYP-INT2.
Z-ERROR is the error code returned in the error parameter of
SEGMENT_ALLOCATE_.
VAL-SEGMENT-ALLOCATE (21).
Conditional Tokens
ZGRD-TKN-XOBJECTFILE is the name of the object file that contains the
SEGMENT_ALLOCATE_ procedure call in internal format.
ZGRD-TKN-SEGMENTID is the extended segment ID requested.
ZGRD-TKN-SEGMENTSIZE is the extended segment size in bytes.
ZGRD-TKN-XFILENAME is the swap-file name in internal format.
ZGRD-TKN-ERRORDETAIL is the detailed information associated with some classes of
errors.
ZGRD-TKN-PIN is the process identification number (PIN) of the process sharing the
segment.
ZGRD-TKN-SEGMENTTYPE is the type of segment requested.

11 -7
SEGMENT_ALLOCATE_ Errors 21: ZGRD-VAL-SEGMENT-ALLOCATE
ZGRD-TKN-MAXSEGSIZE is the maximum size that the segment can grow to using
calls to RESIZESEGMENT.
Effect
The extended segment is not allocated.
Recovery
Follow the recovery procedure for the returned SEGMENT_ALLOCATE_ error status
as described earlier in this section.

11 -8
12 USESEGMENT Errors
This section contains the error list associated with the USESEGMENT procedure.
There are no error codes associated with USESEGMENT; USESEGMENT does not
return error codes.
Note that the USESEGMENT procedure is similar to the SEGMENT_USE_ procedure.
You would typically use the SEGMENT_USE_ procedure on D-series and later
releases, while you used the USESEGMENT procedure on C-series releases. See
Section 13, SEGMENT_USE_ Errors, for the error list associated with the
SEGMENT_USE_ procedure.
Also see Section 10, ALLOCATESEGMENT Errors, for error codes and error lists
associated with the ALLOCATESEGMENT procedure. See Section 11,
SEGMENT_ALLOCATE_ Errors, for error codes and error lists associated with the
SEGMENT_ALLOCATE_ procedure.
Error Lists
subsystem, you might receive a USESEGMENT error list in a response. Subsystems
return such an error list when, in performing your request, they call the USESEGMENT
procedure directly or indirectly and an error occurs on the call.
USESEGMENT error list, identifies the procedure. Its value is ZGRD-VAL-
USESEGMENT (2).
The error list always includes the unconditional tokens listed under its description in
this section. In addition, the error list can include any of the conditional tokens listed
under its description.
constructing a USESEGMENT error list:
description.
This section does not discuss the mechanics of error-list construction. For information
about creating error lists, additional information about tokens and token types, and
definitions of tokens whose names begin with ZSPI-, refer to the SPI Programming
Manual.

12- 1
USESEGMENT Errors 2: ZGRD-VAL-USESEGMENT
2: ZGRD-VAL-USESEGMENT
A condition code less (CCL) was returned by a call to USESEGMENT. Either the
segment ID in ZGRD-TKN-SEGMENTID is not allocated or it cannot be used by a
nonprivileged caller.

Conditional Tokens

ZGRD-TKN-OLDSEGMENT token-type ZSPI-TYP-INT.
Z-ERROR is the error code returned in the status parameter of USESEGMENT.
VAL-USESEGMENT (2).
Conditional Tokens
ZGRD-TKN-OBJECTFILE is the name of the object file that contains the
USESEGMENT procedure call in internal format.
ZGRD-TKN-OLDSEGMENT is the extended data segment ID returned.
Effect
The extended data segment requested is not enabled for use by the caller.
Recovery
Make sure the specified extended data segment ID is already allocated and that the
segment ID is 1023 or less.

12- 2
13 SEGMENT_USE_ Errors
This section contains the error list associated with the SEGMENT_USE_ procedure.
This procedure selects a particular extended data segment to be currently addressable
by the calling process.
Note that the SEGMENT_USE_ procedure is similar to the USESEGMENT procedure.
You would typically use the SEGMENT_USE_ procedure on D-series and later
releases, while you used the USESEGMENT procedure on C-series releases. See
Section 12, USESEGMENT Errors for the error list associated with the USESEGMENT
procedure.
Also see Section 10, ALLOCATESEGMENT Errors, for error codes and error lists
associated with the ALLOCATESEGMENT procedure. See Section 11,
SEGMENT_ALLOCATE_ Errors, for error codes and error lists associated with the
SEGMENT_ALLOCATE_ procedure.
Error Lists
subsystem, you might receive a SEGMENT_USE_ error list in a response.
Subsystems return such an error list when, in performing your request, they call the
SEGMENT_USE_ procedure directly or indirectly and an error occurs on the call.
SEGMENT_USE_ error list, identifies the procedure. Its value is ZGRD-VAL-
SEGMENT-USE (23).
The error list always includes the unconditional tokens listed under its description in
this section. In addition, the error list can include any of the conditional tokens listed
under its description.
constructing a SEGMENT_USE_ error list:
description.
This section does not discuss the mechanics of error-list construction. For information
about creating error lists, additional information about tokens and token types, and
definitions of tokens whose names begin with ZSPI-, refer to the SPI Programming
Manual.

13- 1
SEGMENT_USE_ Errors 23: ZGRD-VAL-SEGMENT-USE
23: ZGRD-VAL-SEGMENT-USE
A call to SEGMENT_USE_ returned a nonzero error as the function value. There are
several possible causes of the error, as explained below.

Conditional Tokens

ZGRD-TKN-OLDSEGMENT token-type ZSPI-TYP-INT.
Z-ERROR is the error code returned in the error return value of SEGMENT_USE_.
VAL-SEGMENT-USE (23).
Conditional Tokens
ZGRD-TKN-XOBJECTFILE is the name of the object file that contains the
SEGMENT_USE_ procedure call in internal format.
ZGRD-TKN-OLDSEGMENT is the previous extended data segment ID in use.
ZGRD-TKN-ERRORDETAIL is the detailed information associated with some classes of
errors.
Effect
The extended data segment requested is not enabled for use by the caller.

13- 2
SEGMENT_USE_ Errors 23: ZGRD-VAL-SEGMENT-USE
Recovery
The recovery procedure depends on the specific error code returned in the error
return value.
• If the error code is 2 (parameter error), the SEGMENT_USE_ call contained an
illegal combination of options. Specifically, the required new-segid parameter was
omitted. Correct the SEGMENT_USE_ call by providing the new-segid
parameter.
• If the error code is 3 (bounds violation), a bounds violation occurred on a reference
parameter. To recover from this error, pass a correct reference address to
SEGMENT_USE_.
• If the error code is 4 (illegal segment ID), the SEGMENT_USE_ call specified a
segment ID (in the new-segid parameter) that is illegal for an unprivileged caller.
That is, the segment ID is not in the range 0 through 2047. To recover from this
error, specify a valid segment ID and then retry the operation. User processes can
specify a segment ID equal to 2047 or less. Only processes supplied by HP can
specify a segment ID greater than 2047.
• If the error code is 5 (segment ID not found), the SEGMENT_USE_ call specified a
segment ID (in the new-segid parameter) that has not been allocated. To recover
from this error, specify a valid, allocated segment ID and then retry the operation.

13- 3
14
Subsystem Programmatic Interface
(SPI) Errors
This section lists and describes the error numbers issued by the Subsystem
Programmatic Interface (SPI) procedures and the error lists associated with those
procedures.
Error Codes
The following pages describe the SPI error numbers. These numbers are the values
returned in the status parameter on calls to the SPI procedures.
When any of these errors (except error 0 or error -1) occurs, the header token ZSPI-
TKN-LASTERR is set to the error number.
0 ZSPI-ERR-OK No error
Cause. The operation completed successfully.

Effect. The requested operation is complete.
-1 ZSPI-ERR-INVBUF Invalid buffer format
Cause. The buffer supplied in the procedure call has an invalid format, as indicated by
one or more of the following:
• The first word of the buffer does not contain the correct message code (ZSPI-VAL-
MSGCODE = -28).
• The buffer is in a format not recognized by the current version of SPI.
• The used length of the buffer (ZSPI-TKN-USEDLEN) is greater than the maximum
length (Z-BUFLEN). Probably SSPUT was called with ZSPI-TKN-RESET-BUFFER
and a maxlen value that is smaller than ZSPI-TKN-USEDLEN.
• The buffer contains a ZSPI-TKN-ENDLIST token but no corresponding list token.
• The position descriptor within the buffer (ZSPI-TKN-POSITION) indicates a current
list that does not begin with a list token. Perhaps an incorrect position descriptor—
one saved from another buffer—was restored to ZSPI-TKN-POSITION in this
buffer.
Effect. The requested operation is not completed. Since the buffer format is invalid,
the last error is not saved.
Recovery. Check for an incorrect buffer parameter or a corrupted buffer.

14- 1
Subsystem Programmatic Interface (SPI) Errors Error Codes
-2 ZSPI-ERR-ILLPARM Illegal parameter value
Cause. A parameter supplied in the procedure call was illegal for one of the following
reasons:
• SSINIT was called with an invalid header type.
• A negative index or count parameter was supplied.
• An attempt was made to use SSPUT or SSGET on a token using a count of zero.
• A call was made to one of the special operations that returns attributes of a token,
but the operation was applied to one of the special tokens. The special tokens that
return attributes are ZSPI-TKN-COUNT, ZSPI-TKN-LEN, ZSPI-TKN-OFFSET, and
ZSPI-TKN-ADDR.
• The program supplied an SPI-defined token code that was invalid for this
procedure call. For example, ZSPI-TKN-DELETE was specified in a call to SSGET,
or ZSPI-TKN-COMMAND was specified in a call to SSPUT.
• An invalid position descriptor was supplied with ZSPI-TKN-POSITION.
Effect. The header token ZSPI-TKN-LASTERR is set to this error number, and the
requested operation is not completed.
Recovery. Correct the parameter in error.
-3 ZSPI-ERR-MISPARM Missing parameter
Cause. This error number indicates that a required parameter was not supplied.
Certain parameters are required only under certain circumstances:
• The ssid parameter is required when calling SSGET with ZSPI-TKN-NEXTCODE
or ZSPI-TKN-NEXTTOKEN if the next token code in the buffer is not qualified by
the default subsystem ID. Always supply a variable for ssid when calling SSGET
with ZSPI-TKN-NEXTCODE or ZSPI-TKN-NEXTTOKEN unless you are certain
that all tokens the program could encounter are qualified by the default subsystem
ID.
• The value parameter is required when calling SSGET with certain standard token
codes (such as ZSPI-TKN-LEN and ZSPI-TKN-OFFSET) or when calling SSPUT
with a token code that has a value (a nonzero token length).
Effect. The header token ZSPI-TKN-LASTERR is set to this error number (unless the
buffer parameter is missing), and the requested operation is not performed.
Recovery. Supply the missing parameter.

14- 2
-4 ZSPI-ERR-BADADDR Illegal parameter address
Cause. A reference parameter has an illegal address for one of the following reasons:
• A parameter has a starting address that is invalid or out of bounds.
• A parameter has an absolute extended address and the caller is nonprivileged.
• A parameter's starting address and length are such that the parameter overlaps the
stack space that is required by the SPI procedure.
• An extended-address parameter refers to the current code space. This can occur if
the parameter is a read-only array located in a user-library space.
Effect. The header token ZSPI-TKN-LASTERR is set to this error number (unless the
bounds error occurs on the buffer parameter), and the requested operation is not
performed.
If the bounds error occurs on the buffer parameter, SPI is unable to find the buffer,
so it cannot set ZSPI-TKN-LASTERR.
Recovery. Correct the parameter declarations to allocate the required storage.
-5 ZSPI-ERR-NOSPACE Buffer full
Cause. This error can occur for one of the following reasons:
• The buffer is full; it cannot contain any more tokens or header information.
• SSPUT was called with ZSPI-TKN-RESET-BUFFER, but maxlen was smaller than
the used length of the buffer. In this case, some information at the end of the
message was lost. Subsequent SPI calls for this buffer will return error -1 (invalid
buffer format).
Recovery. For the first cause, use a larger buffer. For the second cause, recovery is
application-dependent.
-6 ZSPI-ERR-XSUMERR Invalid checksum
Cause. The current buffer checksum does not match the checksum computed on
return from the last SPI call. This error suggests that the buffer has been modified or
damaged.
Recovery. Using a debugging tool such as the Inspect debugger, check for corruption
of the buffer contents.

14- 3
-7 ZSPI-ERR-INTERR Internal error
Cause. This internal error should not occur unless the SPI software malfunctions.
Specific causes include:
• SSGET attempted to return a token value when the program had requested a
token attribute (such as length, offset, address, or count).
• SSGET attempted to return an undefined token attribute.
• On returning, SSGET or SSPUT attempted to set a used length greater than the
buffer length.
• SSPUT received an error when calling SSGET to obtain the default subsystem ID
(ZSPI-TKN-DEFAULT-SSID) from the SPI message header.
Recovery. Report the problem to your service provider, supplying a reproducible test
case.
-8 ZSPI-ERR-MISTKN Token not found
Cause. This error can occur for one of the following reasons:
• The token requested in a call to SSGET was not found.
• An attempt was made to put a ZSPI-TKN-ENDLIST token into the buffer, but no
corresponding list token was found.
Recovery. Corrective action depends on the application. Check for program logic
errors in scanning the buffer. Also check to see whether the token is positioned outside
the current list or preceding the current position.

14- 4
-9 ZSPI-ERR-ILLTKN Illegal token code or map
Cause. An illegal token code or token map was supplied in the procedure call.
Possible causes for this error include:
• The token data type was not recognized. For example, a program used a token
data type not included in the standard SPI definitions. The only token data types
permitted by SPI are those defined by SPI.
• The token length was not a multiple of the basic length associated with the token
data type.
• The token map contained an invalid null-value specification.
• The sum of the lengths of the null-value specifications in the token map was not
equal to the total structure length specified by the map.
• A token map was supplied as a parameter to SSGETTKN, SSPUTTKN, or
SSMOVETKN.
Any of these situations might arise if the program accidentally corrupted the variable
holding the token code or token map.
Recovery. Correct the token code or token map causing the error.
-10 ZSPI-ERR-BADSSID Invalid subsystem ID
Cause. A subsystem ID with an invalid name was supplied as a parameter. The

owner-name field of a subsystem ID must contain an owner name that:
• Begins with a letter
• Contains only letters, hyphens, or digits
• Is left-justified and padded with blanks
Recovery. Check the ssid parameter being supplied in the call, and correct it as
necessary.

14- 5
-11 ZSPI-ERR-NOTIMP Operation not supported
Cause. This operation is not supported in the version of the SPI definitions being
used. (For instance, the operation ZSPI-TKN-ADDR was called to get the address of a
header token.)
Recovery. Check that the token code being supplied in the call is defined for the
version of the SPI definitions being used.
-12 ZSPI-ERR-NOSTACK Insufficient stack space
Cause. An SPI procedure was called with fewer words remaining in the data stack
than were needed by that procedure.
Recovery. Increase the number of stack pages available or reduce the amount of
stack space used. The methods available for taking this action depend on the
programming language.
-13 ZSPI-ERR-ZFIL-ERR File-system error
Cause. A file-system error occurred.

Recovery. Corrective action depends on the error. Check Table 14-1 for an
explanation of the status-1 and status-2 parameters.

14- 6
Table 14-1. Supplementary Status Values

Error status-1 status-2 Explanation
-13 2 x File-system error x received during an open of the
nonresident template file.
-13 3 x File-system error x received during a read of the
nonresident template file.
-14 1 x ALLOCATESEGMENT error x received during
allocation of the private segment.
-14 7 x MOVEX error x received during access of the
private segment.
-15 -2 x Nonresident template file error x, in which x has the
following values:
-1 File code is not 844.
-2 File is not a disk file.
-3 File is not key-sequenced.
-4 File has wrong record size.
-5 File has wrong primary key definition.
-14 ZSPI-ERR-ZGRD-ERR Guardian procedure error
Cause. An error occurred during allocation of the private segment; this error is
returned from ALLOCATESEGMENT or MOVEX.
-15 ZSPI-ERR-INF-FILE Invalid template file
Cause. The template file is invalid.

-16 ZSPI-ERR-CONTINUE More text is available
Cause. The entire representation does not fit in the number of lines specified by the
printlines parameter.
Effect. The amount of text is limited by the printlines parameter setting.
Recovery. Call again to get the next portion.

14- 7
-26 ZSPI-ERR-NO-SCANID See service provider.
Cause. No free scan ID was available. Currently, no more than two scan IDs can be
active simultaneously.
Recovery. No recovery is possible. Report this error to your service provider.
-27 ZSPI-ERR-NO-FORMATID No format ID available
Cause. Currently, no more than one format ID can be active at one time.
Recovery. Recode your program to format only one buffer at a time.

14- 8
Subsystem Programmatic Interface (SPI) Errors Error Lists
Error Lists
If a HP subsystem calls an SPI procedure and an SPI error occurs—that is, the call
returns a nonzero value in status—the subsystem normally attempts to recover from
the error. If it can't recover (for example, when the application program buffer is
corrupted), the subsystem returns information to the application in an error list. The
following pages describe the error list associated with each SPI procedure.
token ZSPI-TKN-PROC-ERR, which is present in every SPI error list, identifies the
procedure.
constructing an error list for an SPI error:
description.
information about creating error lists, for additional information about tokens and token
types, and for definitions of tokens whose names begin with ZSPI-, refer to the SPI
Programming Manual.

14- 9
Subsystem Programmatic Interface (SPI) Errors Error From SSGET or SSGETTKN
Error From SSGET or SSGETTKN

A call to SSGET or SSGETTKN returned an error.

ZSPI-TKN-PARM-ERR token-type ZSPI-TYP-PARM-ERR.
Conditional Tokens
ZSPI-TKN-SSID-ERR token-type ZSPI-TYP-SSID.
ZSPI-TKN-ERROR is the standard SPI token, whose value consists of the fields
Z-SSID and Z-ERROR. Z-SSID is the SPI subsystem identifier ZSPI-VAL-SSID.
Z-ERROR is the status value from the SPI procedure SSGET or SSGETTKN.
ZSPI-VAL-SSGET (2) or ZSPI-VAL-SSGETTKN (3).
ZSPI-TKN-PARM-ERR contains fields ZTOKENCODE, ZINDEX, and ZOFFSET. The
value of ZTOKENCODE is the token code or the first 32 bits of the token map passed
in the call that failed. ZINDEX is the index parameter passed in the call that failed.
ZOFFSET is 0.
Conditional Tokens
ZSPI-TKN-SSID-ERR is the ssid parameter passed in the call that failed. This token
appears only if ssid was omitted in the call that failed.
Effect
The call to SSGET or SSGETTKN fails.
Recovery
Follow the recovery procedure for the returned SPI error code as described earlier in
this section.

14 -10
Subsystem Programmatic Interface (SPI) Errors Error From SSINIT
Error From SSINIT

A call to SSINIT returned an error.

ZSPI-TKN-ERROR is the standard SPI token, whose value consists of the fields
Z-ERROR is the status value from the SPI procedure SSINIT.
ZSPI-TKN-PROC-ERR is the procedure in which the error occurred. Its value is ZSPI-
VAL-SSINIT (1).
Effect
The call to SSINIT fails.
Recovery
this section.

14- 11
Subsystem Programmatic Interface (SPI) Errors Error From SSMOVE or SSMOVETKN
Error From SSMOVE or SSMOVETKN

A call to SSMOVE or SSMOVETKN returned an error.

Conditional Tokens
Z-ERROR is the status value from the SPI procedure SSMOVE or SSMOVETKN.
ZSPI-VAL-SSMOVE (4) or ZSPI-VAL-SSMOVETKN (5).
value of ZTOKENCODE is the token code or the first 32 bits of the token map passed
in the call that failed. ZINDEX is the source-index parameter passed in the call that
failed; ZINDEX is 0 if source-index was omitted in the call that failed. ZOFFSET
is 0.
Conditional Tokens
appears only if ssid was omitted in the call that failed.
Effect
The call to SSMOVE or SSMOVETKN fails.
Recovery
this section.

14 -12
Subsystem Programmatic Interface (SPI) Errors Error From SSNULL
Error From SSNULL

A call to SSNULL returned an error.

Z-ERROR is the status value from the SPI procedure SSNULL.
VAL-SSNULL (6).
value of ZTOKENCODE is the first 32 bits of the token map passed in the call that
failed. ZINDEX and ZOFFSET are 0.
Effect
The call to SSNULL fails.
Recovery
this section.

14 -13
Subsystem Programmatic Interface (SPI) Errors Error From SSPUT or SSPUTTKN
Error From SSPUT or SSPUTTKN

A call to SSPUT or SSPUTTKN returned an error.

Conditional Tokens
Z-ERROR is the status value from the SPI procedure SSPUT or SSPUTTKN.
VAL-SSPUT (7) or ZSPI-VAL-SSPUTTKN (8).
value of ZTOKENCODE is the token code or first 32 bits of the token map passed in
the call that failed. ZINDEX and ZOFFSET are 0.
Conditional Tokens
only appears if ssid was omitted in the call that failed.
Effect
The call to SSPUT or SSPUTTKN fails.
Recovery
this section.

14 -14
15
EDITREAD and EDITREADINIT
Errors
This section lists and describes the errors returned in the status parameter of
EDITREAD and EDITREADINIT procedure calls.
Error Codes
This subsection lists each EDITREAD and EDITREADINIT procedure error code and
0 or greater NO ERROR
Cause. The call was completed successfully. For EDITREAD, a positive value
represents the number of characters that the procedure read in the text line.
Effect. None.
Recovery. No corrective action is required.
-1 END-OF-FILE ENCOUNTERED
Cause. The procedure encountered an end-of-file in the EDIT file.

• For EDITREADINIT, this means the file was empty.
• For EDITREAD, this condition might indicate that the reading of the file is finished.
requested function.
Recovery. Corrective action is application-dependent.
-2 I/O ERROR
Cause. A file-system error occurred.

requested function.
Recovery. Call FILEINFO to obtain the file-system error, and refer to Section 2, File-
System Errors, for corrective action.

15- 1
EDITREAD and EDITREADINIT Errors Error Codes
-3 TEXT FILE FORMAT ERROR
Cause. EDITREADINIT returns this error if any of the following is true:

• The file was not an EDIT file.
• The buffer length was not a power of two in the range 64 through 2048.
• The directory claims the file contains a line with a sequence number greater than
99999.999.
• The file is internally inconsistent.
For EDITREAD, an error exists in the internal format of the file.
requested function.
Recovery. For recovery suggestions for files formatted as EDIT files, see the EDIT
User’s Guide and Reference Manual.
-4 SEQUENCE NUMBER ERROR
Cause. EDITREAD returns this error if the sequence number of the line about to be
read is one of the following:
• Less than that of its predecessor.
• Greater than 99999.999.
• Greater than the largest sequence number in the directory entry for the file page
containing the line about to be read.
requested function.
Recovery. For recovery suggestions for files formatted as EDIT files, see the EDIT
User’s Guide and Reference Manual.

15- 2
EDITREAD and EDITREADINIT Errors Error Codes
-5 CHECKSUM ERROR
Cause. The edit control block was invalid. There are two possible causes:
• The program called EDITREAD before calling EDITREADINIT.
• The program modified the edit control block, which invalidated the control block.
Note. If you accidentally set the edit control block's reposition bit to 1, you will not receive this
error. When the reposition bit is 1, checksum validation is disabled during the call to
EDITREAD.
Effect. The operation terminates.

Recovery. Correct the program error that caused the problem.

15- 3
16 IOEdit Errors
IOEdit allows the run-time libraries of the compilers supported on the operating system
to read and write EDIT format files. An application process can use the following IOEdit
procedures to access files:
BACKSPACEEDIT INITIALIZEEDIT
CLOSEALLEDIT NUMBEREDIT
CLOSEEDIT OPENEDIT
CLOSEEDIT_ OPENEDIT_
COMPLETEIOEDIT PACKEDIT
COMPRESSEDIT POSITIONEDIT
DELETEEDIT READEDIT
EXTENDEDIT READEDITP
GETINCREMENTEDIT UNPACKEDIT
GETPOSITIONEDIT WRITEEDIT
INCREMENTEDIT WRITEEDITP
Like the sequential I/O (SIO) procedures, IOEdit is capable of reading and writing EDIT
files. EDIT files have a file code of 101.
Most error messages returned by IOEdit are the same as the file-system error codes
described in Section 2, File-System Errors. Some have special meaning to IOEdit. The
messages with special meanings are further explained here.
A status or code of 0 indicates that no error occurred.
Error Message Forms

IOEdit returns errors in three different forms:
• Errors returned as positive integers, using the same codes as the file-system
errors described in Section 2, File-System Errors. These errors can assume a
somewhat different meaning when returned by IOEdit, as described below.
• Errors returned as negative integers. These errors are unique to IOEdit and can
occur only from operations on EDIT files.
• Errors that result in the immediate abnormal termination of the calling process. No
recovery is possible. IOEdit does not return to the caller but instead writes a
message to the home terminal and calls ABEND.
The classes of IOEdit errors are discussed in the following subsections.

16- 1
IOEdit Errors File-System Error Codes
File-System Error Codes

Errors reported using positive integers are generally the same as those returned by the
file system. The positive IOEdit error codes that do not have the same meaning as the
file-system errors described in Section 2, File-System Errors, are presented here.
2 (%2) OPERATION NOT ALLOWED ON THIS TYPE OF FILE
Cause. IOEdit returns this error when a file is specified that is not an EDIT file
(unstructured, file code 101) or is not on a disk device.
Effect. IOEdit cannot operate on this file.
10 (%12) FILE/RECORD ALREADY EXISTS
Cause. This error is returned when the same file has already been processed by a
previous call. IOEdit does not support multiple concurrent opens of the same file.
IOEdit returns this error when trying to write a record having the same EDIT line
number as a record already in the file. IOEdit has no procedure corresponding to
WRITEUPDATE; you can rewrite a record only by first deleting it (by calling
DELETEEDIT).
Effect. No action is taken.
11 (%13) FILE NAME NOT IN DIRECTORY OR RECORD NOT

IN FILE, OR THE SPECIFIED TAPE FILE IS NOT
PRESENT ON A LABELED TAPE
Cause. IOEdit returns this error when the file does not exist and read only usage is
specified. Unlike FILE_OPEN_, OPENEDIT_ creates the file if read-write or write only
use is specified and does not return error 11 in that case.
Effect. The requested file was not found.

16- 2
16 (%20) FILE NUMBER HAS NOT BEEN OPENED
Recovery. Most IOEdit procedures return this error when called with the file number of
a file that has not been processed by OPENEDIT_.
Effect. The file was not open, so it could not be processed.
Recovery. Use OPENEDIT or OPENEDIT_ to open the file.
31 (%37) UNABLE TO OBTAIN FILE SYSTEM BUFFER SPACE

33 (%41) I/O PROCESS IS UNABLE TO OBTAIN BUFFER
SPACE
34 (%42) UNABLE TO OBTAIN FILE SYSTEM CONTROL BLOCK
Cause. IOEdit returns these errors when failures occur in the management of space
within the EDIT file segment (EFS). The EFS is analogous to the process file segment
(PFS) used by the file system.
These errors occur only when the caller has more EDIT files open than the limit
specified to INITIALIZEEDIT (the default limit is 30), or when IOEdit needs to enlarge
the EFS but its backing disk is too full to allocate another extent.
Effect. Space is not available to continue processing.
Recovery. Close some EDIT files and try again.
45 (%55) FILE IS FULL
Cause. The current file size is too small. WRITEEDIT returns this error when the
capacity of the file (set by its extent size and maximum number of extents established
when the file was created) is about to be exceeded. Unlike the file system, IOEdit
allows you to recover from this error by calling EXTENDEDIT to increase the file's
capacity to a maximum of 128 megabytes, and then repeating the call to WRITEEDIT
that failed.
Effect. IOEdit cannot continue.
Recovery. Call EXTENDEDIT to increase the file's capacity, then try the call to
WRITEEDIT again.
46 (%56) INVALID KEY SPECIFIED
Cause. Several IOEdit procedures return this error when the EDIT line number
specified in the call is less than -3 or greater than 99999999.
Effect. The request is ignored.
Recovery. Specify a valid EDIT line number.

16- 3
59 (%73) FILE IS BAD
Recovery. Most IOEdit procedures return this error when an automatic attempt to
recover from a warning condition (previously reported by one of the negative error
codes described later in this section) does not succeed. All subsequent calls (except to
CLOSEEDIT) for the same file return error 59.
Effect. The file cannot be used.

16- 4
IOEdit Errors Error Codes Unique to IOEdit
Error Codes Unique to IOEdit

IOEdit can detect some error conditions for which no existing file-system error code
has been defined. These conditions can occur only in operations on EDIT files. IOEdit
uses negative integers to report these conditions.
Error numbers -1 through -5 are returned when IOEdit finds something wrong with the
directory in an EDIT file. This directory corresponds to the set of index blocks in a key-
sequenced file and can be rebuilt by reading the entire file. Rather than doing this,
OPENEDIT_ sets a flag and returns one of the error codes -6 through -10.
The caller can then choose to abandon the file by calling CLOSEEDIT or try to
reconstruct the file's directory by calling any of the other IOEdit procedures. In the
latter case, IOEdit tries to rebuild the directory in memory, without changing the file on
the disk. If this succeeds, all subsequent IOEdit operations work as if the file were
intact. If this does not succeed, this and all subsequent calls to IOEdit procedures
(except CLOSEEDIT) for the same file return error 59.
-1 (inconsistent directory size value)
Cause. A directory error occurred in an EDIT file.

Effect. Processing stops.
Recovery. Proceed as described at the beginning of this subsection.
-2 (record numbers in the directory are out of order)

-3 (a record number in the directory is beyond legal range)

-4 (a block number in the directory is outside the file)


16- 5
-5 (duplicate block numbers in the directory)

-6 (numbers ran out)
Cause. COMPRESSEDIT and EXTENDEDIT return this error when the new record
number of the last record in the file would be larger than 99999999. NUMBEREDIT
returns this error when the new record number of the last record to be renumbered
would not be less than the record number of the next record.
Effect. In either case, the problem is that the increment parameter's value is too large,
and the procedure does nothing.
Recovery. Change the increment parameter to an acceptable value.
-7 (record numbers out of order)
Cause. DELETEEDIT and NUMBEREDIT return this error when the record number of
the first record to be deleted or renumbered is greater than that of the last.
Several other IOEdit procedures return this error when reading a file and finding a
record whose record number is not greater than that of the record preceding it.
Effect. The procedure stops.
Recovery. The file is bad but possibly most of the file's contents could be recovered
by copying the entire file and ignoring only the records having this error.
-8 (record length too large)
Cause. Several IOEdit procedures return this error when reading a file and finding a
record whose length is such that the record would extend beyond the end of its block.
-9 (record format inconsistent)
Cause. Several IOEdit procedures return this error when reading a file and finding a
record whose compression codes disagree with the length of the record.

16- 6
-10 (cannot do renumbering)
Cause. NUMBEREDIT returns this error when the parameters represent disallowed
renumbering: the first new record number is not greater than that of the record
preceding the first record to be renumbered, the new record number increment is zero
or negative, or no records are affected.
-13 (error accessing non-EDIT file line numbers)

-14 (error accessing non-EDIT file change tags)
-15 (change tag should be appended to record)
-16 (both tag and record number error)
-17 (change tag overlays text)
Cause. These errors can occur only when IOEdit is accessing a file other than an
EDIT file. For example, they can occur when a local version control system database is
accessed.
Recovery. This error should not occur; recovery is not possible. Determine the type of
access desired and correct the request.

16- 7
IOEdit Errors Catastrophic Error Messages
Catastrophic Error Messages

Errors for which no recovery is possible result in the immediate abnormal termination
of the calling process. For these conditions, IOEdit does not return to the caller but
instead writes a message to the home terminal and calls ABEND. The catastrophic
error messages are the following.
**** IOEdit internal error, process aborted ****
Cause. A data structure maintained by IOEdit has an inconsistent condition. The

problem is in IOEdit itself, not in the calling program.
Effect. The calling process is terminated abnormally.
**** IOEdit unable to allocate Edit File Segment, process

aborted ****
Cause. If the caller does not explicitly call the INITIALIZEEDIT procedure, the first
IOEdit procedure called does so automatically. In any case, INITIALIZEEDIT first tries
to create the Edit File Segment on the swapping disk volume for the calling process,
and if that fails, it then tries every disk volume in the system. If all attempts fail, IOEdit
issues this message and calls ABEND. If the system is that short of disk space, many
other processes will probably fail in a similar manner.
**** IOEdit stack overflow, process aborted ****
Cause. In the course of accessing caller-provided information that is in user extended

data segments, IOEdit makes local copies on the stack. This can make the program's
stack space requirements much larger than was apparent when the program was
compiled and bound. When the stack is about to overflow due to such an action, IOEdit
issues this message rather than allowing a mysterious trap to occur. The user should
rearrange global primary, global secondary, and the stack space use of the calling
procedures leading to the call to IOEdit to increase the space available for the stack
within IOEdit.

16- 8
IOEdit Errors Catastrophic Error Messages
**** IOEdit needs more pages in user data segment, process

aborted ****
Cause. A program using IOEdit might require more data pages than was apparent
when the program was compiled and bound. When IOEdit makes a local copy on the
stack and these additional pages would cause an address trap because the process
has fewer than 32 data pages (as indicated by the value returned by LASTADDR),
IOEdit issues this message rather than allowing a mysterious trap to occur. The user
should rerun the program with MEM 32 (or more) specified in the RUN command.

16- 9
17 Formatter Errors
The following errors are returned by the FORMATDATA and FORMATDATAX
procedures. The error code is returned as the resulting value from the procedure call.
For additional information about the FORMATDATA[X] procedures, refer to the
Error Codes
This subsection lists each FORMATDATA[X] procedure error code and provides a
description of each code.
267 BUFFER OVERFLOW
Cause. To interpret an edit descriptor, FORMATDATA[X] had to access a character

before the start of the buffer or a character outside the buffer.
Recovery. Correct the format.
268 NO BUFFER
Cause. FORMATDATA[X] required a new buffer, but there were no more buffers.
Recovery. Correct the format, or increase the number of buffers.
270 FORMAT LOOPBACK
Cause. Data items remained to be processed after FORMATDATA[X] reached the end
of a format that contains no repeatable edit descriptors.
Recovery. Include repeatable edit descriptors in the format, or reduce the number of
data items.

17- 1
Formatter Errors Error Codes
271 EDIT ITEM MISMATCH
Cause. In a format-directed operation, an edit descriptor was matched to a data

element that has an incompatible type. For example, the "G" edit descriptor was
associated with a string data element on output, or any edit descriptor except "A" was
associated with a string data element on input.
In list-directed input, a numeric data element was repeated using the r*c form, and
some data element after the first element to which this form applied was a string-type
element.
Recovery. Correct the format, or correct the data list to include missing (or delete
extra) items.
272 ILLEGAL INPUT CHARACTER
Cause. The numeric input field contained an inappropriate character for the
corresponding edit descriptor. For example, a nonnumeric character was entered in a
field being interpreted according to the "I" edit descriptor, or lowercase letters were
used where uppercase letters were required.
Recovery. Correct the format or the data list.
273 BAD FORMAT
Cause. The format contains an edit descriptor that is valid for output but not for input.
For example, I5.5 is invalid for input.
Recovery. Correct the format.
274 NUMERIC OVERFLOW
Cause. A data element's numeric value was so small or so large that

FORMATDATA[X] could not place it in its corresponding data element.
Recovery. Change the format or correct the numeric calculations.

17- 2
18 INITIALIZER Errors
The following error messages are returned by the INITIALIZER procedure. No error
numbers are returned from this procedure call.
For additional information about the INITIALIZER procedure, refer to the Guardian
Programmer’s Guide and the Guardian Procedure Calls Reference Manual.
The following messages indicate fatal coding or internal errors and are issued only
when the program calls the ABEND procedure. The error message is passed along in
the ABEND procedure call to the process’s creator. If the creator process is the TACL
process, the message is passed to the home terminal. If the creator process is not the
TACL process, error handling is application dependent.
Error Messages
This subsection lists each INITIALIZER procedure error message and provides a
description of each message.
Error message: “INITIALIZER: Unable to allocate buffer for

message”
Cause. An error occurred during the allocation of space in a buffer pool. If the
procedure was called from a FORTRAN or COBOL program, the SAVE directive was
not specified.
Effect. The INITIALIZER prepares the error message and calls ABEND.
Recovery. If the procedure was called from a TAL program, this is an internal error. If
bits <0:10> of the flags parameter are 0, report this problem to your service provider.
If the procedure was called from a FORTRAN or COBOL program, specify the MEM 64
option in your RUN command or reduce the number of ASSIGN and PARAM attributes
for your program.

18- 1
INITIALIZER Errors Error Messages
Error message: “INITIALIZER: Invalid FCB format or wrong

number of FCBs specified”
Cause. The cause of this error can be one of the following:

• The file control block (FCB) size is not correct.
• An invalid number of FCBs were specified in either the ALLOCATE^CBS or
ALLOCATE^CBS^D00 DEFINE.
Effect. The INITIALIZER prepares the error message and calls ABEND.
Recovery. Correct the FCB size. The correct sizes for pTAL or TAL programs are as
follows:
DEFINE Used FCB Size for TNS/R FCB Size for TNS or
Native Process Accelerated Process
ALLOCATE^CBS 82 words 60 words
ALLOCATE^CBS^D00 102 words 80 words
If the FCB size was correct, check the number of FCBs in the ALLOCATE^CBS or
ALLOCATE^CBS^D00 DEFINE.
Error message: “INITIALIZER: The NUM^FCBS parameter is

incorrect or not specified”
Cause. The cause of this error can be one of the following:

• The INITIALIZER procedure was called by a TNS/R native process that did not
specify the required num^fcbs parameter.
• A negative number of file control blocks (FCBs) was specified in the num^fcbs
parameter.
• A positive number of file control blocks (FCBs) was specified in the num^fcbs
parameter but the fcbârray parameter was not specified.
Effect. INITIALIZER prepares the error message and calls ABEND.
Recovery. Either correct the value in the num^fcbs parameter or specify the
num^fcbs parameter.
Error message: “INITIALIZER: Timeout reading $RECEIVE”
Cause. A timeout occurred during a wait on $RECEIVE to read the startup sequence.
Recovery. The default waiting period is 60 seconds. The D10 and later versions of
GPLIB offer an optional parameter to specify this timeout period.

18- 2
Error message: “INITIALIZER: Unexpected message from

creator process”
Cause. The message received from the creator process was not a valid message in
the startup sequence.
Recovery. Correct the program that is sending the startup sequence. Refer to the
Guardian Programmer’s Guide for more information.
Error message: “INITIALIZER: Message saving requested but

RUCB missing or undersized”
Cause. This is an internal error.

Recovery. If bits <0:10> of the flags parameter are 0, report this problem to your
service provider.
Error message: “INITIALIZER: Unable to obtain process handle”

Recovery. Report this problem to your service provider.
Error message: “INITIALIZER: Backup takeover and

flags.<12> reset”
Cause. The backup of a process pair returned from CHECKMONITOR and

flags.<12> was reset to 0.
Recovery. Either correct the problem that caused the primary process to terminate,
make sure the primary process does stack checkpointing, or set bit 12 of the flags
parameter to 1.
Error message: “INITIALIZER: Unable to open $RECEIVE”
Cause. $RECEIVE is probably already open.

Effect. INITIALIZER prepares the error message calls ABEND.
Recovery. Make sure $RECEIVE is not open when you call INITIALIZER.

18- 3
Error message: “INITIALIZER: Unexpected error”

Recovery. Report this problem to your service provider.

18- 4
19
Interprocess Command Interpreter
Messages
Application processes use Tandem Advanced Command Language (TACL) to send or
receive the messages described in this section. The TACL product is the command
interpreter supplied by HP for use on the operating system.
Though user-written command interpreters can be used, in this section the TACL
product is assumed to be the only command interpreter. If a user-written command
interpreter is in use, read "command interpreter" instead of "TACL" in the following
messages.
The interprocess command interpreter messages described in this section should not
be confused with the system messages described in Section 20, System Messages.
Though these messages share many of the same message numbers, the command
interpreter messages do not generate a file-system error 6 (system message received)
as do the system messages.
Command interpreter messages are not error messages; they are used to convey
information. When a request for a new process occurs, such as from a
PROCESS_CREATE_ procedure call, the operating system creates a new process.
When the process is created, a message is sent to it by the TACL process. This
message is called the startup message. The startup message carries an indication of
any ASSIGN or PARAM messages to be added. (ASSIGNs and PARAMs establish
certain characteristics of the new process.) When there are no more ASSIGNs or
PARAMs to be processed, the newly created process is ready to be accessed.
Figure 19-1 illustrates the communication between an application process, the TACL
process, and $CMON.
Figure 19-1. Command Interpreter Messages

application <----> TACL <----> $CMON
Messages -1, -2, and -3 are sent by the TACL process to processes that it creates.
Messages -20 and -21 are sent by any process to the TACL process.
Messages -50 through -60 are sent to $CMON, but not all of these messages are sent
by the TACL process. ADDUSER, DELUSER, PASSWORD, and RPASSWORD are
privileged, licensed programs; they send their messages to $CMON. When $CMON is
running, these programs add an additional level of access restriction to the operating
system. When $CMON is not running, their restrictions are not in effect.

19- 1
Interprocess Command Interpreter Messages
The first word of each $CMON message is usually a control word that permits or
disallows an operation. For most $CMON messages, the rest of the reply is text that
the TACL process displays. It typically contains the reason why $CMON did not permit
the operation. $CMON messages are also used to display the greeting at LOGON or a
farewell message at LOGOFF. Two exceptions are the command interpreter messages
-52 (run) and -60 (configuration); these allow $CMON to return information or to display
text. If the first word is zero, the rest of the reply contains data; if the first word is
nonzero, the rest of the reply contains text to be displayed.
The sources of the interprocess command interpreter messages are shown in
Table 19-1.
Table 19-1. Command Interpreter Messages

CI Message Source
-1 Startup Sent by TACL process to the process it created
-2 Assign Sent by TACL process to the process it created
-3 Param Sent by TACL process to the process it created
-20 Wakeup Sent by any process to TACL process
-21 Display Sent by any process to TACL process
-50 Logon Sent to $CMON by TACL process
-51 Logoff Sent to $CMON by TACL process
-52 Run Sent to $CMON by TACL process
-53 Illegal Logon Sent to $CMON by TACL process
-54 Add User Sent to $CMON in adduser^msg
-55 Delete User Sent to $CMON in deluser^msg
-56 Alter Priority Sent to $CMON by TACL process
-57 Password Sent to $CMON in password^msg
-58 Remote Password Sent to $CMON in remotepassword^msg
-59 Prelogon Sent to $CMON by TACL process
-60 Configuration Sent to $CMON by TACL process
The message lengths in this section are subject to change. Do not test the received
length against an expected length.
For more information about command interpreter messages, refer to the Guardian

19- 2
Interprocess Command Interpreter Messages Message Descriptions
This subsection describes the interprocess command interpreter messages.
-1 STARTUP
Cause. This message is received by a new process; it indicates that the new process
was successfully created.
Format. The Startup message always ends with a null byte to terminate the parameter
string. If the resulting message has an odd number of bytes, the TACL process appends
a second null byte. The maximum length possible for a startup message is 596 bytes
(including the trailing null characters).
The form of the startup message is:
STRUCT ci^startup;
BEGIN ! word
INT msgcode; ! [0] -1
STRUCT default;
BEGIN
INT volume [0:3]; ! [1] $default-volume-name
INT subvol [0:3]; ! default-subvol-name
END;
STRUCT infile;
BEGIN
INT volume [0:3]; ! [9] IN parameter file name
INT subvol [0:3]; ! of the RUN command
INT dname [0:3];
END;
STRUCT outfile;
BEGIN
INT volume [0:3]; ! [21] OUT parameter file name
INT subvol [0:3]; ! of the RUN command
INT dname [0:3];
END;
STRING param [0:n-1]; ! [33] parameter string of the RUN

! command (if any) that was
! entered by the operator.
! This is in either of the
! following forms:
!
! parameter-string null [null]
! or
! null null
! n = ( count-read - 66 )
END;

19- 3
Response. If you want your program to receive any existing ASSIGN and PARAM
messages, specify file-system error code 70 in a call to REPLY or specify error code 0
but with a reply of 1 to 4 bytes, where bit 0 of the first byte is set to 1 for ASSIGN
messages and bit 1 is set to 1 for PARAM messages.
-2 ASSIGN
Cause. A new process receives one ASSIGN message for each assignment in effect
at the time the new process was created.
The ASSIGN messages immediately follow the startup message if the new process
does one of the following:
• Replies to the startup message with an error return value of 70. The TACL process
then sends both ASSIGN and PARAM messages.
• Replies to the startup message with an error return value of 0 but with a reply of
from 1 through 4 bytes, where bit 0 of the first byte of the reply is set to 1. If bit 1 of
the first byte of the reply is set to 1, the TACL process also sends a PARAM
message.
Format. The format of the ASSIGN message follows. The message length is 108 bytes.
STRUCT ciâssign; ! ASSIGN message
BEGIN !
INT msg^code; ! [0] -2
!
STRUCT logicalunit; ! PARAMETERS TO ASSIGN COMMAND
BEGIN !
STRING prognamelen; ! [1] length in bytes of name
STRING progname[0:30]; ! {0:31} {program-unit | *}
! (blank filled on right)
STRING filenamelen; ! [17] length in bytes of name
STRING filename[0:30]; ! {0:31} logical-file-name
! (blank filled on right)
END; !
INT(32) fieldmask; ! [33] bit mask to indicate
! which of the following
! fields were supplied
! (1 = supplied):
!
! .<0> = physical-filename
! .<1> = pri-extent-size
! .<2> = sec-extent-size
! .<3> = file-code
! .<4> = exclusion-spec
! .<5> = access-spec
! .<6> = record-size
! .<7> = block-size

19- 4
STRUCT physicalfilename; ! [35] physical-filename

BEGIN !
INT volume [0:3]; !
INT subvol [0:3]; !
INT dfile [0:3]; !
END; !
!
INT primaryextent; ! [47] pri-extent-size
INT secondaryextent; ! [48] sec-extent-size
INT filecode; ! [49] file-code
INT exclusionspec; ! [50] %00 if SHARED }
! %20 if EXCLUSIVE } flag
! %60 if PROTECTED } param
INT accessspec; ! [51] %0000 if I/O } of
! %2000 if INPUT } OPEN
! %4000 if OUTPUT }
INT recordsize; ! [52] record-size
INT blocksize; ! [53] block-size
END;
Response. The application need not respond to the ASSIGN message. When all of the
ASSIGN and PARAM messages have been received, the process is ready for use.

19- 5
-3 PARAM
Cause. A new process receives a PARAM message if any parameters are in effect
when the new process is created.
The PARAM messages immediately follow the ASSIGN messages if the process does
one of the following:
• Replies to the startup message with an error return value of 70. The TACL process
then sends both ASSIGN and PARAM messages.
• Replies to the startup message with an error return value of 0, but with a reply of
from 1 through 4 bytes, where bit 1 of the first byte of the reply is set to 1. If bit 0 of
the first byte of the reply is set to 1, the TACL process also sends ASSIGN
messages.
Format. The format of the PARAM message follows. The maximum message length is
1028 bytes.
STRUCT ci^param; ! PARAM message
BEGIN !
INT msg^code; ! [0] -3
INT numparams; ! [1] number of
! parameters
! included in
! this message
STRING parameters [0:1023]; ! [2] beginning of
! parameters
END;
The field parameters in the above message format is composed of numparams
records of the following form (offsets are given in bytes):
param[0] = length n, in bytes, of
parameter-name
param[1] FOR n = parameter-name
param[n+1] = length v, in bytes, of
parameter-value
param[n+2] FOR v = parameter-value
Response. The application need not respond to the ASSIGN message. When all of the
ASSIGN and PARAM messages have been received, the process is ready for use.

19- 6
-20 WAKEUP
Cause. A process sent a wakeup message to the TACL process.

If the TACL process is paused, the wakeup message causes it to return to the
command input mode (that is, "wake up"). If the TACL process is not paused, it ignores
the wakeup message.
Format. The format of the wakeup message follows. The message length is 2 bytes.
STRUCT wakeup^msg;
BEGIN
INT msgcode; ! -20
END;
Response. None required.
-21 DISPLAY
Cause. A process sends a display message to the TACL process.

The display message causes the TACL process to display the text contained in the
message. The text is displayed just prior to the next time the TACL process prompts for
a command.
The TACL process can store 8 undisplayed, 132-byte messages. If 8 messages are
stored, subsequent messages are rejected with an error 12 indication (file in use).
If the TACL process receives a user message followed by a zero-length message, the
TACL process clears its $RECEIVE buffer and checks the message length.
Format. The format of the display message follows. The message length is 2 bytes plus
the display-text length (in bytes).
The length of the text portion is implied in the write count used to send this message.
STRUCT display^msg;
BEGIN
INT msgcode; ! -21
STRING text [0:n-1]; ! n <= 132
END;
Response. None required.

19- 7
-50 LOGON
Cause. This message is sent to the $CMON process every time the TACL process
tries to log on. If the $CMON process is not running, no $CMON logon restrictions are
in effect.
When a LOGON command is entered, the user name is checked for validity.
Format. The format of the logon message follows. The message length is 54 bytes.
STRUCT logon^msg;
BEGIN
INT userid; ! [1] user ID of user
! logging on
INT cipri; ! [2] initial execution
! priority of TACL
INT ciinfile [0:11]; ! [3] name of the TACL
! command file
INT cioutfile [0:11]; ! [15] name of the TACL
! list file
END;
Response. The $CMON reply indicates whether the user is allowed to log on and
contains an optional display message in the following form:
STRUCT logon^reply;
BEGIN
INT replycode; ! [0] 0 = allow logon
! 1 = disallow logon
STRING !
replytext [0:n]; ! [1] optional message to
! be printed; maximum
END; ! of 132 bytes
The length of the message is 2 bytes plus the reply-text length in bytes. The length of
the reply text is implied in the reply count used when making a reply. If reply-count
= 2, no text is displayed.

19- 8
-51 LOGOFF
Cause. This message is sent to the $CMON process when a LOGOFF command is
entered. It is also sent when a user logs on without first logging off (implicit logoff).
Format. The form of the logoff message follows. The message length is 54 bytes.
STRUCT logoff^msg;
BEGIN
INT userid; ! [1] user ID of user logging
! off
INT cipri; ! [2] initial execution
! priority of TACL
! command file
! list file
END;
Response. The $CMON reply contains an optional display message. If the $CMON
process is not running, the TACL process does not try to write the logoff message. The
format of the reply to the logoff message is:
STRUCT logoff^reply;
BEGIN
INT replycode; ! [0] ignored by TACL
STRING !
replytext [0:131]; ! [1] optional message to be
! printed; maximum of 132
! bytes
END;

19- 9
-52 RUN
Cause. This message is sent to the $CMON process whenever the user tries to start a
process either explicitly (RUN prog-file), implicitly (prog-file), or with the TACL
#NEWPROCESS built-in function.
The RUN parameters IN file, OUT file, LIB file, SWAP file, and the parameter
string are included in the process-creation message sent to $CMON.
Format. The format of the RUN message follows. The length of the message is
determined by the value of paramlen.
STRUCT processcreation^msg;
BEGIN
INT userid; ! [1] user ID of user logged on
INT cipri; ! [2] initial priority of TACL
INT ciinfile [0:11]; ! [3] name of TACL command file
INT cioutfile [0:11]; ! [15] name of the TACL list file
INT progname [0:11]; ! [27] expanded program file
! name
INT priority; ! [39] the value of the PRI RUN
! parameter if supplied;
! otherwise, -1
INT processor; ! [40] the value of the processor RUN
! otherwise, -1
INT proginfile [0:11]; ! [41] the expanded IN file RUN
! otherwise, the default
! IN file
INT progoutfile [0:11]; ! [53] the expanded OUT file RUN
! otherwise, the
! default OUT file
INT proglibfile [0:11]; ! [65] the expanded LIB file
! RUN parameter if
! supplied; otherwise,
! blanks
INT progswapfile [0:11]; ! [77] the expanded SWAP file
! RUN parameter if supplied;
! otherwise, blanks
INT paramlen; ! [89] the length of param. This is
! defined for D42 and later
! releases of the operating
system
STRING param [0:527] ! [90] parameter string of the RUN
! command, which is up to 528
! bytes in length including 2
null
! bytes at the end of the
string.
! This is defined for D42 and

19 -10
! later releases of the

! operating system.
END;
The $CMON process can reply in one of two ways:
• With a run-the-process reply
• With a disallow-process-creation reply
A run-the-process reply contains the process priority and the name of the processor
where the process will run. This reply has the format:
STRUCT processcreation^reply;
BEGIN
INT replycode; ! [0] 0 = create the process
INT progname [0:11]; ! [1] expanded name of program
file
! to be run
INT priority; ! [13] execution priority of the
! new process.
! 0 = PRI option specified by
! user.
! If no PRI option is
! specified, then TACL
! priority minus 1.
! >0 = execution priority
! <0 = PRI option specified by
! user plus negative
! priority offset
returned
! in this field.
! If no PRI option is
! specified by user, then
! TACL priority minus 1
! plus negative priority
! priority offset
returned
! in this field.
! For example, PRI = 150,
! priority = -5, priority
! used = 145).
INT processor; ! [14] processor where new
! process is to run or -1.
! If -1, then the processor
! in which the TACL process
! is running is used.
END;
The values returned in the reply are those used for the process-creation attempt. Any
process-creation errors are seen by the TACL user (no notification is made to
$CMON).

19- 11
A disallow-process-creation reply has the format:

STRUCT processcreation^reply;
BEGIN
INT replycode; ! [0] 1 = disallow process
! creation
STRING !
replytext [0:131]; ! [1] optional message to be
! printed
END;
Response. The length of the message is 2 bytes plus the replytext length in bytes.
The length of the replytext is implied in the reply count used when making a reply.
If reply-count = 2, no text is displayed.
-53 ILLEGAL LOGON
Cause. A user tried to log on three times in a row and failed each time.
Format. The format of the illegal logon message follows. No byte count is returned.
STRUCT illegal^logon^msg;
BEGIN
INT userid; ! [1] user ID of user trying
! to log on
INT ciinfile [0:11]; ! [3] name of the TACL command
! file
INT cioutfile [0:11]; ! [15] name of the TACL list
! file
STRING
logonstring [0:n]; ! [27] the attempted logon
! command string;
! maximum length of 132
! bytes
END;
Response. The $CMON reply message contains an optional display message. The
format for the reply is:
STRUCT illegal^logon^reply;
BEGIN
INT replycode; ! [0] ignored by TACL
STRING replytext [0:n]; ! [1] optional message to be
! printed; maximum length
! of 132 bytes
END;
The TACL process delays for one minute.

19 -12
-54 ADD USER
Cause. A user tried to add another user to the system. Any user can be added if the
current user's security setting allows it. However, if $CMON is not running, no $CMON
add-user restrictions are in effect.
Format. The format of the add-user message is:
STRUCT adduser^msg;
BEGIN
INT userid; ! [1] user ID of user adding the
! new user
! command file
! list file
INT groupname [0:3]; ! [27] the group name of the user
! being added
INT username [0:3]; ! [31] the user name of the user
! being added
INT groupîd; ! [35] the group number of the
! user being added
INT userîd; ! [36] the user number of the
! user being added
END;
Response. The $CMON reply indicates whether the user can be added and contains an
optional display message. The format of the reply message is:
STRUCT adduser^reply;
BEGIN
INT replycode; ! [0] 0 = allow addition of
! user
! 1 = disallow addition
! of user
! of 132 bytes
END;

19 -13
-55 DELETE USER
Cause. A user tried to delete another user from the system. Any user can be deleted if
the current user's security setting allows it. If the $CMON process is not running, then
no $CMON delete-user restrictions are in effect.
Format. The format of the delete-user message is:
STRUCT deluser^msg;
BEGIN
INT userid; ! [1] user ID of user deleting
! user
! command file
INT groupname [0:3]; ! [27] the group name of the user
! being deleted
INT username [0:3]; ! [31] the user name of the user
! being deleted
END;
Response. The $CMON reply indicates whether the user should be deleted and contains
an optional reply message. The format of the reply message is:
STRUCT deluser^reply;
BEGIN
INT replycode; ! [0] 0 = allow deletion of user
! 1 = disallow deletion of
! user
! printed; maximum length of
! 132 bytes
END;

19 -14
-56 ALTER PRIORITY
Cause. A user tried to alter the priority of a process. A user can change the priority of
any process that has the same access ID as that user. Only someone logged on as a
super-group user can change the priority of any process. However, if the $CMON
process is not running, then no $CMON alter-priority restrictions are in effect.
Format. The format of the alter-priority message is:
STRUCT altpri^msg;
BEGIN
INT userid; ! [1] user ID of user altering
! the priority
! command file
INT crtpid [0:3]; ! [27] process ID of the process
! whose priority is to be
! altered
INT progname [0:11]; ! [31] expanded program file name
! of the process whose
! priority is to be altered
INT priority; ! [43] the new priority
INT phandle [0:9]; ! [44] process handle of the process
! whose priority is to be
! altered
END;
Response. The $CMON reply indicates whether the process priority should be changed
and contains an optional display message. The format of the reply message is:
STRUCT altpri^reply;
BEGIN
INT replycode; ! [0] 0 = allow priority to be
! altered
! 1 = disallow priority to
! be altered
! of 132 bytes
END;

19 -15
-57 PASSWORD
Cause. The user tried to change his or her password. Users can change their
passwords at any time. However, if the $CMON process is not running, then no
$CMON password restrictions are in effect.
Format. The format of the password message is:
STRUCT password^msg;
BEGIN
INT userid; ! [1] user ID of the user
! changing the password
! command file
END;
Response. The $CMON reply indicates whether the user password can be changed and
contains an optional display message. The format of the reply message is:
STRUCT password^reply;
BEGIN
INT replycode; ! [0] 0 = allow password to be
! changed
! 1 = disallow password to
! be changed
! printed; maximum length of
! 132 bytes
END;

19 -16
-58 REMOTE PASSWORD
Cause. The user tried to change his or her remote password. Users can change their
remote passwords at any time. However, if the $CMON process is not running, then no
$CMON remote-password restrictions are in effect.
Format. The format of the remote password message is:
STRUCT remotepassword^msg;
BEGIN
INT userid; ! [1] user ID of user changing
! remote password
! command file
INT sysname [0:3]; ! [27] change the remote password
! for this system
! (“*” indicates all systems)
INT rpword [0:3]; ! new value for remotepassword
END;
Response. The $CMON reply indicates whether the user's remote password can be
changed and contains an optional display message. The format of the reply message is:
STRUCT remotepassword^reply;
BEGIN
INT replycode; ! [0] 0 = allow the remote
! password to be
! changed
! 1 = disallow the remote
! password to be
! changed
! of 132 bytes
END;

19 -17
-59 PRELOGON
Cause. The TACL process tried to log on. If the $CMON process is not running, then
no $CMON restrictions are in effect. This message is sent before the TACL process
calls VERIFYUSER.
Format. The form of the prelogon message follows. The length of the message is 72
bytes.
STRUCT prelogon^msg;
BEGIN
INT userid; ! [1] user ID of user logging on:
! 0 if the user is logged off
! or is logged on as
! NULL.NULL (0,0)
INT cipri; ! [2] current priority of TACL
INT ciinfile [0:11]; ! [3] TACL IN file
INT cioutfile [0:11]; ! [15] TACL OUT file
INT loggedon; ! [27] 0 if TACL is currently
! logged off, non-zero if
! TACL is already logged on
INT username [0:7]; ! [28] internal username through
! which the user wants to
! log on
END;
Response. The $CMON reply indicates whether the user can log on and contains an
optional display message. The form of the reply message is:
STRUCT prelogon^reply;
BEGIN
INT replycode; ! [0] 0 = proceed to VERIFYUSER
! 1 = disallow logon
! of 132 bytes
END;

19 -18
-60 CONFIGURATION
Cause. This message is sent to the $CMON process just before either an interactive
TACL process attempts to log on from the logged-off state or a noninteractive TACL
process starts. You can set config^request^type to 1 to obtain configuration data
after log on. If requestcmonuserconfig is set to 1 in the reply message, TACL
should request $CMON for user configuration data after log on. For more information
on writing a command interpreter, refer to the Guardian Programmer’s Guide.
If the $CMON process is not running or is running too slowly, the TACL configuration
remains unchanged from its previous values.
Format. The format of the configuration message is:
STRUCT config^msg;
BEGIN
INT userid; ! [1] current user ID of TACL;
! 0 if logged off or logged on
! as NULL,NULL (0,0)
INT cipri; ! [2] current priority of TACL
INT ciinfile [0:11]; ! [3] IN file of TACL
INT cioutfile [0:11]; ![15] OUT file of TACL
INT config^request^type; ![27] configuration request type
! 0 send default configuration
! 1 send user configuration
END;
The length of the message is 56 bytes.
Response. The $CMON reply contains configuration information or a display message.
The format of the configuration information message is as follows:
STRUCT config^reply;
BEGIN
INT replycode; ! [0] 0
INT count; ! [1] number of INTs that
! follow (currently12);
INT autologoffdelay; ! [2] see #GETCONFIGURATION
INT logoffscreenclear; ! [3] description in the
INT remotesuperid; ! [4] TACL Reference
INT blindlogon; ! [5] Manual for more
INT namelogon; ! [6] information on these
INT cmontimeout; ! [7] parameters.
INT cmonrequired; ! [8]
INT remotecmontimeout; ! [9]
INT remotecmonrequired; ! [10]
INT nochangeuser; ! [11]
INT stoponfemodemerr; ! [12]
INT requestcmonuserconfig; ! [13]
END;

19 -19
The length of the message is 4 bytes plus 2 times the value in the variable count. The
message length is currently 28 bytes.
The format of the display message is as follows:
STRUCT config^text^reply;
BEGIN
INT replycode; ! [0] <> 0
!
STRING !
replytext [0:n]; ! [1] optional message to be
! displayed; maximum of
! 132 bytes
END;
The length of the message is 2 bytes plus the replytext length in bytes. The length
of replytext is implied in the reply count used when making a reply. If count = 2, no
text is displayed.

19 -20
20 System Messages
A system message is an interprocess message that is sent from the operating system
to an application process. System messages are received by an application process
through its $RECEIVE file.
There are two procedures you can use to open $RECEIVE: FILE_OPEN_ and OPEN.
The OPEN procedure, which is superseded by the FILE_OPEN_ procedure, is
supported for compatibility with previous software and should not be used for new
development.
When $RECEIVE is opened with FILE_OPEN_, you can choose to receive
D-series-format system messages (the default action) or C-series-format system
messages. When $RECEIVE is opened with OPEN, you receive C-series-format
messages. Some D-series messages supersede one or more C-series messages,
while other D-series messages support newer features. If a process requests C-series
messages and uses a feature for which there is no C-series message, then the
process receives a D-series message. Table 20-1 lists D-series-format system
messages in numerical order; Table 20-1 lists C-series-format system messages in
numerical order and, for each message, provides the equivalent D-series-format
system message. D-series-format system messages that do not have an equivalent C-
series-format system message are listed at the end of this table.
Application Conversion
D-series-format system messages are available for converted applications to read from
$RECEIVE when using the D-series and later system procedures. See the Guardian
Application Conversion Guide and the Guardian Programmer’s Guide for more detailed
information.
Error Return Conventions

The completion of a read associated with a C-series-format system message returns a
condition code of “greater than” (CCG) and file-system error 6 from FILE_GETINFO_.
D-series-format system messages do not use condition codes to indicate an error.
Instead, each procedure returns an integer error value. If an error condition contains
more information than the procedure can return in an integer parameter, the procedure
returns additional information in an integer error-detail parameter.
For more information about system messages, refer to the Guardian Programmer’s
Guide.
Note. When you read system messages by calling the READUPDATE procedure, you must
reply in a corresponding call to REPLY, even if you have no message to return. If your
application process is performing message queuing, call FILE_GETRECEIVEINFO,
LASTRECEIVE, or RECEIVEINFO immediately following completion of the READUPDATE and
pass the message tag back to the REPLY procedure.

20- 1
System Messages Error Return Conventions
The message lengths discussed in this section are subject to change. Do not test the
received length for equality to an expected length.
The following table lists the D-series-format system messages in numerical order.
Table 20-1. D-Series-Format System Messages (page 1 of 2)
-2 Processor Down
-3 Processor Up
-10 SETTIME
-11 Power On
-12 NEWPROCESSNOWAIT Completion
-13 System Message Buffer Overrun
-21 3270 Device Status Received
-22 Elapsed Time Timeout
-23 Memory Lock Completion
-24 Memory Lock Failure
-26 Process Time Timeout
-32 Process CONTROL
-33 Process SETMODE
-34 Process RESETSYNC
-35 Process CONTROLBUF
-37 Process SETPARAM
-38 Queued Message Cancellation
-41 Nowait DEVICEINFO2 Completion
-100 Remote Processor Down
-101 Process Deletion: ABEND, STOP, or Processor Down
-102 Nowait PROCESS_LAUNCH_ or PROCESS_CREATE_ Completion
-103 Process Open
-104 Process Close
-105 Break on Device
-106 Device Type Inquiry
-107 Subordinate Name Inquiry
-108 Nowait FILE_GETINFOBYNAME_ Completion
-109 Nowait FILENAME_FINDNEXT_ Completion
-110 Loss of Communication With Network Node
-111 Establishment of Communication With Network Node
-112 Job Process Creation
-113 Remote Processor Up

20- 2
System Messages Error Return Conventions
Table 20-1. D-Series-Format System Messages (page 2 of 2)

-121 Pathsend Dialog Abort
-141 Nowait PROCESS_SPAWN_ Completion
-147 Device Information Inquiry
The following table lists the C-series-format system messages in numerical order and,
for each message, provides the equivalent D-series-format system message.
D-series-format system messages that do not have an equivalent C-series-format
system message are listed at the end of the table.
Table 20-2. C-Series and D-Series System Messages Compared (page 1 of 2)

C-Series System Message D-Series System Message
-2 Processor Down -2 Processor Down
-2 Processor Down: Named -101 Process Deletion: Processor Down
Process Deletion
-3 Processor Up -3 Processor Up
-5 Process Deletion: STOP -101 Process Deletion: STOP
-6 Process Deletion: ABEND -101 Process Deletion: ABEND
-8 Change in Status of Network -100 Remote Processor Down
Node -110 Loss of Communication With Network Node
Establishment of Communication With
-111 Network Node
Remote Processor Up
-113
-9 Job Process Creation -112 Job Process Creation
-10 SETTIME -10 SETTIME
-11 Power On -11 Power On
-12 NEWPROCESSNOWAIT -12 NEWPROCESSNOWAIT Completion
Completion
-13 System Message Buffer -13 System Message Buffer Overrun
Overrun
-20 Break on Device -105 Break on Device
-21 3270 Device Status Received -21 3270 Device Status Received
-22 Elapsed Time Timeout -22 Elapsed Time Timeout
-23 Memory Lock Completion -23 Memory Lock Completion
-24 Memory Lock Failure -24 Memory Lock Failure
-26 Process Time Timeout -26 Process Time Timeout
-30 Process Open -103 Process Open
-31 Process Close -104 Process Close
-32 Process CONTROL -32 Process CONTROL

20- 3
System Messages Message Descriptions
Table 20-2. C-Series and D-Series System Messages Compared (page 2 of 2)

C-Series System Message D-Series System Message
-33 Process SETMODE -33 Process SETMODE
-34 Process RESETSYNC -34 Process RESETSYNC
-35 Process CONTROLBUF -35 Process CONTROLBUF
-37 Process SETPARAM -37 Process SETPARAM
-38 Queued Message Cancellation -38 Queued Message Cancellation
-40 Device Type Inquiry -106 Device Type Inquiry
-41 Nowait DEVICEINFO2 -41 Nowait DEVICEINFO2 Completion
Completion
none -107 Subordinate Name Inquiry
none -108 Nowait FILE_GETINFOBYNAME_
Completion
none -109 Nowait FILENAME_FINDNEXT_
Completion
none -141 Nowait PROCESS_SPAWN_ Completion
none -147 Device Information Inquiry
This subsection lists the system messages and provides a description of each
message.
-2 PROCESSOR DOWN
Cause. The operating system did not receive an “I’m alive” message from the
specified processor which was being monitored with the MONITORCPUS procedure.
Compare the named process deletion form of message -2, which is listed separately.
Format. The layout of the received message described in terms of an array of words is:
sysmsg[0] = -2
sysmsg[1] = Processor number
Response. The response, if any, is application dependent.
-2 PROCESSOR DOWN: NAMED PROCESS DELETION
Cause. (C-series-format system message only) The operating system did not receive
an “I’m alive” message from the specified processor which was being monitored with
the MONITORCPUS procedure. This form of the processor down message is sent to
the ancestor of a named process (pair) to indicate that the name has been deleted,
that is, the only process running under that name was in the processor that failed.

20- 4
Format. Although this message has the same number as the preceding message (-2), it
is distinguished by the presence of a dollar sign ($) in the second word. The layout of
the received message described in terms of an array of words is:
sysmsg[0] = -2
sysmsg[1] = $process-name
sysmsg[1] = -1
-3 PROCESSOR UP
Cause. A processor being monitored with the MONITORCPUS procedure was

reloaded.
sysmsg[0] = -3
-5 PROCESS DELETION (STOP)
Cause. (C-series-format system message only) A call to the process-control STOP

procedure deleted a process. The stop message is sent to the ancestor of the process
and the ancestor of the job (GMOM).
Format. The two forms of the stop message are:
• If the deleted process was not named, or if one member of a process pair deletes
the other member, the operating system sends the following form of the message:
sysmsg[0] = -5
sysmsg[1] FOR 4 = Process ID of deleted process
sysmsg[5] = Header size (header ends at sysmsg[19])
sysmsg[6] FOR 4 = Process processor time in microseconds
(a FIXED value)
sysmsg[10] = The job ID; 0 if process has no GMOM
sysmsg[11] = The completion code
sysmsg[12] = Termination information; 0 if the user
did not supply information
sysmsg[13] FOR 4 = The subsystem organization name;
for HP products,this is HP
sysmsg[17] = The subsystem number
sysmsg[18] = The subsystem version
sysmsg[19] = The length of text in bytes
sysmsg[20] FOR n = Up to 80 bytes of text

20- 5
• If the call to the STOP procedure deletes the process name from the process-pair
directory, the operating system sends this message:
sysmsg[0] = -5
sysmsg[1] FOR 3 = The name of the deleted process
sysmsg[4] = -1
sysmsg[5] = The header size (header ends at
sysmsg[19])
(a FIXED value)
sysmsg[12] = Termination information, 0 if the user
sysmsg[13] FOR 4 = Subsystem organization (8 bytes);
for HP products, this is HP
sysmsg[17] = Subsystem number
sysmsg[18] = Subsystem version
sysmsg[19] = Length of text in bytes
sysmsg[20] FOR n = Text (up to 80 bytes)
If an external process caused the termination, the stop message is changed as
follows:
sysmsg[11] = Completion code defaults to 6
sysmsg[12] = Creator access ID
sysmsg[13] FOR 4 = Process ID of the process that caused
the termination
This message indicates that neither member of the process pair exists.
For information about completion codes, refer to Table 20-3 on page 20-21.

20- 6
-6 PROCESS DELETION (ABEND)
Cause. (C-series-format system message only) A process was deleted because of a

call to the process-control ABEND procedure, or because the deleted process
encountered a trap condition and was aborted by the operating system.
The abend message is sent to the ancestor of the process and the ancestor of the job
(GMOM).
Format. The two forms of the abend message are:
• If one member of the process pair is deleted or if the deleted process was not
named, the operating system sends the following message:
sysmsg[0] = -6
sysmsg[1] FOR 4 = Process ID of deleted process
sysmsg[5] = Header size (header ends at
sysmsg[19])
(a FIXED value)
sysmsg[13] FOR 4 = The subsystem organization name;
for HP products, this is HP
sysmsg[17] = The subsystem number
sysmsg[18] = The subsystem version
sysmsg[19] = The length of text in bytes
sysmsg[20] FOR n = Up to 80 bytes of text
• If the operating system deletes the process name from the process-pair directory, it
sends the following message. This message indicates that neither member of the
process pair exists.
sysmsg[0] = -6
sysmsg[1] FOR 3 = The name of the deleted process
sysmsg[4] = -1
sysmsg[5] = The header size (header ends at
sysmsg[19])
(a FIXED value)
sysmsg[13] FOR 4 = Subsystem organization (8 bytes);.
for HP products,this is HP
sysmsg[17] = Subsystem number
sysmsg[18] = Subsystem version
sysmsg[19] = Length of text in bytes
sysmsg[20] FOR n = Text (up to 80 bytes)

20- 7
• If an external process caused the termination, the abend message is changed as

follows:
sysmsg[11] = Completion code defaults to 6
sysmsg[12] = Creator access ID
sysmsg[13] FOR 4 = Process ID of the process that caused
the termination
• If a trap is encountered, the abend message is changed as follows:
sysmsg[11] = Completion code is -1
sysmsg[12] = Termination information is 0
sysmsg[13] FOR 4 = Subsystem organization is 4 words of
blanks
sysmsg[17] = Subsystem number is 0
sysmsg[18] = Subsystem version is 0
sysmsg[19] = Length is 76 bytes of text
sysmsg[20] FOR 76 = Is the following text:
TRAP NO=nn, S=nnnnnn, CS=nn, P=nnnnnn,

ENV=nnnnnn, L=nnnnnn, OCT P=nnnnnnnn
For information about the completion codes, refer to Table 20-3 on page 20-21.
-8 CHANGE IN STATUS OF NETWORK NODE
Cause. (C-series-format system message only) The process that was running on a
system that is part of a network enabled receipt of remote status-change messages by
passing “1” as a parameter to the MONITORNET procedure.
sysmsg[0] = -8
sysmsg[1].<0:7> = System number
sysmsg[1].<8:15> = Number of processors
sysmsg[2] = Current processor-status bit mask
sysmsg[3] = Previous processor-status bit mask

20- 8
-9 JOB PROCESS CREATION
Cause. (C-series-format system message only) The receiving process is the

supervisor of a job and a process running under the supervisor’s GMOM job ID created
a third process.
sysmsg[0] = -9
sysmsg[1] FOR 4 = Job ID
sysmsg[5] = Process ID of the newly created process
-10 SETTIME
Cause. The system manager or operator reset the processor’s internal clock.
If a call to MONITORNEW enabled receipt of SETTIME messages, the operating
system sends the process the following message.
sysmsg[0] = -10
sysmsg[2] FOR 4 = Signed change in microseconds
(FIXED integer)
sysmsg[6] = Reason code
The reason codes are:
Code Meaning
0 Initial setting (Greenwich mean time (GMT) and local civil time (LCT) change).
1 Subsequent adjustment (GMT and LCT time change).
2 Daylight-savings time (LCT time change) transition.
-11 POWER ON
Cause. The processor power failed, then returned.

If a call to MONITORNEW enabled receipt of power on messages, the operating
system sends the process the following message.
sysmsg[0] = -11

20- 9
-12 NEWPROCESSNOWAIT COMPLETION
Cause. A call to the NEWPROCESSNOWAIT procedure finished.

sysmsg[0] = -12
sysmsg[1].<0:7> = NEWPROCESS error
sysmsg[1].<8:15> = NEWPROCESS suberror or file-system
error
sysmsg[2] FOR 2 = Tag passed to NEWPROCESSNOWAIT in
filenames [36:37]
sysmsg[4] FOR 4 = Process ID of the new process
sysmsg[8] = NEWPROCESS error
sysmsg[9] = NEWPROCESS suberror or file-system
error
For more information on the NEWPROCESS error, refer to Section 5, NEWPROCESS
AND NEWPROCESSNOWAIT Errors.
Note. If 119 is returned in sysmsg[1].<8:15>, the file-system error number is greater than 255.
The actual error is in sysmsg[9].
-13 SYSTEM MESSAGE BUFFER OVERRUN
Cause. Some broadcast system messages were not delivered to this process’s
$RECEIVE file because the process did not read $RECEIVE as fast as the system
buffer was filled. Broadcast messages are messages that can go to every process
(system messages -2, -3, -8, -10, and -11).
The system delivers all broadcast messages generated from this time forward.
Format. sysmsg [0] = -13

20 -10
-20 BREAK ON DEVICE
Cause. (C-series-format system message only) The BREAK key was pressed on a
monitored terminal, or a process called SENDBREAKMESSAGE.
If the process specified break monitoring through a call to SETMODE or
SETMODENOWAIT, the operating system sends the process the following message.
sysmsg[0] = -20
sysmsg[1] = Logical device number, in binary, of device
where BREAK was pressed. If a process called
SENDBREAKMESSAGE, this field contains -1.
sysmsg[2] = System number, in binary, of logical device
number or SENDBREAKMESSAGE caller.
sysmsg[3] = The most significant word of the break tag.
sysmsg[4] = The least significant word of the break tag.
-21 3270 DEVICE STATUS RECEIVED
Cause. A call to SETMODE 53 was made by the application to monitor subdevice

status, pass the information to TR3271 by way of a call to SETMODE 51, and issue
this status message.
Format. The layout of the received message described in terms of an array of words
is:
sysmsg[0] = -21
sysmsg[1] = The response ID
sysmsg[2] = The actual 3271 status bytes:
.<0:7> = sense byte
.<8:15> = status byte
sysmsg[3] = A translation of the device status to status
bits; the application might pass
this word directly to TR3271 by way of SETMODE
51 to post the status on a TR3271 subdevice.
-22 ELAPSED TIME TIMEOUT
Cause. A timer set by a call to SIGNALTIMEOUT timed out.

Format. The layout of the received message described in terms of an array of words
is:
sysmsg[0] = -22
sysmsg[1] = parameter1 supplied to SIGNALTIMEOUT
(0 if none)

20- 11
sysmsg[2] FOR 2 = parameter2 supplied to SIGNALTIMEOUT

(0D if none)
-23 MEMORY LOCK COMPLETION
Cause. A call to the privileged procedure LOCKMEMORY waited for memory but
completed successfully before the specified time limit was reached.
sysmsg[0] = -23
sysmsg[1] = parameter1 supplied to LOCKMEMORY
sysmsg[2] FOR 2 = parameter2 supplied to LOCKMEMORY
(if none supplied, 0D)
-24 MEMORY LOCK FAILURE
Cause. A call to the privileged procedure LOCKMEMORY waited for memory but
timed out without completing the lock.
sysmsg[0] = -24
sysmsg[1] = parameter1 supplied to LOCKMEMORY
(if none supplied, 0)
sysmsg[2] FOR 2 = parameter2 supplied to LOCKMEMORY
(if none supplied, 0D)
-26 PROCESS TIME TIMEOUT
Cause. A timer set by a call to SIGNALPROCESSTIMEOUT timed out.

sysmsg[0] = -26
sysmsg[1] = parameter1 supplied to
SIGNALPROCESSTIMEOUT (0 if none)
sysmsg[2] FOR 2 = parameter2 supplied to
SIGNALPROCESSTIMEOUT (0D if none)

20 -12
-30 PROCESS OPEN
Cause. (C-series-format system message only) Either the process was opened by
another process or the backup process of a process pair opened a process (a process
receives two process open messages when opened by a process pair).
Format. The operating system sends the process the following message, provided the
process has opened its $RECEIVE file to receive file management system messages
either by setting the FILE_OPEN_ procedure parameter options.<15> to 0 or the
OPEN procedure parameter flags.<1> to 1:
sysmsg[0] = -30
sysmsg[1] = flags parameter to caller's OPEN
sysmsg[2] = sync-or-receive-depth parameter to
caller's OPEN
sysmsg[3] FOR 4 = 0 if normal OPEN; process ID of primary
process if an open by a backup process
sysmsg[7] = 0 if normal OPEN, negative of the file
number of file if an open by a backup
process
sysmsg[8] = Process accessor ID of opener
sysmsg[9] FOR 4 = Optional first qualified name of named
process or blanks
sysmsg[13] FOR 4 = Optional second qualified name of named
process or blanks
sysmsg[17].<14> = Set to 1 if the opener's process access
identification (given in the message)
has not been verified in the receiver's
node (although it passes the
remote password test)
sysmsg[17].<15> = Set to 1 if the opener is on a
different node from the receiving
process
Response. Obtain the process ID of the opener by a call to LASTRECEIVE or
RECEIVEINFO. Corrective action, if any, is application dependent.
-31 PROCESS CLOSE
Cause. (C-series-format system message only) Another process closed the receiver
process. The closing process can also be the backup process of a process pair, so a
process receives two process close messages when closed by a process pair.
sysmsg[0] = -31
Response. Obtain the process ID of the closer by a call to LASTRECEIVE or
RECEIVEINFO. Corrective action, if any, is application dependent.

20 -13
-32 PROCESS CONTROL
Cause. Another process called the CONTROL procedure while referred to the receiver
process file.
sysmsg[0] = -32
sysmsg[1] = Operation parameter to caller's CONTROL
sysmsg[2] = Parameter parameter to caller's CONTROL
Response. Obtain the process ID of the caller to CONTROL by a call to
LASTRECEIVE or RECEIVEINFO. The response, if any, is application dependent.
-33 PROCESS SETMODE
Cause. Another process referred to the receiving process in a call to the SETMODE or
SETMODENOWAIT procedure.
sysmsg[0] = -33
sysmsg[1] = function parameter to caller's SETMODE or
SETMODENOWAIT
sysmsg[2] = param1 parameter to caller's SETMODE
or SETMODENOWAIT
sysmsg[3] = param2 parameter to caller's SETMODE
or SETMODENOWAIT
If the receiving process can handle requests for last parameter information (by setting
param1.<15> in a SETMODE 80 call), the flags word is included:
sysmsg[4] = Flags word:
.<13> = 1 if param1 was supplied in the call to
SETMODE
.<14> = 1 if param2 was supplied in the call
to SETMODE
.<15> = 1 if last-params was supplied in the
call to SETMODE
Response. Obtain the process ID of the caller to SETMODE or SETMODENOWAIT
by a call to LASTRECEIVE or RECEIVEINFO. The response, if any, is application
dependent.

20 -14
A process that receives the extended (five-word) SETMODE message can return a
value to be placed in the last-params parameter of the SETMODE caller. The
process should call REPLY with a buffer that has the format:
replymsg[0] = -33
replymsg[1] = The previous value of param1
replymsg[2] = The previous value of param2
Supplying a reply when none is needed will not cause an error.
-34 PROCESS RESETSYNC
Cause. Another process called the RESETSYNC procedure while referring to the
receiver process file. A call to the CHECKPOINT procedure might contain an implicit
call to RESETSYNC.
sysmsg[0] = -34
Response. Obtain the process ID of the caller to RESETSYNC by a call to
LASTRECEIVE or RECEIVEINFO. A server process using the sync ID mechanism
should clear its local copy of the sync ID value. Corrective action, if any, is application
dependent.
-35 PROCESS CONTROLBUF
Cause. Another process called the CONTROLBUF procedure while referring to the
receiver process file.
sysmsg[0] = -35
sysmsg[1] = operation parameter to the caller's
CONTROLBUF
sysmsg[2] = count parameter to the caller's
CONTROLBUF
sysmsg[3] FOR n = buffer data from caller's CONTROLBUF,
where n is the number of words in
buffer
Response. Obtain the process ID of the caller to CONTROLBUF by a subsequent call
to LASTRECEIVE or RECEIVEINFO. The response, if any, is application dependent.

20 -15
-37 PROCESS SETPARAM
Cause. Another process referred to the receiving process in a SETPARAM call.

The operating system sends this message only if the receiving process has indicated it
will accept such messages by doing one of the following:
• Omitting the FILE_OPEN_ procedure options parameter, or setting
options.<15> = 0
• Setting flags.<1> = 1 in its call to OPEN on $RECEIVE
• Setting param1.<14> to 1 in a call to SETMODE 80
sysmsg[0] = -37
sysmsg[1] = function parameter to SETPARAM
sysmsg[2] = Flags word:
.<14> = 1 if param-array was supplied in
the call to SETPARAM
.<15> = 1 if last-param-array was supplied
in the call to SETPARAM
sysmsg[3] = param-count parameter specified in the call
to SETPARAM
sysmsg[4:n] = param-array parameter to SETPARAM with a
length of param-count bytes
Response. To return a value in the last-param-array parameter of the
SETPARAM caller, the receiving process should call REPLY with a buffer that has the
following format:
replymsg[0] = -37
replymsg[1] = The value in the range 0 through 256 for
last-param-count
replymsg[2:n] = The value for last-param-array, with a
length of last-param-count bytes
If the format of the reply is incorrect, error code 2 is returned to both the caller of
REPLY and the caller of SETPARAM.

20 -16
-38 QUEUED MESSAGE CANCELLATION
Cause. A pending message was canceled. (A pending message is a message read by

READUPDATE but not yet replied to.)
This message can occur at random relative to other messages. For example, if an
opener calls CLOSE or FILE_CLOSE_ while it has an outstanding operation, the
operational message is canceled and the close message is sent. The receiving
process cannot assume that the cancellation message appears on $RECEIVE before
the close message (or vice versa).
The absence of a cancellation message does not mean that the sending process
received a reply. A cancellation can occur at any time, including just before a reply or
while the reply data is awaiting delivery; in these cases, the cancellation message does
not appear.
Do not use the cancellation notice to determine the delivery status of replies.
Format. Message -38 can be received only if SETMODE 80 param.<13> has been set
to 1:
sysmsg[0] = -38
sysmsg[1] = The message-tag of the canceled message
Response. Call REPLY for the referenced message.
-40 DEVICE TYPE INQUIRY
Cause. (C-series-format system message only) Another process called DEVICEINFO,

DEVICEINFO2, or FILEINFO to request the device type or the physical record length
from the receiving process. This message is sent only to subtype 30 processes. Do not
attempt to use subtype 30 processes to emulate disk devices (device type 3). Calls
related to disk files (such as FILEINFO) return file-system error 2 even if the process is
emulating a disk file.
sysmsg[0] = -40
sysmsg[1:4] = The first qualifier name in internal
format (blank padded)
sysmsg[5:8] = The second qualifier name in internal
format (blank padded)
If the calling process did not specify qualifiers to the process name, sysmsg[1:4] and
sysmsg[5:8] are blank.
Response. The subtype 30 process should call REPLY with the needed information in a
buffer of the following format:
replymsg[0] = -40
replymsg[1] = The device type word:
.<0:3> = 0

20 -17
.<4:9> = device type

.<10:15> = device subtype
replymsg[2] = Physical record length
If the message response is incorrectly formatted, the DEVICEINFO[2] or FILEINFO
caller receives the default values (device type and subtype of zero) and the REPLY
caller receives error code 2.
-41 NOWAIT DEVICEINFO2 COMPLETION
Cause. The DEVICEINFO2 nowait option (options.<13>) was specified to obtain

device information. The device information is returned only in system message -41 on
$RECEIVE.
sysmsg[0] = -41
sysmsg[1:2] = Tag value from tag-or-timeout parameter
sysmsg[3] = File-system error code
sysmsg[4] = The devtype value
sysmsg[5] = The physical-recordlen value
sysmsg[6] = The diskprocess-version value

20 -18
The device information returned in words 4 through 6 is valid only if the file-system
error code (word 3) is zero. Words 1 and 2 return the value passed to DEVICEINFO2
to help you identify this particular completion when there are simultaneous inquiries.
-100 REMOTE PROCESSOR DOWN
Cause. The remote operating system has declared a processor to be down, and the
process had called MONITORNET to monitor remote status changes.
sysmsg[0] = -100
sysmsg[1] FOR 2 = Node number
sysmsg[4] = Length of node name, in bytes
sysmsg[5] FOR 3 = Reserved
sysmsg[8] FOR * = Node name (including the \)
This message is not sent if the last processor in a node fails; that case is reported as a
node failure (message -110).
-101 PROCESS DELETION
Cause. A process (pair) has terminated; the cause of termination is indicated in the
message. This message is sent:
• To the mom process if it exists and termination was not caused by processor
failure
• To the GMOM (NetBatch job ancestor) if it exists and termination was not caused
by processor failure
• To the ancestor if the terminated process was a named process without a backup.
In this case, the message is sent even if the process is terminated by a processor
failure.
If the same process fulfills multiple roles (for example, ancestor and GMOM), then it
receives only one copy of the deletion message.
If the terminated process was created by the NEWPROCESS procedure, or if the
“send to any ancestor” create option (create-options.<9> in the
PROCESS_CREATE_ call) was used, the deletion message is delivered to any
process with the same name as the original ancestor, regardless of sequence number.
Otherwise, the deletion message is delivered only to the original instance of a named
ancestor process, as indicated by having the original sequence number. All the
processes in an unbroken string of primary and backup processes with the same name
are considered part of the same instance and have the same sequence number.

20 -19
sysmsg[0] = -101
sysmsg[1] FOR 10 = Process handle of terminated process
(a FIXED value)
sysmsg[15] = Process job ID; 0 if the process is
not part of a job
sysmsg[16] = Completion code
sysmsg[17] = Termination information (0 if none
supplied)
The next three items contain the SPI subsystem ID:
sysmsg[18] FOR 4 = Subsystem organization: "HP" for
HP supplied subsystems; blank if
not supplied
sysmsg[22] = Subsystem number: zero if not supplied
sysmsg[23] = Subsystem version: zero if not supplied
sysmsg[24] FOR 10 = Process handle of external process
causing termination; null process
handle (all words set to -1) if none
sysmsg[34] = Length in bytes of termination text
sysmsg[35] = Offset in bytes (from beginning of
message) of process descriptor
of terminated named process (pair)
sysmsg[36] = Length in bytes of process descriptor
of terminated named process (pair)
sysmsg[37].<0:13> = Reserved
sysmsg[37].<14> = OSS system type: 1 if the terminated
process was an OSS process; 0 if
the terminated process was a Guardian
process
sysmsg[37].<15> = Abend: termination caused by ABEND if 1,
STOP if 0
sysmsg[38] FOR 2 = OSS process ID
sysmsg[40] = Reserved
sysmsg[41] FOR * = Termination text (80-byte limit). The
length of this item is in sysmsg[34].
It is zero length if no text was
supplied.
sysmsg[ ] FOR * = Process descriptor of terminated named
process (pair). The offset of this
item is in sysmsg[35]. The length of
this item is in sysmsg[36]. If the
terminated process is unnamed or is
the recipient's backup, this is zero
length.

20 -20
The returned process descriptor for a terminated named process (pair) is in the form:
\node.$name:seqno
Table 20-3 shows HP completion codes; we recommend that you use positive
completion codes in the same way. You can specify a completion code with any
positive value in a PROCESS_STOP_, STOP, or ABEND procedure call. Negative
completion codes are reserved for HP use.
For an OSS process terminating as a result of an exit() function call, the completion
code is set to the low-order 8 bits of the exit status.
Table 20-3. Completion Codes (page 1 of 4)

Completion
Code Definition
0 Normal, voluntary termination with no errors. This code is the default for
PROCESS_STOP_ (if abnormal termination is not specified) and STOP if no
completion code is specified, and for the OSS exit() function if no exit
status is specified.
1 Normal, voluntary termination with WARNING diagnostics. For example, if
the process is a compiler, the compilation terminated with WARNING
diagnostics after building a complete object file.
2 Abnormal, voluntary termination with FATAL errors or diagnostics. For
example, if the process is a compiler, the compilation terminated with FATAL
diagnostics and either an object file was not built or, if built, might be
incomplete. A complete listing is generated.
3 Abnormal, voluntary, but premature termination with FATAL errors or
diagnostics. For example, if the process is a compiler, the compilation
terminated with FATAL diagnostics, with either no object file or an incomplete
object file being built and an incomplete listing generated (the compiler quit
compiling prematurely).
4 Process never got started. This completion code exists primarily for the use
of the command interpreter or other command language interpreters that can
act as the executor process of a batch job. This code allows the executor
process to detect that a process associated with a RUN statement never got
started. In that sense, this completion code is a “fake” completion code. The
command interpreter acts as though it received a termination message from
the process that it tried to create, when in fact it received an error returned by
the procedure or OSS function that launched the process. The command
interpreter then makes the completion code and the error returned by the
procedure or OSS function that launched the process available for
evaluation, for example, by a batch job executor process.
5 Process calls PROCESS_STOP_ (with abnormal termination specified) or
ABEND on itself. This code is the default completion code for the
PROCESS_STOP_ procedure (when abnormal termination is specified) and
the ABEND procedure.

20 -21

Completion
Code Definition
6 PROCESS_STOP_, STOP, or ABEND was called to delete a process by an
external, but authorized, process. The system includes this completion code
in the process deletion message. If the process cannot be stopped, the
request is saved so that when the process calls SETSTOP this completion
code is sent with the process deletion message. The user ID, the
PCBCRAID (CAID) and the process ID of the process that caused the
termination, are included in the termination message.
7 Restart this job. This completion code is used by the NetBatch scheduler
and an executor process. The executor process sets its completion code to
this value upon termination; the scheduler interprets this completion code
and restarts a “restartable” job.
8 Code 8 is the same as code 1, normal termination, except that the user must
examine the listing file to determine whether the results are acceptable.
Completion code 8 is typically used by compilers.
9 The kill() or raise() OSS function generated a signal that stopped the
process. The termination information provides the signal number.
Note that if a signal is delivered to a signal handler that stops the process,
the completion code will be determined by the handler. For example, when a
signal stops a TNS/R native C program, a different completion code is
returned as set by the signal handler installed by the Common Run-Time
Environment (CRE).
-1 A trap was detected in a Guardian TNS process. If the system detects the
absence of a trap handler routine or encounters another trap in a trap
handler, then in addition to an abnormal termination, this completion code is
returned automatically in the process deletion (ABEND) message. The
contents of the text string vary with the state of the process. The first nine
characters are "TRAPNO=nn" with nn representing the trap number in
decimal. Then the text identifies the code space, including the TNS code
segment index when appropriate, and indicates whether the process was
privileged. Finally, the text displays key registers, depending upon the
execution mode of the process at the time of its termination: P or pc, L, and
S for TNS or accelerated mode; pc and sp for TNS/R native mode.
Examples:
Invalid address in TNS mode:
TRAPNO=00: (UC.00) P=%000012 L=%000001 S=%000003
Instruction failure in accelerated mode:
Arithmetic overflow (division by zero) in accelerated mode, privileged:
TRAPNO=02: (acc UC, Priv) pc=%h7042370C L=%023520 S=%023526
Limits-exceeded in TNS/R native mode, privileged:
TRAPNO=05: (SCr, Priv) pc=0x808E2EDC sp=0x5FFFFF00

20 -22

Completion
Code Definition
-2 This code is returned by the system when a process has terminated itself but
the system is unable to pass along the requested completion code and the
associated termination information due to a resource problem in the system.
-3 This code is returned by the system when a process terminating itself passed
bad parameters to PROCESS_DELETE_, STOP, or ABEND. In this case,
some or all of the information requested in the completion code message
may not be present. Since the process is stopping itself, it is stopped.
-4 This code is returned by the system when a processor failure caused the
name of a process to be deleted (that is, the only process running under that
name was in the processor that failed).
-5 A communications or resource failure occurred during the execution of one of
the OSS exec set of functions; or an initialization failure of the new process
occurred when it was too late for the exec set of functions to return an error
to its caller.
-6 An OSS process or TNS/R native process terminated when it caused a
hardware exception. The termination information field of the message
contains the signal number.
The termination text is in the message for all processes. However, while the
TACL command interpreter displays the termination text when it is present in
the message for a process created by TACL, OSS utilities such as osh
typically do not.
The text shows the signal number and name, identifies the code space, and
indicates whether the process was privileged. For a TNS/R native process,
the text displays the pc and sp registers. For an OSS process, it shows
registers appropriate to the mode, as for completion code -1.
Examples:
Invalid address in TNS/R native mode:
Signal 11, SIGSEGV: (UCr) pc=0x700024F0 sp=0x4FFFFC68
-7 An OSS process or TNS/R native process terminated as a result of a

corrupted stack frame or register state.
-8 An OSS process or TNS/R native process terminated because of insufficient
user stack space for signal delivery. Stack overflow generates completion
code -8, which is otherwise like completion code -6.
Example:
Stack overflow in TNS/R native mode:
Signal 25, SIGSTK: (UCr) pc=0x70000394 sp=0x4FEFFE18
-9 An OSS process or TNS/R native process terminated because of insufficient
PRIV stack space for signal delivery. The termination information field of the
message contains the signal number.

20 -23

Completion
Code Definition
-10 An OSS process or TNS/R native process terminated because it was unable
to obtain resources for signal delivery. The termination information field of
the message contains the signal number.
-11 An OSS process or TNS/R native process terminated because it attempted
to resume from a nonresumable signal. The termination information field of
the message contains the signal number.
-12 One of the OSS exec or tdm_exec set of functions executed successfully.
The OSS process ID continues to exist as it migrates to another process
handle, but the original process handle is deleted. Call
PROCESS_GETINFOLIST_ to obtain the new process handle of the OSS
process.
-13 The OSS open() or dup() function performed by the PROCESS_SPAWN_
procedure failed. The termination information in sysmsg[17] contains the
OSS errno for the error that occurred. The subsystem ID in sysmsg[18]
contains the null value. The termination text in sysmsg[41] can contain
additional information.

20 -24
-102 NOWAIT PROCESS_LAUNCH_ OR PROCESS_CREATE_ COMPLETION
Cause. A nowait call to the PROCESS_LAUNCH_ OR PROCESS_CREATE_

procedure finished.
sysmsg[0] = -102
sysmsg[1] FOR 2 = nowait-tag supplied to PROCESS_CREATE_
sysmsg[3] FOR 10 = process-handle of new process
sysmsg[13] = error
sysmsg[14] = error-detail
sysmsg[15] = Length in bytes of process descriptor
of new process
sysmsg[20] FOR * = Process descriptor of new process
(the length of this item is in
sysmsg[15])
If the new process is unnamed, then the returned process descriptor is in unnamed
form:
\node.$:cpu:pin:seqno
If the new process is named, then the returned process descriptor is in named form:
\node.$name:seqno
When a named process creates its backup, the returned process descriptor is in
named form.
If this message returns an error value indicating that process creation failed, then it
does not return a valid process handle or a process descriptor; the process descriptor
length is 0.

20 -25
-103 PROCESS OPEN
Cause. The receiving process was opened by another process.

sysmsg[0] = -103
sysmsg[1] = Access mode (from the access
FILE_OPEN_ parameter)
sysmsg[2] = Exclusion mode (from the exclusion
sysmsg[3] = Nowait depth (from the nowait
sysmsg[4] = Sync depth (from the
sync-or-receive-depth FILE_OPEN_
parameter)
sysmsg[5] = Open options (from the options
sysmsg[6] = User ID of opener (process access ID)
sysmsg[7] = Miscellaneous:
.<0:12> = currently undefined; subject
to change
.<13> = the opener's user ID has not
been verified locally (in the
receiver's node0); in any case
it will have passed a remote
password check
.<14> = the opener is on a different
node from the receiver.
.<15> = backup open: this is an open
by a backup
sysmsg[8] FOR 10 = For a backup open, this contains the
process handle of the primary process;
for a normal open, this is a null
process handle
sysmsg[18] = The length in bytes of the qualifier
name given below
sysmsg[19] = The offset in bytes from the beginning
of the message to the beginning of the
opener process name appearing below
sysmsg[20] = The length in bytes of the opener
process name appearing below
sysmsg[21] = For a backup open, the file number
used by the primary. It is typically,
but not always, the same as the
backup. Unlike the old open message,
this value is never negative.
sysmsg[22] = The creator access ID of the opener.
Unlike the process access ID given
earlier in the message, this ID is not

20 -26
verified by remote password checking.

opener home terminal name
sysmsg[24] = The length in bytes of the opener home
terminal name appearing below
sysmsg[25] FOR 5 = Reserved, subject to change
sysmsg[30] FOR * = The qualifier portion of the name used
to open the process, in external form
(for example, "#PORT2.CTL"). The
length of this item is in sysmsg[18].
It is zero length if no qualifier was
given.
sysmsg[ ] FOR * = For a named opener, this is the
process name in external process
descriptor form (system, name, and
sequence number). If the opening
process is unnamed, the length is
zero. In either case, the opener's
process handle is available from
FILE_GETRECEIVEINFO_. The offset and
length of the name are in sysmsg[19]
and sysmsg[20].
sysmsg[ ] FOR * = The opener's home terminal name in
external form. This is zero
length if the opening operating system
version is earlier than C10. The
offset and length of the name are in
sysmsg[23] and sysmsg[24].
Response. If the receiver wishes to reject the open, it should call REPLY with an
appropriate file-system error code (> 9). If the application is tracking openers, the
opener identified from FILE_GETRECEIVEINFO_ should be added to its table and, in
the case of a backup open, associated with the primary open. If the application wishes
to have a particular value (typically an open table index) returned in the OpenLabel field
from FILE_GETRECEIVEINFO_ on later messages from this opener, it should call
REPLY with the desired value in the reply data buffer in the following form:
repmsg[0] = -103
repmsg[1] = OpenLabel value

20 -27
-104 PROCESS CLOSE
Cause. The receiving process was closed by another process.

sysmsg[0] = -104
sysmsg[1] = tapedisposition parameter to FILE_CLOSE_
Response. If the application is tracking openers, the receiver should call
FILE_GETRECEIVEINFO_ to obtain the sender’s process handle, which can then be
used to search the application’s tables for the entry to be deleted.
-105 BREAK ON DEVICE
Cause. The BREAK key was pressed on a terminal or other device for which the
application has enabled break monitoring.
sysmsg[0] = -105
sysmsg[1] = File number of the receiver's open
file to the terminal that indicated
break (or -1 if unavailable).
Before D00, this is the file number
to that device (or, if there is more
than one, it can be the number of
any of the files). The device must
be open.
sysmsg[2] FOR 2 = The break tag value specified with
SETPARAM (if used)
Response. The application should take action in one of the ways described in the

20 -28
-106 DEVICE TYPE INQUIRY
Cause. The receiving process (which must have subtype 30 to receive this message)
was the subject of a call to FILE_GETINFOBYNAME_, DEVICEINFO, or a similar
function requesting device-type information.
sysmsg[0] = -106
sysmsg[4] = Length in bytes of the qualifier part
of the file name being inquired about
(zero if none given)
sysmsg[5] FOR * = The qualifier part of the file name
being inquired about, in external
form (the length is given in the
previous field)
Response. The subtype 30 process should REPLY with the needed information in the
following form:
repmsg[0] = -106
repmsg[1] = Device type
repmsg[2] = Device subtype
repmsg[3] FOR 3 = Reserved, must be filled with -1
repmsg[6] = Physical record length

20 -29
-107 SUBORDINATE NAME INQUIRY
Cause. The receiving process (which must have requested these messages by an
explicit call to PROCESS_SETINFO_) is being queried for subordinate names by
another process calling FILENAME_FINDNEXT_.
sysmsg[0] = -107
sysmsg[1] = Flags:
.<0:14> = Currently undefined; subject
to change
.<15> = Skip if same; if set and the
start name given below
exists, skip it and return
the following name. If
equal to 0, return the start
name if it exists.
sysmsg[2] = Length in bytes of the start name
(zero if no start name supplied)
pattern appearing below
sysmsg[4] = Length in bytes of the pattern
sysmsg[8] FOR * = The start name: the qualifier part of
the file name, in external form, at
which to start searching for a name to
sysmsg[2]
sysmsg[ ] FOR * = The pattern: the qualifier part of
the pattern for which a name is to be
returned (the offset and length are
given above)
Response. The process should search its list of subordinate names, starting with (or
after, if the Skip If Same option is used) the start name (or from the beginning, if the
start name is zero length). The process should return the first name that matches the
pattern (as determined by FILENAME_MATCH_). If there are no matching names, a
REPLY with error 1 should be made. When replying with a name, subtype 30 processes
must supply more information than normal processes, as indicated below; for normal
processes, these fields are ignored. To return a matching name, the process should
REPLY with a data buffer in the following form:
repmsg[0] = -107
repmsg[1] = For subtype 30 processes: device type
repmsg[2] = For subtype 30 processes: device
subtype
repmsg[3] FOR 3 = For subtype 30 processes: reserved,
must be filled with -1

20 -30
repmsg[6] = Returned name length in bytes

repmsg[7] FOR * = Returned name (external form
qualifier, as for example
"#PORT1.CTL")
-108 NOWAIT FILE_GETINFOBYNAME_ COMPLETION
Cause. The receiving process called FILE_GETINFOBYNAME_ specifying that the

information be obtained in a nowait manner, and the information is now available.
sysmsg[0] = -108
sysmsg[1] FOR 2 = The tag value from the timeout-or-tag
parameter to FILE_GETINFOBYNAME_
sysmsg[3] = The file-system error code giving the
resultant status of the request
sysmsg[4] FOR 5 = The type information (device type,
and so on) having the same layout as
described for the typeinfo parameter
of FILE_GETINFOBYNAME_
sysmsg[9] = The physical record length value
sysmsg[10] = The flag value from the flags
parameter to FILE_GETINFOBYNAME_
.<0:14> = Currently undefined; subject
to change
.<15> = 0 signifies a Guardian file
.<15> = 1 signifies an OSS file
Response. The process can use the information returned (if sysmsg[3] indicates no
error occurred).

20 -31
-109 NOWAIT FILENAME_FINDNEXT_ COMPLETION
Cause. The receiving process called FILENAME_FINDNEXT_ on a searchid that

had a nowait operation specified for it by FILENAME_FINDSTART_. The results of the
find request are being reported.
sysmsg[0] = -109
sysmsg[1] = The searchid on which the request was
made
sysmsg[2] = The file-system error code giving the
resultant status of the request
sysmsg[3] FOR 5 = The type information (device type, and
so on) having the same layout as
described for the entityinfo parameter
of FILENAME_FINDNEXT_. If sysmsg[2]
<> 0 the value of this field is
undefined.
sysmsg[8] = Length in bytes of the returned name
(zero if none present because of an
error)
sysmsg[9] FOR 2 = The tag parameter from
FILENAME_FINDNEXT_
sysmsg[14] FOR * = The returned name, in external form
with a length given in sysmsg[8]
Response. The process can use the information returned (if sysmsg[2] indicates no
error occurred).
-110 LOSS OF COMMUNICATION WITH NETWORK NODE
Cause. The remote node has either gone down or become partitioned from this node,
and the process had called MONITORNET to enable reception of remote status
change messages.
sysmsg[0] = -110
sysmsg[4] FOR 2 = Node identifier
sysmsg[7] FOR n = Node name (including the \)

20 -32
-111 ESTABLISHMENT OF COMMUNICATION WITH NETWORK NODE
Cause. The remote node has established connection with this node, and the process
had called MONITORNET to enable reception of remote status change messages.
sysmsg[0] = -111
sysmsg[7] FOR n = Node name (including the \)
Response. The response, if any, is application-dependent.
-112 JOB PROCESS CREATION
Cause. The receiving process is the supervisor of a job and a process running under
the supervisor’s GMOM job ID created a third process.
sysmsg[0] = -112
sysmsg[1] = Job ID
sysmsg[2] FOR 10 = Process handle of the newly created
process
-113 REMOTE PROCESSOR UP
Cause. The remote operating system has reloaded a processor, and the process had
called MONITORNET to monitor remote status changes.
sysmsg[0] = -113
sysmsg[8] FOR * = Node name (including the \)
-121 PATHSEND DIALOG ABORT
Cause. A Pathsend dialog has been aborted for one of the following reasons:
• The requester aborted the dialog explicitly by calling the
SERVERCLASS_DIALOG_ABORT_ procedure.
• The requester process abended.

20 -33
• The requester canceled the last server-class send operation in the dialog.
Format. The operating system sends the server process the following message,
provided the server has opened its $RECEIVE file to receive file-management system
messages either by setting the FILE_OPEN_ procedure parameter options.<15> to
0 or by setting the OPEN procedure parameter flags.<1> to 1:
sysmsg [0] = -121
Response. The server process should reply to this system message with an error value
of either FEOK (0) or FEEOF (1); these values direct the LINKMON process to release
the link for re-use. To identify the dialog to which this message applies, the server must
first call FILE_GETRECEIVEINFO_ to obtain the file number and process handle
associated with this message. For more information about Pathsend dialogs, refer to the
NonStop TS/MP Pathsend and Server Programming Manual.

20 -34
-141 NOWAIT PROCESS_SPAWN_ COMPLETION
Cause. A call to the PROCESS_SPAWN_ procedure was completed.

sysmsg[0] = -141
sysmsg[14] FOR 2 = nowait-tag supplied to PROCESS_SPAWN_
sysmsg[16] FOR 2 = ZSYS-DDL-PROCESSRESULTS.Z-LEN of
process-results supplied to
PROCESS_SPAWN_
sysmsg[18] FOR 10 = ZSYS-DDL-PROCESSRESULTS.Z-PHANDLE of
process-results (process handle of
the new process)
sysmsg[28] FOR 2 = ZSYS-DDL-PROCESSRESULTS.Z-PID of
process-results (OSS pid of the
new process)
sysmsg[30] FOR 2 = ZSYS-DDL-PROCESSRESULTS.Z-ERRNO of
process-results (OSS errno)
sysmsg[32] = ZSYS-DDL-PROCESSRESULTS.Z-TPCERROR of
process-results (Guardian error)
sysmsg[33] = ZSYS-DDL-PROCESSRESULTS.Z-TPCDETAIL of
process-results (Guardian error
detail)
If this message returns an error value indicating that process creation failed, then it
does not return a valid process handle or OSS process ID.

20 -35
-147 DEVICE INFORMATION INQUIRY
Cause. The receiving process (which must have subtype 30 to receive this message)
was the subject of a call to CONFIG_GETINFO_BYLDEV_,
CONFIG_GETINFO_BYNAME_, or a similar procedure requesting physical device
information. This message is used only on G-series releases.
sysmsg[0] = -147
sysmsg[1] = Message version (must be equal to
ZSYS-VAL-SMSG-CONFIGINFO-VERS)
sysmsg[2] FOR 4 = Device name for which configuration info
is needed; blank-filled (can be all blanks)
sysmsg[6] FOR 4 = Subdevice name; blank-filled (can be all
blanks)
sysmsg[10] FOR 4 = Secondary subdevice name qualifier;
blank-filled (can be all blanks)
Response. The receiver responds to system message -147 as follows:
If the receiver... it calls REPLY with file-system
error...
Does not handle message -147 2 (operation not allowed)
Does not recognize the version of the 565 (malformed request denied)
received message
Does not recognize the device name 14 (no such device)
or one of its qualifiers
If none of the preceding conditions apply, the receiver returns the requested
information by calling REPLY with file-system error 0 (operation successful) and with
data placed in the reply data buffer in the following form:
repmsg[0] = -147
repmsg[1] = Message version (must be equal to
ZSYS-VAL-SMSG-CONFIGINFO-VERS)
repmsg[2] FOR 2 = Device type; 32-bit value (can be -1D
for null)
repmsg[4] = Device subtype (can be -1 for null)
repmsg[5] = Device record size (must contain a
positive value)
repmsg[6] = Logical status; the possible values are
defined in the file ZCOMDDL.
repmsg[7] = Length in bytes of symbolic name of
physical hardware excluding the trailing
null (can be from 1 to 63)
repmsg[8] FOR * = Symbolic name of physical hardware (null
terminated). This is the same name that
is held in the CONFIG file and is the same
as the SCF object name.
repmsg[40] = Length in bytes of subsystem manager name
20 -36
excluding the trailing null (can be from 0

to 47)
repmsg[41] FOR * = Subsystem manager name (null terminated);
can be either a simple name (for example,
$PDEV1) or a complete process descriptor
repmsg[65] FOR 10 = Process handle for primary I/O process;
can be null (that is, all words set to -1)
repmsg[75] FOR 10 = Process handle for secondary I/O process;
can be null (that is, all words set to -1)
repmsg[85] FOR 10 = Reserved for future use
repmsg[95] = Length in bytes of device-specific
information; must be greater than or equal
to 0 and less than 2048
repmsg[96] FOR * = Device-specific information; as they are
made available, structures defined by the
various subsystems are described in the
appropriate subsystem manuals.
Note. When evaluating the size of the data structure that contains everything preceding the
device-specific information (as declared in ZSYS* files), TAL and pTAL give a different result
from C. For example, if this data is declared as a structure called data, the correct length in
TAL and pTAL would be $len(data)-1; the correct length in C would be sizeof(data)-2.

20 -37
21 Traps and Signals
Certain critical problems can cause a process to be unable to continue executing.
These are typically the result of coding errors, but other conditions, such as the lack of
a system resource (for example, memory), can also prevent normal process execution.
Such conditions are reported as traps to TNS processes and as signals to TNS/R
native processes. A trap is a software mechanism that stops process execution. A
signal is a software interrupt that can notify a process of other events, such as timer
expiration, as well as of critical error conditions.
The set of signals that are used in the Guardian environment is known as TNS/R
native signals. This set is a subset of a larger set of signals used in the Open System
Services (OSS) environment. Most of the TNS/R native signals are caused by the
same conditions that cause traps to occur in TNS processes. TNS processes in the
Guardian environment do not receive signals, except under rare circumstances that
cause a trap 8 to be generated.
In this section, equivalent trap and signal conditions are described together. Signals
are commonly referred to by name, where each name is a literal that stands for a 32-bit
signal number.
Table 21-1 lists the TNS/R native signals by name, in alphabetic order, along with their
signal numbers, the equivalent trap numbers, and brief trap descriptions.
Table 21-1. TNS/R Native Signal Names, Signal Numbers, Trap Numbers, and
Trap Descriptions
Signal Trap
Signal Name Number Number Trap Description
SIGABRT 6D none (not available to TNS processes)
SIGALRM 14D none (not available to TNS processes)
SIGFPE 8D 2 Arithmetic overflow
SIGILL 4D 1 Instruction failure
SIGLIMIT 27D 5 Limits exceeded
SIGMEMERR 22D 13 Uncorrectable memory error
SIGMEMMGR 24D 11 Memory manager read error
SIGNOMEM 23D 12 No memory available
SIGSEGV 11D 0 Illegal address reference
SIGSTK 25D 3 Stack overflow
SIGTIMEOUT 26D 4 Process loop-timer timeout
other signals -- 8 (signal delivered to TNS process)

21-1
Traps and Signals Trap Handling
Trap Handling
When a trap occurs, control passes to the Debug or Inspect debugger by default.
However, by using the ARMTRAP system procedure, a program can specify a trap
handler to be executed when the trap occurs or, alternatively, that a trap causes the
program to abend. The actions of a trap handler typically include either terminating the
process or, in certain cases, resuming execution.
For information about the Debug and Inspect debuggers, refer to the Debug Manual
and the Inspect Manual.
For a description of the ARMTRAP procedure, see the Guardian Procedure Calls
Reference Manual. For an explanation of how to create your own trap handler with
ARMTRAP, see the Guardian Programmer’s Guide.
Signal Handling
When a process receives a TNS/R native signal, the default action is for the process to
abend. Alternatively, the program can call the SIGACTION_INIT_ procedure to specify
that another action should result, such as the execution of a signal handler. If a signal
handler has been specified, it is executed when the process receives a signal. The
actions of a signal handler typically include either terminating the process or, in certain
cases, resuming execution.
For additional information about signals and signal handling, refer to the Guardian
Programmer’s Guide. For a description of the SIGACTION_INIT_ procedure, see the
Guardian Procedure Calls Reference Manual.
Signals and Trap Numbers

This subsection lists each signal by name with its equivalent trap number and provides
a description of each. In general, it includes only signals that are generated by the
system; additional signals can be generated by the raise() function. The list is in
order by trap number. Signals that have no equivalent trap are listed at the end of this
subsection.
SIGSEGV, Trap 0 Illegal address reference
Cause. An address was specified that was not within either the virtual code area or
the virtual data area allocated to the process.
Note. If a TNS process has fewer than 32 pages of data stack, then a situation such as an
infinitely recursive TNS procedure causes a trap 0 instead of a trap 3.
References to privileged 32-bit addresses from unprivileged code are reported as SIGSEGV or
trap 0.

21- 2
Traps and Signals Signals and Trap Numbers
Effect. For trap 0, process control passes to the Debug or Inspect debugger, or to the
trap handler designated by a call to the ARMTRAP procedure, or else the process is
abended. See Trap Handling on page 21-2 for further information.
For the signal SIGSEGV, the default or specified signal handling occurs. See Signal
Handling on page 21-2 for further information.
Recovery. If the illegal address reference occurred because of an erroneous pointer
value, correct the program. For TNS processes, if the address is lower than 65535.
refer to the recovery information for trap 3.
SIGILL, Trap 1 Instruction failure
Cause. The instruction failure is caused by one of the following:

• There was an attempt to execute a code word that is not a valid instruction.
• A nonprivileged process tried to either execute a privileged instruction or access a
privileged memory address.
• An EXTENSIBLE procedure was called with a calling sequence that involved a
mismatch. Most commonly, the mismatch is the following: The calling program was
compiled using an external declaration that specified more parameters than were
declared for the procedure being called, and the calling program passed one of
these unsupported parameters.
• The program has executed a deliberate failure instruction emitted by the compiler
to cause a trap when an error situation was detected at run time. (For example, a
pTAL program executed an END statement without encountering a RETURN
statement.)
Effect. For trap 1, process control passes to the Debug or Inspect debugger or to the
For the signal SIGILL, the default or specified signal handling occurs. See Signal
Recovery. Check your application program.

21- 3
SIGFPE, Trap 2 Arithmetic overflow
Cause. Overflow trapping is enabled, and an arithmetic overflow occurred. In the TNS
environment, overflow traps are enabled when the T bit (ENV.<8>) is set to 1. In
TNS/R native mode, overflow checking is controlled statically at compile time.
Arithmetic overflow occurs for one of the following reasons:
• The result of a signed arithmetic operation could not be represented with the
number of bits available for the particular data type.
• A division operation with a divisor of zero was attempted, or unsigned integer
division resulted in a quotient that exceeded the word size.
• The control variable of a CASE expression, or a CASE statement with no
OTHERWISE label, did not match any of the defined cases.
For the signal SIGFPE, the default or specified signal handling occurs. See Signal
Recovery. If the overflow occurred because of an erroneous computation, correct the
program.
If overflow is legitimate at this point, either use unsigned arithmetic (for 16-bit values)
or disable overflow traps during the computation. One example of when overflow is
sometimes legitimate is an address computation.
For trap 2, the program can use a trap handler to detect the overflow and recover. The
recovery action can be either to resume the computation or to transfer control to a
recovery point. However, resuming the computation at the point of the trap is not an
option if the trap occurred in an operating-system procedure.
Note that before returning to the program at the trap point or elsewhere, the trap
handler must clear the V bit (bit 10) in the trap handler’s copy of the ENV register at L[-
1]. Otherwise, the process will abend.
A signal handler cannot resume the process at the site of the overflow.

21- 4
SIGSTK, Trap 3 Stack overflow
Cause. The cause of the stack overflow varies with the type of process.
• In TNS or accelerated mode, a stack overflow results when one of the following
problems occurs:
• The value in the TNS stack register, S, exceeds 32767. Typically, an attempt
was made to execute a procedure or subprocedure whose stack marker,
sublocal data area, or local data area extends into the upper 32K of the user
data segment.
• There was not enough remaining virtual data space for an operating-system
procedure to execute.
The amount of virtual data space available is the lesser of 'G'[32767] and the upper
bound of the process's virtual data area (the number of data pages specified when
the process was created or run).
Operating-system procedures require approximately 350 words of user-data stack
space to execute.
Note
If you have fewer than 32 pages of data stack, then a situation such as an infinitely
recursive TNS procedure causes a trap 0 instead of a trap 3.
• In TNS/R native mode, the cause of the stack overflow is as follows:

A process running in TNS/R native mode has set the stack pointer register (SP) to
point beyond the allocated RISC stack but within the same memory region, and the
process has attempted to access a memory address between the end of the stack
and the address designated by the SP. Typically, this situation occurs when a
procedure or subprocedure was called with its activation record extending beyond
the end of the RISC stack.
For the signal SIGSTK, the process abends or enters the Debug or Inspect debugger.
A SIGSTK signal cannot be delivered to a signal handler because there is no space on
the stack for invoking the handler. See Signal Handling on page 21-2 for further
information.
Recovery. Recovery is application dependent. If the problem is infinite recursion of
procedure calls, correct the program.

21- 5
For TNS processes, you can increase the size of the virtual data area with one of the
following:
• The TAL compiler DATAPAGES directive
• The MEM parameter of the command interpreter RUN command
• The memory-pages parameter of either the NEWPROCESS or
PROCESS_CREATE_ procedure, the Z^MEMORY^PAGES field of the
param-list parameter of the PROCESS_LAUNCH_ procedure, or the
Z^MEMORYPAGES field of the process-extension parameter of the
PROCESS_SPAWN_ procedure.
If there is simply too much data for the TNS stack segment, you can revise the
program by putting some of its global or local data into an extended data segment.
For TNS/R processes, the main stack grows dynamically as needed, up to a maximum.
You can increase the maximum when the process is created, up to a limit.
For further information about managing memory, refer to the Guardian Programmer’s
Guide.
SIGTIMEOUT, Trap 4 Process loop-timer timeout
Cause. The time limit specified in the latest call to SETLOOPTIMER has expired.
For the signal SIGTIMEOUT, the default or specified signal handling occurs. See
Signal Handling on page 21-2 for further information.
Note that the loop-timer trap or signal is disabled after the trap or signal is generated;
the loop-timer trap or signal will not recur unless it is rearmed by another call to
SETLOOPTIMER.
Recovery. Recovery is application dependent. It is possible to resume from both trap
4 and SIGTIMEOUT, continuing the program from the site of the interruption.
SIGLIMIT, Trap 5 Limits exceeded
Cause. This condition occurs for one of the following reasons:

• The process called an operating-system procedure that tried to place a process
identification number (PIN) greater than 255 into a 1-byte field.
• A TNS Guardian process invoked a procedure or function that is allowed only for
an OSS or TNS/R native process.

21- 6
For the signal SIGLIMIT, the default or specified signal handling occurs. See Signal
Recovery. Run the process at a low PIN or recode the process.
Trap 8 (Signal delivered to TNS process)
Cause. A trap 8 is generated when, under very rare circumstances, a signal is

delivered to a TNS process.
Effect. Process control passes to the Debug or Inspect debugger or to the trap
handler designated by a call to the ARMTRAP procedure, or else the process is
Recovery. Contact your system manager.
SIGMEMMGR, Trap 11 Memory manager read error
Cause. A hard (unrecoverable) read error occurred while the program was trying to
bring a page in from virtual memory.
For the signal SIGMEMMGR, the default or specified signal handling occurs. See
If the trap or signal occurred because the disk containing the swap file went down, the
process abends, regardless of the selected trap or signal handling.
Recovery. Recovery is application dependent. Usually, this condition occurs when
both the primary and backup paths to a disk are down. Contact your system manager
to resolve the problem.
SIGNOMEM, Trap 12 No memory available
Cause. This error occurs for one of the following reasons:

• A page fault occurred, but no physical memory page is available for overlay.
• Disk space cannot be allocated when the program is referencing an extensible
segment.
For the signal SIGNOMEM, the default or specified signal handling occurs. See Signal

21- 7
Recovery. If the program was using extensible segments, either free up disk space on
the swap volume or specify a swap file on a different volume.
SIGMEMERR, Trap 13 Uncorrectable memory error
Cause. An uncorrectable memory error occurred.

For the signal SIGMEMERR, the default or specified signal handling occurs. See
Recovery. If possible, rerun your application. Otherwise, contact your system
manager.
SIGABRT (No trap) Abnormal termination
Cause. A procedure running within the process called abort() or

raise(SIGABRT).
Effect. For the signal SIGABRT, the default or specified signal handling occurs. See
SIGALRM (No trap) Timer expiration
Cause. A procedure called alarm() and the specified time has elapsed.
Effect. For the signal SIGALRM, the default or specified signal handling occurs. See
Recovery. Recovery is application dependent. It is possible to resume from SIGALRM
and to continue the program from the site of the interruption.

21- 8
Traps and Signals Error Lists
Error Lists
subsystem, you might receive a trap-out error list in a response. HP subsystems return
such an error list when, in performing your request, they encounter either a trap
condition or an exception condition that causes a signal to be generated.
Each error list always includes the unconditional tokens listed under its description. If
you are designing a subsystem that uses SPI, include all unconditional tokens listed in
the error list's description.
information about creating error lists, for additional information about tokens and token
types, and for definitions of tokens whose names begin with ZSPI-, refer to the SPI
Programming Manual.

21- 9
Traps and Signals 5: ZGRD-VAL-TRAPOUT
5: ZGRD-VAL-TRAPOUT
A user-enabled trap or signal handler started execution following the detection of a trap
or exception condition while a process was executing.

Conditional Tokens

ZGRD-TKN-ENTRYPOINTLABEL token-type ZSPI-TYP-STRING.
ZGRD-TKN-BINDTIMESTAMP token-type ZSPI-TYP-TIMESTAMP.
ZGRD-TKN-STACKENV token-type ZSPI-TYP-UINT.
ZGRD-TKN-SREGISTER token-type ZSPI-TYP-UINT.
ZGRD-TKN-PREGISTER token-type ZSPI-TYP-UINT.
ZGRD-TKN-EREGISTER token-type ZSPI-TYP-UINT.
ZGRD-TKN-LREGISTER token-type ZSPI-TYP-UINT.
ZGRD-TKN-SIGNAL-NUM token-type ZSPI-TYP-INT2.
ZGRD-TKN-TRAP-TEXT token-type ZSPI-TYP-STRING.
Z-ERROR is the trap or signal number.
ZSPI-TKN-PROC-ERR is the procedure code. Its value is ZGRD-VAL-TRAPOUT (5).
Conditional Tokens
ZGRD-TKN-XOBJECTFILE is the name of the program file associated with the process
that encountered the trap or signal.
ZGRD-TKN-ENTRYPOINTLABEL is the internal procedure entrypoint label of the
trapping procedure (IPIL entry).
ZGRD-TKN-BINDTIMESTAMP is the program file's nld utility or Binder timestamp.
ZGRD-TKN-STACKENV is the value of the stack-marker ENV register at the time the
trap occurred. This token is applicable only to TNS processes.
ZGRD-TKN-SREGISTER is the value of the S register at the time of the trap. This token
is applicable only to TNS processes.
ZGRD-TKN-PREGISTER is the value of the P register at the time of the trap. This token

21 -10
Traps and Signals 5: ZGRD-VAL-TRAPOUT
ZGRD-TKN-EREGISTER is the value of the E register at the time of the trap. This token
ZGRD-TKN-LREGISTER is the value of the L register at the time of the trap. This token
ZGRD-TKN-SIGNAL-NUM is the signal number of the signal that was received. This
token is applicable only to TNS/R native processes.
ZGRD-TKN-TRAP-TEXT is a text representation of the process state. It is
recommended that your program use the HIST_INIT_ and HIST_FORMAT_
procedures to produce the text, as these procedures are capable of handling and
formatting the process state of both TNS and TNS/R native processes. To produce the
most concise text, specify the options HO_Init_uContext and HO_OneLine in the call to
HIST_INIT_ and specify the options HF_base and HF_LocLineTNS (or
HF_LocLineRISC, as appropriate) in the call to HIST_FORMAT_. For further
information about the HIST_INIT_ and HIST_FORMAT_ procedures, refer to the
Guardian Procedure Calls Reference Manual.
Effect
The process might not be able to continue executing.
Recovery
Take corrective action as indicated by the returned trap or signal number. Traps and
signals are described earlier in this section.

21-11
22 OSS Error Information
This section describes how to find more information on Open System Services (OSS)
errors.
Brief OSS Error Information

Guardian procedures that make use of OSS functions can return OSS errno values in
addition to Guardian file-system error codes.
You can obtain a short explanation of any OSS errno value or Guardian file-system
error code at the TACL prompt by entering:
ERROR number
If you want to scan a list of all the errors, enter:
ERROR -1
Detailed OSS Error Information

Obtaining a more detailed explanation of an OSS error requires two steps:
1. In the errnoh file, look up the OSS errno symbolic name that corresponds to the
OSS errno value returned.
2. To find the error description, look up the OSS errno symbolic name in the Open
System Services System Calls Reference Manual.
Example
For example, to find a description for OSS errno value 4028 from the Guardian
environment, perform the following steps:
1. Using an editor, search the errnoh file for the OSS errno symbolic name
corresponding to 4028:
> edit $system.system.errnoh read
TEXT EDITOR - T9601B30 - (08MAR87)
CURRENT FILE IS \MYSYS.$SYSTEM.SYSTEM.ERRNOH
*list unseq "4028"
#define ENOSPC 4028 /* No space left on device */
...
*exit
>
The OSS errno symbolic name corresponding to 4028 is ENOSPC.
2. Refer to the Open System Services System Calls Reference Manual to find the
description of ENOSPC:
[ENOSPC] No space left on device. During the write( ) function on a regular file or
when extending a directory, there is no free space left on the device.
22- 1
Index
A Error codes (continued)
INITIALIZER procedure 18-1/18-4
ALLOCATESEGMENT procedure IOEdit procedures 16-2/16-4
error codes 10-1
NEWPROCESS procedure 5-1/5-19
error list 10-5
NEWPROCESSNOWAIT
guidelines for using 10-4 procedure 5-1/5-19
OSS 22-1
C PROCESS_CREATE_
C language error numbers for file-system procedure 6-1/6-35
errors 2-3 PROCESS_GETINFOLIST_
Command interpreter messages 1-3, procedure 7-1
19-3/19-20 PROCESS_GETPAIRINFO_
description of 19-1 procedure 8-1
Completion codes 20-21 PROCESS_LAUNCH_
C-series procedure errors 1-2 procedure 6-1/6-41
PROCESS_SPAWN_ procedure 6-1,
9-1, 9-2
D SEGMENT_ALLOCATE_
DEFINE procedure error codes 4-1 procedure 11-1
DSM/SM errors sequential I/O (SIO)
See DSM/Storage Manager errors procedures 3-1/3-7
DSM/Storage Manager errors 2-1 Subsystem Programmatic Interface
(SPI) 14-1/14-8
E Error lists 1-3
ALLOCATESEGMENT procedure 10-5
EDITREAD procedure error codes 15-1
guidelines for using 10-4
EDITREADINIT procedure error
codes 15-1 file system 2-96/2-157
errno values for file-system errors 2-3 conditional tokens 2-96
Error codes guidelines for using 2-96
ALLOCATESEGMENT procedure 10-1 ZFIL-VAL-AWAITIO 2-99
DEFINE procedure 4-1 ZFIL-VAL-AWAITIOX 2-129
EDITREAD procedure 15-1 ZFIL-VAL-CHECKCLOSE 2-100
EDITREADINIT procedure 15-1 ZFIL-VAL-CHECKMONITOR 2-101
file system 2-5/2-95 ZFIL-VAL-CHECKOPEN 2-103
C-series and D-series handling 2-2 ZFIL-VAL-CHECKPOINT 2-104
errno, corresponding values of 2-3 ZFIL-VAL-
guidelines for using 2-1 CHECKPOINTMANY 2-106
used by IOEdit 16-2/16-4 ZFIL-VAL-
FORMATDATA[X] procedure 17-1 CHECKPOINTMANYX 2-130

Index -1
Index E
Error lists (continued) Error lists (continued)

ZFIL-VAL-CHECKPOINTX 2-132 ZFIL-VAL-
ZFIL-VAL-CHECKSWITCH 2-108 READUPDATELOCK 2-119
ZFIL-VAL-CLOSE 2-145 ZFIL-VAL-
READUPDATELOCKX 2-136
ZFIL-VAL-CONTROL 2-109
ZFIL-VAL-READUPDATEX 2-137
ZFIL-VAL-CREATE 2-111
ZFIL-VAL-READX 2-138
ZFIL-VAL-DEVICEINFO2 2-146
ZFIL-VAL-REPLY 2-120
ZFIL-VAL-DISK-REFRESH 2-153
ZFIL-VAL-REPLYX 2-139
ZFIL-VAL-FILENAME-
COMPARE 2-152 ZFIL-VAL-SETMODE 2-121
ZFIL-VAL-FILENAME- ZFIL-VAL-
FINDFINISH 2-156 SETMODENOWAIT 2-122
ZFIL-VAL-FILENAME- ZFIL-VAL-WRITE 2-124
FINDNEXT 2-155 ZFIL-VAL-WRITEREAD 2-125
ZFIL-VAL-FILENAME- ZFIL-VAL-WRITEREADX 2-140
FINDSTART 2-155 ZFIL-VAL-WRITEUPDATE 2-127
ZFIL-VAL-FILE-CLOSE 2-150 ZFIL-VAL-
ZFIL-VAL-FILE-CREATE 2-157 WRITEUPDATEUNLOCK 2-128
ZFIL-VAL-FILE- ZFIL-VAL-
CREATELIST 2-147 WRITEUPDATEUNLOCKX 2-141
ZFIL-VAL-FILE- ZFIL-VAL-WRITEUPDATEX 2-143
GETINFOBYNAME 2-150 ZFIL-VAL-WRITEX 2-144
ZFIL-VAL-FILE- NEWPROCESS procedure 5-21
GETOPENINFO 2-152 guidelines for using 5-20
ZFIL-VAL-FILE- NEWPROCESSNOWAIT
GETRECEIVEINFO 2-151 procedure 5-22
ZFIL-VAL-FILE-OPEN 2-148 guidelines for using 5-20
ZFIL-VAL-FILE-OPEN- PROCESS_CREATE_ procedure 6-37
CHKPT 2-147
ZFIL-VAL-FILE-PURGE 2-149
PROCESS_LAUNCH_ procedure 6-39
ZFIL-VAL-FILE-RENAME 2-154
ZFIL-VAL-KEYPOSITION 2-112
PROCESS_SPAWN_ procedure 9-7
ZFIL-VAL-KEYPOSITIONX 2-133
ZFIL-VAL-OPEN 2-113
SEGMENT_ALLOCATE_
ZFIL-VAL-POSITION 2-115 procedure 11-7
ZFIL-VAL-PURGE 2-114 guidelines for using 11-6
ZFIL-VAL-READ 2-116 SEGMENT_USE_ procedure 13-2
ZFIL-VAL-READLOCK 2-117 guidelines for using 13-1
ZFIL-VAL-READLOCKX 2-134 signal error list 21-10
ZFIL-VAL-READUPDATE 2-118 guidelines for using 21-9

Index -2
Index F
Error lists (continued)

SSGET procedure 14-10
F
File system
SSGETTKN procedure 14-10
error codes 2-5/2-25
SSINIT procedure 14-11
C-series and D-series handling 2-2
SSMOVE procedure 14-12
errno, corresponding values of 2-3
SSMOVETKN procedure 14-12
SSNULL procedure 14-13
used by IOEdit 16-2/16-4
SSPUT procedure 14-14
error lists 2-96/2-157
SSPUTTKN procedure 14-14
conditional tokens 2-96
Subsystem Programmatic Interface
(SPI) guidelines for using 2-96
general description of 14-9 ZFIL-VAL-AWAITIO 2-99
guidelines for using 14-9 ZFIL-VAL-AWAITIOX 2-129
SSGET 14-10 ZFIL-VAL-CHECKCLOSE 2-100
SSGETTKN 14-10 ZFIL-VAL-CHECKMONITOR 2-101
SSINIT 14-11 ZFIL-VAL-CHECKOPEN 2-103
SSMOVE 14-12 ZFIL-VAL-CHECKPOINT 2-104
SSMOVETKN 14-12 ZFIL-VAL-
CHECKPOINTMANY 2-106
SSNULL 14-13
ZFIL-VAL-
SSPUT 14-14
CHECKPOINTMANYX 2-130
SSPUTTKN 14-14
ZFIL-VAL-CHECKPOINTX 2-132
trap error list 21-10
ZFIL-VAL-CHECKSWITCH 2-108
ZFIL-VAL-CLOSE 2-145
USESEGMENT procedure 12-2
ZFIL-VAL-CONTROL 2-109
ZFIL-VAL-CREATE 2-111
ZGRD-VAL-
ZFIL-VAL-DEVICEINFO2 2-146
ALLOCATESEGMENT 10-5
ZFIL-VAL-DISK-REFRESH 2-153
ZGRD-VAL-NEWPROCESS 5-21
ZFIL-VAL-FILENAME-
ZGRD-VAL-
COMPARE 2-152
NEWPROCESSNOWAIT 5-22
ZFIL-VAL-FILENAME-
ZGRD-VAL-PROCESS-CREATE 6-37
FINDFINISH 2-156
ZGRD-VAL-PROCESS-LAUNCH 6-39
ZFIL-VAL-FILENAME-
ZGRD-VAL-PROCESS-SPAWN 9-7 FINDNEXT 2-155
ZGRD-VAL-SEGMENT- ZFIL-VAL-FILENAME-
ALLOCATE 11-7 FINDSTART 2-155
ZGRD-VAL-SEGMENT-USE 13-2 ZFIL-VAL-FILE-CLOSE 2-150
ZGRD-VAL-TRAPOUT 21-10 ZFIL-VAL-FILE-CREATE 2-157
ZGRD-VAL-USESEGMENT 12-2 ZFIL-VAL-FILE-
CREATELIST 2-147

Index -3
Index G
File system (continued) File system (continued)

ZFIL-VAL-FILE- ZFIL-VAL-WRITEX 2-144
GETINFOBYNAME 2-150 OSS error code information 22-1
ZFIL-VAL-FILE- FORMATDATA[X] procedure error
GETOPENINFO 2-152 codes 17-1
ZFIL-VAL-FILE-
GETRECEIVEINFO 2-151
ZFIL-VAL-FILE-OPEN 2-148
G
G-series procedure errors 1-1
ZFIL-VAL-FILE-OPEN-
CHKPT 2-147
ZFIL-VAL-FILE-PURGE 2-149 H
ZFIL-VAL-FILE-RENAME 2-154 H-series procedure errors 1-1
ZFIL-VAL-KEYPOSITION 2-112
ZFIL-VAL-KEYPOSITIONX 2-133 I
ZFIL-VAL-OPEN 2-113 INITIALIZER procedure error
ZFIL-VAL-POSITION 2-115 codes 18-1/18-4
ZFIL-VAL-PURGE 2-114 Interprocess command interpreter
ZFIL-VAL-READ 2-116 messages 1-3, 19-3/19-20
ZFIL-VAL-READLOCK 2-117 description of 19-1
ZFIL-VAL-READLOCKX 2-134 Interprocess messages, general description
of 1-3
ZFIL-VAL-READUPDATE 2-118
Interprocess system messages
ZFIL-VAL-
See System messages
READUPDATELOCK 2-119
IOEdit procedures
ZFIL-VAL-
READUPDATELOCKX 2-136 catastrophic error messages 16-8/16-9
ZFIL-VAL-READUPDATEX 2-137 file-system error codes used
by 16-2/16-4
ZFIL-VAL-READX 2-138
unique error codes used by 16-5/16-7
ZFIL-VAL-REPLY 2-120
ZFIL-VAL-REPLYX 2-139
ZFIL-VAL-SETMODE 2-121 M
ZFIL-VAL- Messages
SETMODENOWAIT 2-122 command interpreter 19-3/19-20
ZFIL-VAL-WRITE 2-124 interprocess, general description of 1-3
ZFIL-VAL-WRITEREAD 2-125 system
ZFIL-VAL-WRITEREADX 2-140 See System messages
ZFIL-VAL-WRITEUPDATE 2-127
ZFIL-VAL- N
WRITEUPDATEUNLOCK 2-128
NEWPROCESS procedure
ZFIL-VAL-
error codes 5-1/5-19
WRITEUPDATEUNLOCKX 2-141
error list 5-21
ZFIL-VAL-WRITEUPDATEX 2-143

Index -4
Index O
NEWPROCESS procedure (continued) Procedures (continued)

guidelines for using 5-20 PROCESS_CREATE_
NEWPROCESSNOWAIT procedure error codes 6-1/6-35
error codes 5-1/5-19 error list 6-37
error list 5-22 PROCESS_GETINFOLIST_ error
guidelines for using 5-20 codes 7-1
PROCESS_GETPAIRINFO_ error
codes 8-1
O PROCESS_LAUNCH_
OSS error codes 22-1 error codes 6-1/6-35
PROCESS_SPAWN_
error list 6-39
procedure 9-2/9-5
PROCESS_SPAWN_
error codes 6-1, 9-1, 9-2
P error list 9-7
Procedure errors 1-1 Guardian error codes 6-1, 9-1
C-series 1-2 OSS error codes 9-2/9-5
D-series 1-1 SEGMENT_ALLOCATE_
Procedures error codes 11-1
ALLOCATESEGMENT error list 11-7
error codes 10-1 SEGMENT_USE_ error list 13-2
error list 10-5 sequential I/O (SIO) error
DEFINE error codes 4-1 codes 3-1/3-7
EDITREAD error codes 15-1 Subsystem Programmatic Interface
EDITREADINIT error codes 15-1 (SPI)
FORMATDATA[X] error codes 17-1 error codes 14-1/14-8
INITIALIZER error codes 18-1/18-4 error lists 14-9/14-14
IOEdit USESEGMENT error list 12-2
catastrophic error PROCESS_CREATE_ procedure
messages 16-8/16-9 error codes 6-1/6-41
file-system error codes used error list 6-37
by 16-2/16-4 guidelines for using 6-36
unique error codes used PROCESS_GETINFOLIST_ procedure
by 16-5/16-7 error codes 7-1
NEWPROCESS PROCESS_GETPAIRINFO_ procedure
error codes 5-1/5-19 error codes 8-1
error list 5-21 PROCESS_LAUNCH_ procedure
NEWPROCESSNOWAIT error codes 6-1/6-41
error codes 5-1/5-19 error list 6-39
error list 5-22 guidelines for using 6-36

Index -5
Index S
PROCESS_SPAWN_ procedure SIGNOMEM signal 21-7

error codes 6-1, 9-1, 9-2 SIGSEGV signal 21-2
error list 9-7 SIGSTK signal 21-5
guidelines for using 9-6 SIGTIMEOUT signal 21-6
Guardian error codes 6-1, 9-1 SIO
OSS error codes 9-2/9-5 See Sequential I/O (SIO) procedures
SPI
See Subsystem Programmatic Interface
S (SPI)
SEGMENT_ALLOCATE_ procedure SSGET procedure error list 14-10
error codes 11-1 SSGETTKN procedure error list 14-10
error list 11-7 SSINIT procedure error list 14-11
guidelines for using 11-6 SSMOVE procedure error list 14-12
SEGMENT_USE_ procedure error list 13-2 SSMOVETKN procedure error list 14-12
guidelines for using 13-1 SSNULL procedure error list 14-13
Sequential I/O (SIO) procedure error SSPUT procedure error list 14-14
codes 3-1/3-7 SSPUTTKN procedure error list 14-14
SIGABRT signal 21-8 Subsystem Programmatic Interface (SPI)
SIGALRM signal 21-8 error codes 14-1/14-8
SIGFPE signal 21-4 error lists 14-9/14-14
SIGILL signal 21-3 general description of 14-9
SIGLIMIT signal 21-6 guidelines for using 14-9
SIGMEMERR signal 21-8 SSGET 14-10
SIGMEMMGR signal 21-7
SSGETTKN 14-10
Signals 1-4, 21-1
SSINIT 14-11
and corresponding traps 21-1
SSMOVE 14-12
error list 21-10
SSMOVETKN 14-12
SSNULL 14-13
SIGABRT 21-8
SSPUT 14-14
SIGALRM 21-8
SSPUTTKN 14-14
SIGFPE 21-4
System messages 1-3, 20-4/20-37
SIGILL 21-3
C-series compared with D-series 20-3
SIGLIMIT 21-6
description of 20-1
SIGMEMERR 21-8
D-series 20-2
SIGMEMMGR 21-7
signal handling 21-2
signal names and numbers 21-1
T
SIGNOMEM 21-7 Traps 1-4, 21-1
SIGSEGV 21-2 and corresponding signals 21-1
SIGSTK 21-5 each described 21-2/21-8
SIGTIMEOUT 21-6 error list 21-10

Index -6
Index U
Traps (continued) ZFIL-VAL- (continued)

guidelines for using 21-9 FILE-RENAME 2-154
trap handling 21-2 KEYPOSITION 2-112
KEYPOSITIONX 2-133
U OPEN 2-113
POSITION 2-115
USESEGMENT procedure error list 12-2
guidelines for using 12-1 PURGE 2-114
READ 2-116
READLOCK 2-117
Z READLOCKX 2-134
ZFIL-VAL- READUPDATE 2-118
AWAITIO 2-99
READUPDATELOCK 2-119
AWAITIOX 2-129
READUPDATELOCKX 2-136
CHECKCLOSE 2-100
READUPDATEX 2-137
CHECKMONITOR 2-101
READX 2-138
CHECKOPEN 2-103
REPLY 2-120
CHECKPOINT 2-104
REPLYX 2-139
CHECKPOINTMANY 2-106
SETMODE 2-121
CHECKPOINTMANYX 2-130
SETMODENOWAIT 2-122
CHECKPOINTX 2-132
WRITE 2-124
CHECKSWITCH 2-108
WRITEREAD 2-125
CLOSE 2-145
WRITEREADX 2-140
CONTROL 2-109
WRITEUPDATE 2-127
CREATE 2-111
WRITEUPDATEUNLOCK 2-128
DEVICEINFO2 2-146
WRITEUPDATEUNLOCKX 2-141
DISK-REFRESH 2-153
WRITEUPDATEX 2-143
FILENAME-COMPARE 2-152
WRITEX 2-144
FILENAME-FINDFINISH 2-156
ZGRD-VAL-
FILENAME-FINDNEXT 2-155 ALLOCATESEGMENT 10-5
FILENAME-FINDSTART 2-155 NEWPROCESS 5-21
FILE-CLOSE 2-150 NEWPROCESSNOWAIT 5-22
FILE-CREATE 2-157 PROCESS-CREATE 6-37
FILE-CREATELIST 2-147 PROCESS-LAUNCH 6-39
FILE-GETINFOBYNAME 2-150 PROCESS-SPAWN 9-7
FILE-GETOPENINFO 2-152 SEGMENT-ALLOCATE 11-7
FILE-GETRECEIVEINFO 2-151 SEGMENT-USE 13-2
FILE-OPEN 2-148 TRAPOUT 21-10
FILE-OPEN-CHKPT 2-147 USESEGMENT 12-2
FILE-PURGE 2-149

Index -7
Index Z

Index -8

Guardian Procedure Errors and Messages Manual

Uploaded by

Copyright:

Available Formats

Guardian Procedure Errors and Messages Manual

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Guardian Procedure Errors and Messages Manual

Uploaded by

Copyright:

Available Formats

Guardian Procedure

Errors and Messages

Part Number Published

Index Figures Tables

2. File-System Errors (continued)

Guardian Procedure Errors and Messages Manual—522628-011

2. File-System Errors (continued)

3. Sequential I/O Errors

5. NEWPROCESS AND NEWPROCESSNOWAIT Errors

6. Process Creation Errors

Guardian Procedure Errors and Messages Manual—522628-011

9. PROCESS_SPAWN_ Open System Services (OSS) Errors

10. ALLOCATESEGMENT Errors

11. SEGMENT_ALLOCATE_ Errors

12. USESEGMENT Errors

13. SEGMENT_USE_ Errors

14. Subsystem Programmatic Interface (SPI) Errors

15. EDITREAD and EDITREADINIT Errors

16. IOEdit Errors

Guardian Procedure Errors and Messages Manual—522628-011

17. Formatter Errors

18. INITIALIZER Errors

19. Interprocess Command Interpreter Messages

20. System Messages

21. Traps and Signals

22. OSS Error Information

Guardian Procedure Errors and Messages Manual—522628-011

Guardian Procedure Errors and Messages Manual—522628-011

Part Number Published

New and Changed Information

Guardian Procedure Errors and Messages Manual—522628-011

Changes to the H06.21/J06.10 Manual

Changes to the H06.20/J06.09 Manual

° Errors 700 on page 2-76 to 711 on page 2-79.

° Error 734 on page 2-83.

° Error 735 on page 2-84.

° Error 899 on page 2-86.

Changes to the H06.14/J06.03 Manual

Changes to the H06.11 Manual

° 3502 (%6656) FEREQUESTALLOCATIONFAILURE on page 2-87

° 538 SIOERR^EXTSIZE^OVERFLOW: Extent size is greater than 65535 pages

° 539 SIOERR^PAGEWRITE^OVERFLOW: The highest possible page in EDIT

° 2081 (%004041) Bad internal format. on page 4-9

Guardian Procedure Errors and Messages Manual—522628-011

Who Should Read This Manual

How This Manual Is Organized

Guardian Procedure Errors and Messages Manual—522628-011

the error list returned by the PROCESS_SPAWN_ procedure are listed in

Guardian Procedure Errors and Messages Manual—522628-011

Guardian Procedure Errors and Messages Manual—522628-011

[ ] Brackets. Brackets enclose optional syntax items. For example:

Punctuation. Parentheses, commas, semicolons, and other symbols not previously

Guardian Procedure Errors and Messages Manual—522628-011

Notation for Messages

Guardian Procedure Errors and Messages Manual—522628-011

Notation for Management Programming Interfaces

Change Bar Notation

Guardian Procedure Errors and Messages Manual—522628-011

HP Encourages Your Comments

Guardian Procedure Errors and Messages Manual—522628-011