File Utility Program - FUP Reference Manual - 321 Pages
File Utility Program - FUP Reference Manual - 321 Pages
(FUP) Reference
Manual
Abstract
This manual describes the command syntax and error messages for the File Utility
Program (FUP).
Product Version
T6553 D45, G08, and H01
Supported Release Version Updates (RVUs)
This publication supports J06.03 and all subsequent J-series RVUs, H06.03 and all
subsequent H-series RVUs, G06.23 and all subsequent G-series RVUs, and D46.00
and all subsequent D-series RVUs, until otherwise indicated by its replacement
publications.
Legal Notices
What’s New in This Manual ix
Manual Information ix
New and Changed Information ix
About This Manual xiii
Organization of This Manual xiii
Additional Information xiv
Notation Conventions xiv
1. FUP Overview
Starting a FUP Process 1-2
At the TACL Prompt 1-2
Interactively 1-3
From a Command File 1-3
Command Files 1-3
Run Options 1-4
Using FUP Custom Files 1-4
FUP Custom File Guidelines 1-4
FUP Custom File Example 1-5
Interrupting or Terminating a FUP Process 1-5
Entering a FUP Command 1-5
FUP Command Guideline 1-7
FUP Command Examples 1-7
Specifying Files 1-8
Listfile Parameter 1-8
Fileset-list Parameter 1-9
Fileset Parameter 1-9
Wild-Card Option 1-10
Qualified File Sets 1-11
Creating Files 1-19
Examples of Creating Files 1-19
Hewlett-Packard Company—523323-012
i
Contents 1. FUP Overview (continued)
2. FUP Commands
! 2-4
! Guidelines 2-4
! Examples 2-4
Commands Related to ! 2-5
? 2-5
? Guidelines 2-5
Commands Related to ? 2-6
ALLOCATE 2-6
ALLOCATE Guidelines 2-6
ALLOCATE Examples 2-7
Commands Related to ALLOCATE 2-8
ALLOW 2-9
ALTER 2-9
ALTER Parameters for All File Types 2-10
ALTER Parameters for Files With Alternate-Key Fields 2-13
ALTER Parameters for Partitioned Files 2-15
ALTER Parameter for Unstructured Files 2-18
ALTER Guidelines 2-18
ALTER Examples 2-20
Commands Related to ALTER 2-20
BUILDKEYRECORDS 2-21
BUILDKEYRECORDS Guidelines 2-22
BUILDKEYRECORDS Example 2-23
3. FUP Messages
A. DEFINE Tables
B. FUP Command Summary
C. FUP Command Syntax Summary
Glossary
Index
Examples
Example 2-1. COPY Command Listing Format 2-56
Example 2-2. INFO Listing Format 2-85
Example 2-3. INFO Listing Format 2-97
Example 2-4. Short INFO for SQL/MX Table Using ANSI Names 2-97
Example 2-5. Short INFO for OSS Files With POSIX ACL 2-98
Example 2-6. DETAIL Format for SQL Tables and Indexes and for Enscribe and
OSS Files 2-99
Example 2-7. DETAIL Format for SQL/MP View 2-104
Example 2-8. INFO STATISTICS Listing Format 2-112
Example 2-9. INFO EXTENTS Listing Format 2-115
Example 2-10. FUP LISTLOCKS DETAIL Listing Format 2-120
Example 2-11. FUP LISTOPENS Listing Format 2-125
Tables
Table 1-1. File Format Codes 1-23
Table 1-2. FUP Commands and SQL/MP Files 1-25
Table 1-3. FUP Commands and SQL/MX Files 1-27
Table 2-1. Response to ALLOW ABENDS ON or OFF 2-28
Table 2-2. System File Code Definitions 2-88
Tables (continued)
Table 2-3. Extent-Size Rounding 2-179
Table A-1. How FUP Input Options Work With TAPE DEFINEs A-1
Table A-2. How FUP Output Options Work With TAPE DEFINEs A-2
Table B-1. FUP Control Commands B-1
Table B-2. FUP Informational Commands B-2
Table B-3. FUP Security Management Commands B-2
Table B-4. FUP File Management Commands B-2
Abstract
This manual describes the command syntax and error messages for the File Utility
Program (FUP).
Product Version
T6553 D45, G08, and H01
Supported Release Version Updates (RVUs)
This publication supports J06.03 and all subsequent J-series RVUs, H06.03 and all
subsequent H-series RVUs, G06.23 and all subsequent G-series RVUs, and D46.00
and all subsequent D-series RVUs, until otherwise indicated by its replacement
publications.
Document History
Part Number Product Version Published
523323-004 FUP D45 and G08 April 2004
523323-007 FUP D45, G08, and H01 July 2005
523323-009 FUP D45, G08, and H01 July 2006
523323-010 FUP D45, G08, and H01 November 2006
523323-012 FUP D45, G08, and H01 October 2009
• Added the TYPE variable and its description to the key column example on page
2-101.
• Added a note on page 2-101 that the key column TYPE and LENGHT is not
displayed for SQL/MX objects .
• Updated these commands to document the FUP support for SQL/MX ANSI names:
° TRUSTED
° TRUSTME
° TRUSTSHARED
• Added a general guideline to the DUP[LICATE] command in FUP commands.
• Added command names to guidelines and examples.
• Added the TRUST variable to Example 2-6, DETAIL Format for SQL Tables and
Indexes and for Enscribe and OSS Files, on page 2-99.
Additional Information
For more information about FUP and its associated components, see:
Manual Description
5200 Optical Storage Facility Describes the FUP commands that require special
(OSF) Reference Manual considerations when used with the 5200 OSF (for D-
series RVUs only)
Enscribe Programmer’s Guide Provides descriptions of structured disk files and the file
attributes specified in FUP commands.
File Utility Program (FUP) Explains how to execute commands from within an
Management Programming application program by using the Subsystem
Manual Programmatic Interface (SPI).
Guardian Disk and Tape Explains tape handling procedures and describes the
Utilities Reference Manual BACKCOPY, BACKUP, DCOM, DSAP, RESTORE, and
TAPECOM utilities.
Guardian Procedure Errors Describes any file-system (or other) errors referenced by
and Messages Manual number in the FUP error messages.
Guardian User’s Guide Provides task-oriented instructions for using FUP and
DEFINEs. It also includes a basic FUP introduction.
SQL/MP Reference Manual Describes the SQLCI commands used on SQL files.
Because this function is similar to the FUP process, this
manual also describes using FUP at the SQLCI prompt.
SQL/MX Reference Manual Describes the syntax of SQL language elements.
Safeguard User’s Guide Describes the basic security tasks that FUP requires.
TACL Reference Manual Describes the RUN command options and all the other
TACL commands and functions.
SQL/MX Glossary Describes terms specific to SQL/MX.
Notation Conventions
Hypertext Links
Blue underline is used to indicate a hypertext link within text. By clicking a passage of
text with a blue underline, you are taken to the location described. For example:
This requirement is described under Backup DAM Volumes and Physical Disk
Drives on page 3-2.
UPPERCASE LETTERS. Uppercase letters indicate keywords and reserved words. Type
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
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
{ } 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 can 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 [ , 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…"
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 this example, no
spaces are 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 / ] LINE
[ , attribute-spec ]…
Bold Text. Bold text in an example indicates user input typed at the terminal. For example:
ENTER RUN CODE
?123
CODE RECEIVED: 123.00
The user must press the Return key after typing the input.
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 can
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:
proc-name trapped [ in SQL | in SQL file system ]
{ } 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 can 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:
obj-type obj-name state changed to state, caused by
{ Object | Operator | Service }
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
% notation precedes an octal number. The %B notation precedes a binary number.
The %H notation precedes a hexadecimal number. For example:
%005400
%B101111
%H2F
P=%p-register E=%e-register
UPPERCASE LETTERS. Uppercase letters indicate names from definition files. Type 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
!r. The !r notation following a token or field name indicates that the token or field is
required. For example:
ZCOM-TKN-OBJNAME token-type ZSPI-TYP-STRING. !r
!o. The !o notation following a token or field name indicates that the token or field is
optional. For example:
ZSPI-TKN-MANAGER token-type ZSPI-TYP-FNAME32. !o
Note. FUP does not support mixing of both SQL/MX objects (using their ANSI names) and
Guardian objects in the same command. Therefore, the user must use two separate
commands, one for Guardian names and another for ANSI names.
Note. For more information about Enscribe files and structured query language (SQL) files,
see the Enscribe Programmer’s Guide, the SQL/MP Reference Manual, the SQL/MX
Reference Manual , and the Guardian User’s Guide .
Topic Page
Starting a FUP Process 1-2
Using FUP Custom Files 1-4
Interrupting or Terminating a FUP Process 1-5
Entering a FUP Command 1-5
Specifying Files 1-8
Using DEFINEs With FUP 1-21
Handling Different Types of Files 1-22
Note. For information about entering FUP commands, see Entering a FUP Command on
page 1-5. For descriptions of each FUP command and its corresponding syntax, see Section 2,
FUP Commands. For a description of command files and run options, see From a Command
File on page 1-3.
The FUP command (INFO *) in this example instructs FUP to list the file information of
each file in your current subvolume. After FUP executes the command, control of the
terminal returns to TACL. A separate FUP process starts and completes for each
command you enter.
Interactively
Entering FUP commands within FUP (interactively) saves time if you are going to enter
a series of commands. Type the term FUP (without any commands or options) at the
TACL prompt, and press the RETURN key to start an interactive FUP process:
1> FUP
File Utility Program - T6553D45 - (13OCT2000) SYSTEM \WEST
Copyright Tandem Computers Incorporated 1981, 1983, 1985-2000
-
The FUP process is ready to receive commands interactively when its sign-on banner
and prompt (a hyphen) are first displayed. You can then type a FUP command at each
subsequent FUP prompt:
-INFO *
After FUP executes the command, the FUP prompt reappears. Type another FUP
command at the subsequent prompt, or use the EXIT command to return control of the
terminal from FUP to TACL:
-EXIT
Command Files
To create a command file that contains FUP commands, use a text editor (such as
TEDIT). This example shows a command file (ALLSUBS):
-- FUP commands for obtaining a list of all
-- subvolumes in $DISK1, $DISK2, and $DISK3
--
-- Last modified 5/17/01
--
SUBVOLS $DISK1 -- Contains manufacturing files
SUBVOLS $DISK2 -- Contains administrative files
SUBVOLS $DISK3 -- Contains all other files
The example command file (ALLSUBS) uses a FUP command (SUBVOLS) to list the
subvolumes in three different disk volumes ($DISK1, $DISK2, and $DISK3). It also
includes comment lines to help identify the file and explain the operations that are
performed.
Note. You can enter two dashes (--) or a less-than symbol (<) to indicate a comment. FUP
ignores any text that follows these punctuation marks until the end of the command line.
Run Options
The TACL environment includes a set of predefined commands—including the TACL
RUN command. You must use a run option when you start a FUP process with a
command file. The two run options that FUP uses most often are IN (for specifying an
input file) and OUT (for an output file).
Note. For more information about the TACL RUN command, see the TACL Reference Manual.
Type the term FUP and the command file and run option you want, and press
RETURN:
In this example, the FUP process starts when you use a command file (ALLSUBS) and
run option (IN). The command in this example writes to the terminal because there is
no OUT file. After FUP executes the last command in the command file, control of the
terminal returns to TACL.
Caution. For software product revisions (SPRs) earlier than T6553ABQ, FUP can cause a
processor halt if it receives a bad startup message. This situation occurs mainly when FUP is
started programmatically and the FUP STARTUPMESSAGE is accidentally corrupted, or when
FUP is started through SCF after the “DEFAULT STARTUP MESSAGE” gets corrupted when a
SYSTEM \system-name is executed. To avoid this situation, install T6553ABQ or a later
SPR.
• By default, FUP does not echo the commands for either file. To start echoing,
specify CONFIGURE ECHO OBEY in either file.
• If there are no existing custom files when FUP begins, default custom files are
automatically created for your security.
• Custom files must be EDIT files.
• Errors encountered during the processing of a custom file could cause the FUP
session to terminate.
run-options
are any of the available options for the TACL RUN command. They must be
separated from each other by commas, and enclosed on the command line with
slashes (/). Although each option is available when you run FUP from the TACL
prompt, only one (the OUT run option) is also available within FUP. These run
options are used most often with the FUP process:
IN filename
names a disk file, nondisk device, or process from which FUP reads
commands. Any IN disk file that you specify must be an EDIT file, unstructured
file (132 byte records), or Enscribe-structured file. You cannot specify an SQL
file as an IN disk file.
If you omit this option, FUP uses the IN file name that is in effect for the current
TACL process— usually the home terminal.
OUT listfile
names a nondisk device, process, or a disk file to which FUP directs its listing
output (unless you use the OUT option of a subsequent FUP command to
direct the output somewhere else). You must use a Guardian file name, a
SPOOL DEFINE, or MAP DEFINE as the OUT list file to run FUP. Any
specified OUT disk file must be an EDIT file, unstructured file, or Enscribe-
structured file. You cannot specify an SQL file as an OUT disk file. You can use
this option to specify output to a TACL variable.
If you omit this option, FUP uses the OUT list file that is in effect for the current
TACL process—which is usually the home terminal.
If listfile does not exist, FUP creates it as an EDIT file with a maximum
record length of 132 characters.
If listfile does exist, FUP appends output to it.
If listfile is an unstructured disk file, each of its records is 132 characters
long, and any partial lines are blank-filled to column 132.
For a description of using a SPOOL DEFINE for listfile, see Using
DEFINEs With FUP on page 1-21.
command
is a FUP command. You can enter only one FUP command in a TACL command. If
you need additional screen space to enter the command, end the command line
with an ampersand (&) and press RETURN. Continue the command on the next
line at the new TACL prompt.
If you type FUP followed by command, the FUP process terminates after executing
the command and returns control of the terminal to TACL.
If you type FUP without command, you must terminate the FUP process using the
EXIT command after you finish your FUP activity.
While the command is executing, FUP displays any information about the ongoing
process and then returns you to the FUP prompt (hyphen) when the process is
complete.
• To avoid the interactive FUP process shown in the previous example for the DUP
command (and the PURGE option) at the TACL prompt, and purge FILE2 before it
duplicates the contents of FILE1 to FILE2:
TACL1> FUP DUP FILE1, FILE2, PURGE
Specifying Files
FUP commands make it easy to:
• Create, display, and duplicate files
• Load data into files
• Alter file characteristics
• Purge files
Before you use FUP to create or manage files, become familiar with the various file
types and the methods used to specify them.
Note. For more information on the different types of Enscribe-structured files, see the Enscribe
Programmer’s Guide.
The different types of files in the Guardian environment include disk files (containing
data, code, or text), nondisk devices (terminals, printers, or tape drives), spooler files
(code 129), and processes (programs that are running). Disk file names have four
parts: node, volume, subvolume, and file identifier. The names of nondisk devices and
processes must begin with a dollar sign ($), followed by one to five alphanumeric
characters (and additional qualifiers if applicable).
Note. For more information on file-naming conventions, see the Guardian User’s Guide .
Three common syntax terms appear throughout the FUP commands. Each of these
terms provides a way to specify the file or files you want to affect with a FUP
command. The syntax terms are:
• listfile
• fileset-list
• fileset
Note. For information on when FUP commands do not apply to SQL files, see the command
descriptions in Section 2, FUP Commands.
Listfile Parameter
The listfile parameter refers to a nondisk device, a process, an existing disk file,
or a spooler file (code 129) to which you direct the output of a FUP command. It always
appears with the OUT keyword in the syntax descriptions for the TACL RUN command
and in the FILENAMES, FILES, HELP, INFO, LISTLOCKS, LISTOPENS, RELOAD,
SHOW, STATUS, SUBVOLS, SUSPEND, and VOLS commands for FUP.
If you omit this option, FUP uses the OUT list file that is in effect for the current TACL
process—usually the home terminal.
If a list file does not exist, FUP creates it as an EDIT file with a maximum record length
of 132 characters.
If a list file does exist, FUP appends output to it.
If the list file is an unstructured disk file, each of its records is 132 characters long, and
any partial lines are blank-filled to column 132.
You can specify a SPOOL DEFINE for listfile. For a description of using a SPOOL
DEFINE, see Using DEFINEs With FUP on page 1-21.
If you enter only a partial file name, FUP expands it using the current default node,
volume, and subvolume names.
Fileset-list Parameter
The fileset-list parameter refers to one or more file sets. It can contain up to ten
different sets of files enclosed in parentheses. The form for fileset-list is:
fileset is:
[[[ \node.] $volume. ] subvolume. ] file-id [ qualified-expr
]
Fileset-list Examples
• To specify the files MYFILE, MYSRC, and MYOBJ in the current default subvolume
on the current default volume:
(MYFILE, MYSRC, MYOBJ)
• To specify all files in the current default subvolume and all files in all the
subvolumes on the volume $VOL1:
(*, $VOL1.*.*)
• To specify all files in all subvolumes on the current default volume and all files on
the volume $VOL2:
(*.*, $VOL2.*.*)
Fileset Parameter
The fileset parameter refers to a set of files. It can be one file, all the files in a
subvolume, all the files on a volume, a subset specified by the wild-card option, or a
subset produced by a qualified file set. The file or files can be fully qualified or can
include wild cards or qualifiers. The form for fileset is:
node
is the name of a node in an Expand network. If you omit node, FUP uses the
current default node: either the node that was in effect when you started FUP or
the node you specified during the last FUP SYSTEM or VOLUME command.
volume
is the name of a volume (disk drive). If you omit volume, FUP uses the current
default volume: either the volume that was in effect when you started FUP or the
volume you specified during the last FUP VOLUME (or SYSTEM) command. If you
specify volume, you must also specify subvolume.
subvolume
is the name of the subvolume. It consists of alphanumeric characters, wild-card
characters (* or ?*), or a combination of both. If you omit subvolume, FUP uses
the current default subvolume: either the subvolume that was in effect when you
started FUP or the subvolume you specified during the last FUP VOLUME (or
SYSTEM) command.
For compatibility with previous versions of FUP, the subvolume name can be the
default in most commands. If the subvolume is omitted, FUP uses the current
default subvolume and issues a warning that this option might be deleted in future
versions of FUP.
file-id
is the name of an Enscribe or SQL disk file. It consists of alphanumeric characters,
wild-card characters (* or ?), or a combination of both. Table 1-2 on page 1-25
shows the interaction between FUP commands and SQL files.
qualified-expr
specifies additional restrictions to a file set. It extends the power of the wild-card
option in file names by specifying attributes of the file. A file set with
qualified-expr is called a qualified file set. For more information, see Qualified
File Sets on page 1-11.
Wild-Card Option
You can use asterisks (*) or question marks (?) as wild-card characters to help specify
files. The asterisk can represent from zero through eight unspecified characters in the
position you place it. The question mark represents only one character in the position
you place it.
Wild-Card Guidelines
• The only FUP commands that allow you to use wild-card characters to specify a
volume name are VOLS, SUBVOLS, INFO, FILENAMES, and FILES.
• You cannot use more than eight characters (including wild-card characters) in any
portion of the file name (volume, subvolume, or file identifier).
• The only valid use of the wild-card option in a destination file-set specification is a
single asterisk (*) in the subvolume or file ID position. The DUP MYFILE, *
command is valid, but the DUP MYFILE, MY* command is not.
Wild-Card Examples
• To specify the file MYFILE on the current FUP default volume and subvolume:
MYFILE
• To specify all the file IDs that begin with MY on the current default volume and
subvolume:
MY*
• To specify all the file IDs that begin with MY, are followed by one character, and
end with ILE on the current default volume and subvolume:
MY?ILE
• To specify all the files in the subvolume MYSVOL on the current default volume:
MYSVOL.*
• To specify all the files in all the subvolumes on the volume $VOL1:
$VOL1.*.*
• To specify all the files in all the subvolumes on the current default volume:
*.*
• To specify all the files in the subvolume MYTOWN on the volume $BRDWAY on
the \NY node:
\NY.$BRDWAY.MYTOWN.*
Note. File sets can use qualified file-set syntax except where explicitly prohibited in the
descriptions of each command in Section 2, FUP Commands.
qualifier is:
EXCLUDE fileset
FROM CATALOG[S] catalog-list
START file-id
WHERE expression
catalog-list is:
expression is:
( expression )
expression AND expression
expression OR expression
NOT expression
file-attribute
OWNER = user-id
timestamp-field time-conditional time-value
FILECODE conditional-number
EOF conditional-number
file-attribute is:
ALTKEY
AUDITED
BROKEN
COLLATION
CORRUPT
CRASHOPEN
ENSCRIBE
ENTRYSEQUENCED
FORMAT1|FORMAT2
INDEX
[ SHORTHAND | PROTECTION ] VIEW
KEYSEQUENCED
LICENSED
OPEN
PROGID
RELATIVE
[ PRIMARY | SECONDARY ] PARTITION
ROLLFORWARDNEEDED
SAFEGUARD
SQL
SQLPROGRAM
TABLE
UNSTRUCTURED
user-id is:
group-name.user-name
group-name.*
group-number, user-number
group-number,*
timestamp-field is:
CREATIONTIME
EXPIRATIONTIME
LASTOPENTIME
MODTIME
time-conditional is:
AFTER | >
BEFORE | <
time-value is:
[ date ] time
[ time ] date
time is:
hh:mm[:ss]
date is:
dd mmm yyyy
mmm dd yyyy
qualifier
is qualifying criteria that you can specify to include files or objects in a file set. You
can specify qualifiers in any order, but use each qualifier only once per file-set list.
EXCLUDE fileset
excludes the files you specify from a process. This specification is identical to
the NOT option. You can include wild-card characters in this specification.
catalog-list
specifies a volume and subvolume (or a DEFINE name of a file) containing
descriptions of the SQL objects (tables, indexes, views, or SQL programs).
START file-id
specifies a starting position within a file set or file-set list. The process starts on
the first file after the file ID you specify. Although the wild-card option is not
valid for this specification, the file ID does not have to correspond to an existing
file. This option is useful for restarting a process that was interrupted.
WHERE expression
provides additional qualifying criteria for files or objects. The order of
precedence in evaluating expressions is parentheses ( ), NOT, AND, and OR.
file-attribute
is an additional parameter to the WHERE qualifier. You must use the complete
parameter name in the command. Abbreviations are not permitted.
ALTKEY
adds, replaces, or alters an alternate-key specification.
AUDITED
specifies files that are audited by the HP NonStop Transaction
Management Facility (TMF).
BROKEN
specifies files that are broken. The file needs media recovery because an
I/O or consistency check failure occurred the last time it was open. A good
example of its use is: WHERE NOT BROKEN.
COLLATION
specifies an SQL collation object.
CORRUPT
specifies any files with questionable contents. The destination file is
marked CORRUPT automatically by FUP during DUP or LOAD operations.
If the DUP or LOAD operation is completed abnormally, the CORRUPT flag
remains set for the file. A good example of its use is: WHERE NOT
CORRUPT.
CRASHOPEN
specifies files not closed normally by the disk process. A good example of
its use is: WHERE NOT CRASHOPEN.
ENSCRIBE
specifies Enscribe files.
ENTRYSEQUENCED
specifies entry-sequenced files.
FORMAT1
describes files that are smaller than 2 GB minus 1 MB size limit. A Format
1 file cannot exceed the 2 GB minus 1 MB size limit.
FORMAT2
describes files that are bigger than 2 GB minus 1 MB size limit. A Format 2
file can exceed the 2 GB minus 1 MB size limit.
Note. The Format 1 files were created on systems running RVUs preceding D46.00
or G06.00. The Format 2 files were created on systems running D46.00, G06.00, or
later RVUs.
INDEX
specifies SQL indexes.
KEYSEQUENCED
specifies key-sequenced files.
LICENSED
specifies LICENSED files. These files are executable object files that run in
privileged mode.
OPEN
specifies files that are in the OPEN state.
PROGID
specifies files that have the PROGID flag set.
RELATIVE
specifies relative files.
ROLLFORWARDNEEDED
specifies any files that need an HP Transaction Management Facility
(TMF) rollforward process. A good example is: WHERE NOT
ROLLFORWARDNEEDED.
SAFEGUARD
specifies files protected by Safeguard at the file level.
SQL
specifies all types of SQL files except SQL program files.
SQLPROGRAM
specifies SQL program files.
TABLE
specifies SQL tables.
UNSTRUCTURED
specifies unstructured disk files.
OWNER = user-id
specifies user identification for the form specified.
timestamp-field
helps select a file based on a specific time:
CREATIONTIME
selects a file based on when it was created.
EXPIRATIONTIME
selects a file based on when it expires and can be purged. You must use
this option to select dates set with the NOPURGEUNTIL timestamp in FUP
ALTER.
LASTOPENTIME
selects a file based on when it was last opened.
MODTIME
selects a file based on when it was last modified.
time-conditional
selects a file based on a specific time:
[ AFTER | > ]
selects a file after the time you specify.
[ BEFORE | < ]
selects a file before the time you specify.
time-value
selects a file based on a specific time. The default is 00:00:00 (midnight) of the
current date:
hh:mm[:ss]
hh is the hour. mm is the minutes. ss is the seconds.
date is:
dd mmm yyyy
mmm dd yyyy
mmm is always the first three characters in the name of the month.
yyyy must be four digits for the year. Valid examples are:
1 JAN 2001 06:30
JAN 1 2001 06:30
6:30 1 JAN 2001
20 OCT 1999 07:29:30
06:15 JAN 1 1998
23:30:00 DEC 31 2000
FILECODE conditional-number
selects a file based on its file code. conditional-number represents a file
code (such as 101) and is preceded by one of these symbols:
< (less than)
<= (less than or equal to)
= (equal to)
> (greater than)
>= (greater than or equal to)
<> (not equal to)
EOF conditional-number
selects a file based on the number of bytes it contains. conditional-
number represents the number of bytes (you provide) and is preceded by one
of these symbols:
< (less than)
<= (less than or equal to)
= (equal to)
> (greater than)
>= (greater than or equal to)
<> (not equal to).
Note. Keywords for qualified file sets must be specified completely. Abbreviations are
not permitted. For example, use KEYSEQUENCED (not K), ENTRYSEQUENCED
(not E), RELATIVE (not R), UNSTRUCTURED (not U), and FILECODE (not CODE).
Note. Qualifiers appear immediately after the file set they are qualifying. This does
not have to be the end of the FUP command.
Creating Files
You can create files using FUP commands.
FUP
-ALTER filename, PART (sec-partition-num,
[\node.]$volume
[ , pri-extent-size [ , sec-extent-size] ]
• To create a partitioned alternate-key sequenced file using the SET command:
FUP
-SET TYPE K
-SET CODE 1001
-SET EXT (32,8)
-SET REC 54
-SET BLOCK 4096
-SET KEYLEN 2
-SET ALTKEY ("LO", KEYOFF 42, KEYLEN 4)
-SET ALTKEY ("VN", KEYOFF 46, KEYLEN 8)
-SET ALTFILE (0,invalt)
-SET PART (1, $ade001,5,5)
-CREATE INV
CREATED- $STORE1.SVOL1.INV
CREATED- $STORE1.SVOL1.INVALT
Partitioning Files
You can partition existing files using FUP commands.
Related Commands
COMMAND Function
CREATE Creates a file using the current file-creation parameter values that have
been defined with a SET command
PURGE Deletes a file
LOADALTFILE Creates an alternate-key file from a primary file
LISTOPENS Lists all processes that now have one or more designated files open
DEFINE names on NonStop systems always begin with the “=” character.
Note. For more information about DEFINEs, see the Guardian User’s Guide .
SPOOL DEFINEs
You can specify a SPOOL DEFINE in these situations:
• As listfile with the OUT option in:
MAP DEFINEs
You can specify a MAP DEFINE wherever FUP permits a file name. It is sometimes
easier to use a DEFINE name such as =CUSTOMERS than an actual file name such as
\SF.$ACCNTS.CURRNT.CUSTNMES.
TAPE DEFINEs
You can specify a TAPE DEFINE as:
• out-filename (destination) in a BUILDKEYRECORDS command
• in-filename (source) or out-filename (destination) in a COPY command
• in-filename (source) in a LOAD command
Note. Do not use TAPE DEFINE attributes that conflict with your FUP command parameters.
For more information about how FUP input and output options work with TAPE DEFINEs, see
Appendix A, DEFINE Tables.
At the time of creation, you can specify the file format, as Table 1-1 shows.
Note. FUP supports Format 2 SQL partitions in the G06.13 and later RVUs. FUP supports
fallback from G06.13 to as far back as G06.03. If you are running an RVU prior to G06.03, you
cannot fall back to your previous RVU.
FUP commands display error 584 when they encounter Format 2 SQL partitions on a system
that has fallen back from G06.13 to an earlier RVU. FUP continues to process the remaining
files.
When you specify Format 0, the system decides the format of the file based on other
file attributes. The system chooses Format 2 when the block size is over 4 KB or when
a partition size is over 2 GB minus 1 KB. For unstructured, relative, and entry-
sequenced files, the system chooses Format 2 when the total file size is 4 GB or more.
By default, Format 1 is assumed.
The FUP INFO command provides the value of the file format for existing files.
You likely need to change existing applications to access Format 2 files over 4 GB in
size that are not key-sequenced. These files require the use of a 64-bit primary key,
and system interfaces preceding G06.00 or D46.00 support only 32-bit primary keys.
Applications are unlikely to require any changes to access key-sequenced Format 2
files.
Except for key-sequenced files, all partitions of a file must have the same format
(Format 1 or Format 2). You cannot create a key-sequenced file with a mix of partition
formats. However, you can replace the partitions one at a time with equivalent Format
2 partitions if the block and record sizes are the same for all partitions.
You should convert a key-sequenced partition to Format 2 when it approaches its
capacity limit:
1. Create a new Format 2 replacement partition on a different volume.
2. Load the data into the new partition from the Format 1 partition.
3. Change the file partition information to point to the new Format 2 partition instead
of the Format 1 partition.
If Format 2 files are created on a system running D46.00, G06.00, or subsequent
RVUs, fallback to an RVU prior to D46.00 or G06.00 precludes access to the contents
of those files. Access to Format 2 files in an Expand network from RVUs before D46.00
or G06.00 will not be possible.
Note. The FUP SET, RESET, and SHOW commands do not apply to SQL/MP files. They
apply only to Enscribe files. The FUP control commands described in Appendix B, FUP
Command Summary, do not operate on SQL/MP files.
Table 1-2 shows the FUP commands that do not apply to all the different types of
SQL/MP files and lists any alternative methods for manipulating the files.
Table 1-2. FUP Commands and SQL/MP Files (page 1 of 2)
Applicable to SQL/MP File Type
SQLCI
FUP Command Table View Index Partition Catalog Equivalent
ALLOCATE No No No No No CREATE
TABLE
CREATE
INDEX
ALTER TABLE
ALTER INDEX
ALTER No No No No No ALTER TABLE
ALTER INDEX
BUILDKEYRECORDS No No No No No CREATE
TABLE
CHECKSUM Yes No Yes Yes No
COPY No No No No No COPY utility
CREATE No No No No No CREATE
TABLE
CREATE
INDEX
DEALLOCATE No No No No No ALTER TABLE
ALTER INDEX
DUP No No No No No DUP utility
FILES Yes Yes Yes Yes Yes
FILENAMES Yes Yes Yes Yes Yes
GIVE No No No No No ALTER
command
SECURE utility
INFO Yes Yes Yes Yes Yes FILEINFO
utility
LICENSE No No No No No
LISTOPENS Yes 1 Yes Yes Yes
LOAD No No No No No LOAD utility
LOADALTFILE No No No No No LOAD utility
PURGE No No No No No DROP
command
PURGE utility
PURGEDATA No No No No No
RELOAD Yes No Yes Yes Yes
1 = Yes for protection views; No for shorthand views
indexes that depend on the table are loaded automatically. No SQL/MP commands
load indexes directly.
• For compatibility with the FUP PURGE command, substitute the keyword PURGE
for DROP in the SQL/MP DROP utility.
!
Executes an existing command again.
-num
executes a command that appears before the current command. For example, use
!-3 to execute the third command prior to the current one.
num
is the number of a command line. For example, use !2 to execute the second
command of the current FUP session.
string
is the first character or characters of a previous command. For example, use !DUP
$ to execute the most recent DUP command that begins with a volume name.
quoted
is a string enclosed in either single or double quotation marks. FUP searches every
character in the command buffer—not just the first characters—until it finds the
string. For example, use !“\MAUI” to execute the most recent command that
referenced the system \MAUI.
! Guidelines
• If you use the ! command without a number or text string, FUP executes the last
command you entered again.
• FUP displays the specified command before it is executed.
• To display command line numbers or recent commands, use the FUP HISTORY
command.
! Examples
• To immediately view the FUP command entered on line eight and execute it again:
-!8
• To execute the command issued two commands prior to the current command:
-!-2
• To execute the last FUP INFO command again:
-!INFO
Commands Related to !
COMMAND Function Page
HISTORY Displays previous FUP commands 2-80
FC Modifies a previous command 2-71
? Displays a previous FUP command 2-5
?
Displays a specific command.
-num
displays a command that appears before the current command. For example, use
?-3 to display the third command prior to the current one.
num
is the number of a command line. For example, use ?2 to display the second
command of the current FUP session.
string
is the first character or characters of a previous command. For example, use ?DUP
$ to display the most recent DUP command that begins with a volume name.
quoted
is a string enclosed in either single or double quotation marks. FUP searches every
character in the command buffer—not just the first characters—until it finds the
string. For example, use ?“\MAUI” to display the most recent command that
referenced the system \MAUI.
? Guidelines
• To display the last command you entered, use the ? command without a number or
text string.
• To display command line numbers of recent commands, use the FUP HISTORY
command.
Commands Related to ?
COMMAND Function Page
HISTORY Displays previous FUP commands 2-80
FC Modifies a previous FUP command 2-71
! Reexecutes a previous FUP command 2-4
ALLOCATE
Allocates file extents for a disk file. This command applies only to Enscribe files.
fileset-list
is a list of disk files for which extents are to be allocated. You can specify
qualified-fileset for this fileset-list.
num-extents
is the total number of extents to be allocated to the file.
For nonpartitioned Disk Process 2 (DP2) files and key-sequenced partitioned files,
specify num-extents as a value from 1 through maximum-extents.
maximum-extents is the number of extents set with the MAXEXTENTS file
attribute when the file was created or last altered. The default value for
maximum-extents is 16. For more information, see SET on page 2-166.
For partitioned files that are not key-sequenced, specify num-extents as a value
from 1 through 16 multiplied by the number of partitions. Each partition is allocated
16 extents (beginning with the start of the file) until the total num-extents are
allocated.
Note. Extents 0 through 15 are in partition zero, extents 16 through 31 are in partition one,
extents 32 through 47 are in partition two, and so on.
PARTONLY
allocates extents to any primary and secondary partitions of partitioned files in
fileset-list. If a primary partition name is referenced, the extents are
allocated only to the primary partition. If you omit PARTONLY, FUP allocates
extents to all the partitioned files in fileset-list. PARTONLY has no effect on
nonpartitioned files.
ALLOCATE Guidelines
• num-extents has a different significance for different file types. For
key-sequenced partition files, it is the number of extents allocated in each partition.
For nonpartitioned files and partitioned files that are not key sequenced, it is the
total number of extents to be allocated to the file.
An example is trying to allocate 12 additional extents to a nonpartitioned file that
already has four extents allocated and trying to allocate 12 additional extents for
the primary partition of a key-sequenced file that already has four extents
allocated. Although you need to specify 16 for num-extents in both instances,
you must include the PARTONLY option with the key-sequenced file specification
so that the 12 extents are allocated only to the primary partition.
Note. You can allocate more than 16 extents on the last partition only when you use the
PARTONLY option with the ALLOCATE command.
• The ALLOCATE command cannot handle SQL files that are not SQL object files.
You must use the SQLCI CREATE and SQLCI ALTER commands.
• To allocate volume directory extents for fileset-list, you must use this file
name syntax:
$volume.SYS00.DIRECTRY
ALLOCATE Examples
To create an unstructured file and allocate 10 file extents for it:
1. To create an unstructured file when the default file-creation attributes are enabled:
-CREATE YRFILE
2. Use the FUP INFO, DETAIL command to see that no file extents are initially
allocated for the file:
-INFO YRFILE, DETAIL
$BOOKS1.COMLANG.YRFILE 14 April 2001, 9:00
ENSCRIBE
TYPE U
FORMAT 1
EXT ( 2 PAGES, 2 PAGES )
MAXEXTENTS 16
BUFFERSIZE 4096
OWNER 8,44
SECURITY (RWEP): NUNU
DATA MODIF: 14 April 2001, 08:59
CREATION DATE: 14 April 2001, 08:59
LAST OPEN: NEVER OPENED
EOF 0 (0.0 % USED)
FILE LABEL: 214 (5.2 % USED)
EXTENTS ALLOCATED: 0
This display indicates it is a DP2 file because:
• The listing includes the MAXEXTENTS, BUFFERSIZE, CREATION DATE, and
LAST OPEN attributes.
• The extent size for YRFILE is not the one page (2,048 bytes) FUP default.
During the file-creation process, DP2 rounds up the extent size (to 2 pages or
4,096 bytes) because the extent size of DP2 files must always be an integral
multiple of the BUFFERSIZE (for unstructured files) or of the BLOCK size (for
structured files).
To create an unstructured DP2 file with one-page extents, you must specify a
BUFFERSIZE of 2048 bytes with either the FUP SET or FUP CREATE command.
3. To allocate 10 file extents for YRFILE:
-ALLOCATE YRFILE, 10
4. Use the FUP INFO YRFILE, DETAIL command to see that the extents are now
allocated:
$BOOKS1.COMLANG.YRFILE 14 April 2001, 09:05
ENSCRIBE
TYPE U
FORMAT 1
EXT ( 2 PAGES, 2 PAGES )
MAXEXTENTS 16
BUFFERSIZE 4096
OWNER 8,44
SECURITY (RWEP): NUNU
DATA MODIF: 14 April 2001, 08:59
CREATION DATE: 14 April 2001, 08:59
LAST OPEN: 14 April 2001, 09:04
EOF 0 (0.0 % USED)
FILE LABEL: 214 (5.2 % USED)
EXTENTS ALLOCATED: 10
ALLOW
Sets the number of errors and warnings that FUP allows before it stops executing FUP
commands. If this number is exceeded, the current FUP command is aborted.
If you reach the error or warning limit while entering FUP commands interactively, FUP
terminates the current command, displays a warning message, displays its hyphen
prompt, and continues to accept commands.
If you reach the error or warning limit while FUP is executing commands from a
command file, FUP stops executing.
The ALLOW command became an option of the CONFIG[URE] command with the D30
product version of FUP. However, for compatibility purposes, any FUP product versions
prior to D30 continue to recognize the ALLOW option as command syntax. For more
information, see CONFIG[URE] on page 2-26.
ALTER
Changes some characteristics of an Enscribe disk file’s label. This command affects
only the file label. It does not create or purge files, and does not insert, delete, or move
records. It applies only to Enscribe files.
filename
is the name of the file that you want to alter. FUP expands a partial file name by
adding the current default names for system, volume, and subvolume. You cannot
use wild-card characters in filename or specify qualified-fileset for it.
alter-option
names the file characteristic you want altered. Available options depend on the file
type:
For all file types:
[ NO ] AUDIT
[ NO ] AUDITCOMPRESS
[ NO ] BUFFERED
CODE file-code
LOCKLENGTH generic-lock-key-length
MAXEXTENTS maximum-extents
NOPURGEUNTIL timestamp
[NO] REFRESH
RESETBROKEN
RESETCORRUPT
[ NO ] SERIALWRITES
[ NO ] VERIFIEDWRITES
[ NO ] AUDIT
designates the file as audited or nonaudited by TMF. If you specify AUDIT, FUP
marks the file as an audited file. If you specify NO AUDIT, FUP marks the file as a
nonaudited file. To use the AUDIT option, you must have read and write access to
a file (or be the super ID 255,255).
Note. For more information, see the TMF Planning and Configuration Guide.
[ NO ] AUDITCOMPRESS
sets the mode of producing audit-checkpoint messages (using compressed or
entire messages) for audited files. The default is NO AUDITCOMPRESS.
[ NO ] BUFFERED
sets the mode of handling write requests to the file using buffered or write-through
cache. BUFFERED specifies a buffered cache. NO BUFFERED specifies a write-
through cache. The default is BUFFERED for audited files and NO BUFFERED for
nonaudited files.
Caution. If you use the buffered-cache option on a DP2 file that TMF does not audit, a system
failure or disk-process takeover can cause the loss of buffered updates to the file. An
application might not detect this loss, or handle the loss correctly unless it is modified to do so.
CODE file-code
alters the file code. Specify file-code as an integer from 0 through 65,535. The
default file-code is zero. File codes 100 through 999 are reserved for use by
HP.
Note. For a list of reserved file codes, see Table 2-2, System File Code Definitions, on
page 2-88.
LOCKLENGTH generic-lock-key-length
is the lock-key length. The generic lock-key length determines the grouping of
records that share a single lock. The value specified must be between zero and the
key length of the file.
MAXEXTENTS maximum-extents
sets the maximum number of extents to allocate (for nonpartitioned files only).
Specify maximum-extents as an integer from 16 through n, where n is a
maximum value determined by the amount of free space in the file label. FUP
rounds any value you set from 1 through 15 up to 16. The absolute maximum is
978 extents. The default is 16.
You cannot always allocate all the extents specified by maximum-extents. The
actual number of extents that you can allocate depends on the amount of space in
the file label. If there are alternate keys or partitions, the maximum number of
extents allowed is fewer than 978.
If you specify MAXEXTENTS and the file is Format 1, you must consider the
primary and secondary extent sizes to avoid exceeding the maximum file size for a
Format 1 file. When the primary and secondary extent size with the specified
MAXEXTENT size is larger than two gigabytes, the ALTER operation rejects the
request with file-system error 21 (illegal count specified) if the file is Format 1.
When the primary and secondary extent size with the specified MAXEXTENT size
is larger than four gigabytes, the ALTER operation rejects the request with file-
system error 12 (file in use) if the file is:
• format 2 file that is not key-sequenced and not partitioned
• format 2 partitioned file that is key-sequenced and the operation is performed
on other than the base partition
• currently less than 4 gigabytes
• currently opened by another process
If the file is a partitioned file that is not key-sequenced, the ALTER operation
rejects the request with file-system error 2 (operation not allowed on this type of
file).
NOPURGEUNTIL [ timestamp ]
lets you change the expiration date of a file.
timestamp is:
[ date ] [ , ] [ time ]
where date is specified as:
{ dd mmm yyyy | mmm dd yyyy }
where dd (day) is an integer from 1 through 31, and mmm (month) is one of:
JAN, FEB, MAR, APR, MAY, JUN,
JUL, AUG, SEP, OCT, NOV, DEC
and yyyy (year) is a four-digit integer in the range 1900 to 3999.
If date is omitted, timestamp defaults to the current date.
time is specified as:
hh:mm[:ss]
values for the hour (hh) are from 0 to 23 (only the 24-hour clock is
supported).
If time is omitted, timestamp defaults to the beginning of the day (0:00:00 or
midnight).
If seconds (ss) is omitted, timestamp defaults to zero.
If both date and time are omitted, timestamp returns to the parameter
default setting at the time the file was created.
If timestamp contains a date prior to Jan 1, 1900 or after Dec 31, 4000, FUP
displays BAD TIMESTAMP and no date in the INFO DETAIL response screen.
[ NO ] REFRESH
causes the file label to be copied to disk whenever the file control block is marked
as dirty. This situation occurs if the end of file or a free-space block changes. The
file label is not written to disk if the only change is updating the LAST MODIFIED
field. The default is NO REFRESH. (The label is not copied to disk.)
RESETBROKEN
resets the BROKEN flag in the file label for a nonaudited file. Fix the broken parts
of the file before using RESETBROKEN.
RESETCORRUPT
causes the corrupt flag to be reset (turned off).
[ NO ] SERIALWRITES
sets the mode of writes to the mirror: serial or parallel. The default is NO
SERIALWRITES.
[ NO ] VERIFIEDWRITES
sets the mode of file writes: verified or unverified. The default is NO
VERIFIEDWRITES.
key-specifier
is a 2-byte value that identifies this alternate-key field. Specify
key-specifier as either a one or two-character string in quotation marks:
"[c1]c2"
or as an integer from -32,768 through 32,767:
{ -32,768 : 32,767 }
You can use any characters for key-specifier except zero. If you omit c1,
then c1 is treated as a zero.
Note. If you add a new key specifier that references an undefined key-file number,
you must include the ALTFILE option to define the alternate-key file.
altkey-param
specifies attributes of the alternate-key file.
FILE key-file-number
sets the key-file number for key-specifier. Specify key-file-number
as an integer in the range 0 through 255. This number is related to an
actual file by the ALTFILE create-param. The default
key-file-number is zero.
[ NO ] INSERTIONORDER
specifies whether to use insertion-ordered alternate-key sequencing. The
default, NO INSERTIONORDER, specifies to order alternate key records
of files with duplicate key values by their primary key sequence and not
their order of insertion.
An insertion-ordered alternate key cannot share an alternate-key file with
other keys of different lengths or with other keys that are not insertion
ordered.
All the nonunique alternate keys of a file must have the same duplicate-key
ordering attribute. A file with this specification must not have both insertion-
ordered alternate keys and standard (duplicate ordering by primary key)
nonunique alternate keys.
KEYLEN key-length
sets the key length for key-specifier. To create a key-sequenced file,
you must specify a KEYLEN, or the creation attempt fails.
KEYOFF key-offset
sets the key offset for key-specifier. The default is zero.
Note. For more information about null values, see the Enscribe Programmer’s
Guide.
[ NO ] UNIQUE
specifies whether to set key-specifier as a unique key. The default is
NO UNIQUE.
[ NO ] UPDATE
specifies whether to set automatic updating for the alternate-key file
represented by key-specifier.
The NO UPDATE option prevents the file system from automatically
updating the specified alternate-key file when you write to the main file.
Although you usually want to keep alternate-key files synchronized with
their main files, you sometimes might want to keep files unsynchronized.
For example, you might have two files pointing to the same alternate-key
file but only want updates from one of the two written to it.
The default is UPDATE.
DELALTFILE key-file-number
deletes the reference to an alternate-key file but does not purge the file. The
alternate-key file must not be referenced by any existing key-specifier. After
you execute an ALTER command with this option, the remaining key-file numbers
and references to them are adjusted. The numbers begin with zero and are
contiguous (0,1, 2, and so on).
DELALTKEY key-specifier
deletes an alternate-key specification. You cannot access the file through
key-specifier after you execute this option. The key-specifier parameter is
a 2-byte value that you already specified (in the ALTKEY parameter) to identify the
alternate-key field. The value is passed to the KEYPOSITION procedure when it is
referenced by this key field.
Specify key-specifier as a one or two-character string in quotation marks:
"[c1]c2"
Or specify it as an integer from -32,768 through 32,767:
{ -32,768 : 32,767 }
You can use any characters for key-specifier except zero. If you omit c1, then
c1 is treated as a zero.
PART
alters secondary partition specifications for partitioned files. You can specify extent
sizes if you are adding a new partition or if you are altering an existing partition of a
key-sequenced file. You must specify each secondary partition separately.
If the sec-partition-num already exists, for all file types, you can change
volume.
You can redefine the extent sizes for key-sequenced files only. If the partition
contains data, you must perform additional operations on the partition to complete
the extent-size change.
You can add a new, additional partition to any files that are not key sequenced. You
cannot add partitions to key-sequenced files.
sec-partition-num , \node.$volume
names the volume where this secondary partition is to reside. Specify
sec-partition-num as an integer from 1 through 15 to designate the
secondary partition. Specify node and volume as the names of the node and
volume to contain the partition. The file name and the subvolume of the
primary partition are specified when the file is created.
Although FUP lets you specify any number in the range 1 through 15 for
sec-partition-num, FUP changes it to a standard DP2 number that starts
at zero when the file is created.
Note. Example 2-6 on page 2-99 shows how the DP2 number is listed in the INFO
DETAIL command.
pri-extent-size , sec-extent-size
alters the primary and secondary extent sizes. The default value for
pri-extent-size is one page (2,048 bytes). If you specify a value of zero or
do not specify a value, sec-extent-size defaults to the
pri-extent-size value. You cannot define primary or secondary extents as
zero pages when altering files.
For partitioned unstructured files where one or more partitions reside on a 512
byte sector disk enclosure in an external disk enclosure 1, DP2 has additional
restrictions. You must specify both the pri-extent-size and the sec-
extent-size so that they can be explicitly divisible by 14. DP2 does not
automatically round the size up. Additionally, you must specify the same pri-
extent-size for all partitions and the same sec-extent-size for all
partitions.
1. Examples of such disks are JBOD (sub-type = 53) and ESS (sub-type = 52).
0:maximum [ PAGE[S] ]
specifies the extent size in pages (2,048-byte units). Possible values are:
Format 1 Format 2
0:65,535 [ PAGE[S] ] 0:512,000,000 [ PAGE[S] ]
Note. If you specify an extent size over 65,535 pages, you must assign Format 2
to your files. For more information about Format 2 files, see Handling Different
Types of Files on page 1-22.
Because the minimum extent size is one page (2,048 bytes), one page is
also allocated if you specify zero extents.
0:maximum BYTE[S]
specifies the extent size in bytes. Possible values of maximum are:
Format 1 Format 2
0:134,215,680 BYTE[S] 0:2,147,483,647 BYTE[S]
The FUP process rounds the extent size up to the next full page. If you
specify 2,047 bytes, FUP allocates one page. If you specify 2,049 bytes,
FUP allocates two pages.
0:maximum REC[S]
specifies the extent size based on the current settings for
record-length (REC), data-block-length (BLOCK),
index-block-length (IBLOCK), key-field lengths, and compression
settings. Possible values are:
Format 1 Format 2
0:134,215,680 REC[S] 0:2,147,483,647 REC[S]
The FUP process rounds the extent size up to the next full page.
0:maximum MEGABYTE[S]
specifies extent sizes in million-byte units. Possible values are:
Format 1 Format 2
0:134 MEGABYTE[S] 0:2147 MEGABYTE[S]
The FUP process rounds the extent size up to the next full page.
PARTONLY
specifies that any changes you make with the ALTER command apply only to the
indicated file partition. If you reference a primary partition name, the extents are
altered only in the primary partition.
BUFFERSIZE unstructured-buffer-size
is the internal buffer size to use when accessing the specified file. You can set the
BUFFERSIZE file attribute with this command (FUP ALTER) or with the FUP
CREATE and FUP SET commands. Possible values for unstructured-
buffer-size (in bytes) are:
FUP rounds the actual buffer size up to the nearest valid DP2 block size.
ODDUNSTR
changes an even unstructured file to an odd unstructured file.
Unstructured Enscribe files can be even or odd. FUP rounds up any odd byte
count that you give to an even unstructured file (for reading, writing, or positioning).
This is the default for unstructured files.
FUP does not round odd unstructured files up. You always read, write, or position
at the byte count you give.
To change an odd unstructured file to an even unstructured file, copy the odd file
into a new file that was created as even unstructured.
ALTER Guidelines
• To receive the current file attributes for any file you want to alter, use the FUP
INFO command:
-INFO filename , DETAIL
• To alter a file, you must have both read and write access to it.
• Changing the AUDIT option for DP2 files also changes the default value of the
BUFFERED attribute.
If you specify NO AUDIT, the BUFFERED option and file label default is set to NO
BUFFERED. If you specify AUDIT, the BUFFERED option is set ON. If you have
explicitly set the BUFFERED attribute, that value remains unchanged.
• Adding the AUDIT attribute to a file causes audit records to be written for the file.
• If you use the AUDIT option and the volume containing the primary file, or any of its
secondary partitions or alternate-key files containing at least one automatically
updated alternate key, is not audited, the request fails. You receive file-system
error 80 (invalid operation on audited file or nonaudited disk volume). If these files
are all audited, the labels of the alternate-key files are updated to reflect the audit
option.
• To add an alternate key to a structured file that does not have any alternate-key
files specified, your ALTER command must specify both the new alternate key and
the new alternate-key file. This is also true if you try to delete an alternate key with
the DELALTFILE option of the ALTER command.
For example, to add the alternate key “aa” to the file FRED and specify the
alternate-key file AFILE:
-ALTER FRED, ALTFILE (0, AFILE), ALTKEY ("aa", FILE 0,&
-KEYOFF 0, KEYLEN 5)
• Altering the NULL or UPDATE attribute of an alternate key for a file does not
change the actual contents of the alternate-key file. You must update the data
(usually with the FUP LOADALTFILE command) to make the alternate-key file
completely consistent with the primary-key file. If your application does not require
complete consistency, you might not need to reload the alternate-key file.
• Altering the UNIQUE attribute of an alternate key for a file makes the file
description inconsistent with its alternate-key files. Subsequent attempts to open
the file cause file-system error 4 (failure to open an alternate-key file). To use the
file after the UNIQUE attribute for an alternate key is altered, purge the alternate-
key file, re-create it, and adjust its key length. Files with the UNIQUE attribute have
a different key length than files with the NO UNIQUE attribute.
Note. For more information about the storage of alternate keys in alternate-key files, see
the Enscribe Programmer’s Guide.
• When you alter an alternate key from NO UNIQUE to UNIQUE, the contents of the
records in the primary file are not examined. The file could contain records with
duplicate values in the alternate-key field while you are changing its description.
The FUP ALTER command does not recognize the existence of duplicates, but the
duplicates cause an error when the alternate-key file is loaded.
• If the extent sizes of a secondary partition are altered, this warning message
appears:
USING SPECIFIED EXTENT SIZES: USER MUST ENSURE
CONSISTENT PARTITIONS
• The ALTER command cannot handle SQL files that are not SQL object files. To do
this, you must use SQLCI TABLE and SQLCI INDEX.
• The ALTER command returns error 197 (an SQL error has occurred) against
SQL-compiled objects.
• All partitions of a file are created with the same format version. For files that are
not key-sequenced, a partition created independently must have the same format
as all other partitions of the file.
Caution. If you use the ALTER command to change a partition to a different format from the
rest of the partitions, the file system reports errors. FUP does not check for such errors.
ALTER Examples
• To assign file code 10 to MYFILE1:
-ALTER MYFILE1, CODE 10
• To cause the file label for MYFILE2 to be updated whenever the file control block
(FCB) changes:
-ALTER MYFILE2, REFRESH
• To assign the alternate-key file MYFILE4 to MYFILE3 and give it key-file number
two (ALTFILE 2 must already be a defined attribute of MYFILE3):
-ALTER MYFILE3, ALTFILE (2 , MYFILE4)
• To delete the alternate key "ab" from MYFILE5:
-ALTER MYFILE5, DELALTKEY "ab", DELALTFILE 0
BUILDKEYRECORDS
Generates the alternate-key records for specified key fields of a structured disk file and
writes those records to a designated file. This command applies only to Enscribe files.
Although the output of BUILDKEYRECORDS can be the actual destination alternate-
key file, it is more efficient to use a LOADALTFILE command to load alternate-keys.
You can use BUILDKEYRECORDS to generate the alternate-key records, store them
in another file, and then load them into the destination alternate-key file with a LOAD
command. This approach is useful when limited system resources prevent a
LOADALTFILE operation.
key-specifier-list is:
key-specifier
( key-specifier [ , key-specifier ] )
out-option is:
BLOCKOUT out-block-length
DENSITYOUT density
EBCDICOUT
PAD [ pad-character ]
RECOUT out-record-length
[ NO ] REWINDOUT
SKIPOUT num-eofs
TAPEMODE mode
[ NO ] UNLOADOUT
XLATE [ translation-table-name ]
XLATEIN [ translation-table-name ]
XLATEOUT [ translation-table-name ]
primary-filename
names an existing primary file whose alternate-key records are to be generated.
You must have previously defined alternate-key fields for the primary file. You
cannot use wild-card characters or specify qualified-fileset for
primary-filename.
out-filename
names an existing file where the alternate-key records generated by this command
are to be written. out-filename can be an existing disk file, a nondisk device, a
process, a tape or SPOOLER (code 129) file, or a spool DEFINE name. You
cannot use wild-card characters or specify qualified-fileset for
out-filename.
key-specifier-list
names one or more alternate-key fields of the primary file whose corresponding
alternate-key records are to be generated.
key-specifier
is a 2-byte value that identifies the alternate-key field. Specify it as a one or
two-character string in quotation marks:
"[c1]c2"
Or as an integer from -32,768 through 32,767:
{ -32,768 : 32,767 }
You can use any characters for key-specifier except zero. If you omit c1,
then c1 is treated as a zero.
out-option
specifies the format and control of out-filename. out-option is any one of:
BLOCKOUT out-block-length
DENSITYOUT density
EBCDICOUT
PAD [ pad-character ]
RECOUT out-record-length
[ NO ] REWINDOUT
SKIPOUT num-eofs
TAPEMODE mode
[ NO ] UNLOADOUT
XLATE [ xlate-table-name ]
XLATEIN [ xlate-table-name ]
XLATEOUT [ xlate-table-name ]
Note. For a complete description of these options, see out-option on page 2-45.
BUILDKEYRECORDS Guidelines
• BUILDKEYRECORDS causes the primary file to be read sequentially according to
its primary-key field.
For each record read from the primary file, BUILDKEYRECORDS generates one
or more alternate-key records (corresponding to the number of key specifiers that
are named) and writes them to the out file.
If you name more than one key specifier, BUILDKEYRECORDS generates the
corresponding alternate-key file records in the order of the ASCII collating
sequence for the key specifiers.
BUILDKEYRECORDS Example
To generate the alternate-key records for MYFILE using key specifications “ab” and
“cd”, and write the alternate-key records to $TAPE:
-BUILDKEYRECORDS MYFILE, $TAPE, ("ab","cd")
CHECKSUM
Recomputes the checksum value for blocks of data in disk files. Use this command
when recovering from a software-detected checksum error.
fileset-list
is a list of files whose checksum values are to be recomputed. fileset-list can
include Enscribe files (structured or unstructured) and SQL files. You can specify
qualified-fileset for fileset-list.
Any SQL views (protection or shorthand) in fileset-list are ignored, and a
warning is issued.
PARTONLY
specifies that FUP is to compute checksum values for all the primary and
secondary partitions encountered in fileset-list and for all nonpartitioned
files. If a primary partition name is referenced, the secondary partitions of the file
are not checksummed.
CHECKSUM Guidelines
• Although CHECKSUM reads each block of data (from each file specified by
fileset-list) and recomputes a checksum value for each one, it rewrites only
blocks whose checksum values are incorrect.
• Checksum errors usually indicate a potential data integrity problem. CHECKSUM
recomputes the checksum value for blocks of data but does not fix any data that
might have changed.
• If a Peripheral Utility Program (PUP) DOWN or STOPOPENS command was
executed for the volume named in fileset-list, CHECKSUM aborts.
• If an open file or SQL view is encountered during processing, CHECKSUM skips it,
and a warning message appears.
• If the fileset-list includes an asterisk (*) to specify all files in the file set, and
if the PARTONLY option has not been specified, CHECKSUM skips secondary
partitions of partitioned files (Enscribe and SQL).
• CHECKSUM accepts SQL catalogs but not SQL views.
• The maximum number of files in the fileset-list that can be checksummed at a time
is 2,147,483,647.
CHECKSUM Examples
• To recompute checksum values for all the files on volume $TEST1 (except
secondary partitions of partitioned files):
-CHECKSUM $TEST1.*.*
If partitioned files exist on $TEST1, FUP skips the secondary partitions because
the wild-card character (*) is specified for the files in fileset-list, and the
PARTONLY option is not included in the command.
• To recompute checksum values for all the files on volume $TEST1 (including any
existing partitioned files):
-CHECKSUM $TEST1.*.*, PARTONLY
• In this example, assume that a checksum error for $DATA appears on the operator
console and that $DATA contains these SQL and Enscribe files:
° Checksum values are recomputed for SQL files KSTABLE and KSTABLEI, for
the SQL object program file OFILE, and for the Enscribe files SFILE and
UFILE.
° SQL views PVIEW and SVIEW are skipped, and a warning message appears.
CONFIG[URE]
Customizes your FUP configuration information. You can use the CONFIG
abbreviation for this command. The COPY, DUP[LICATE], LOAD, LOADALTFILE, and
RELOAD configuration commands modify the defaults of their corresponding FUP
commands. These modified defaults become applicable when the corresponding
command is executed. The commands apply to other FUP commands as specified.
config-command is:
command-option | environment-option
command-option is:
environment-option is:
config-command
specifies FUP configuration information:
allow-option is:
{ num } [ [SEVERE ] ERRORS ]
{ num } [ WARNINGS ]
[ ABENDS [ ON | OFF ] ]
num
is the maximum number of errors or warnings that can occur before a
FUP command is aborted. The default ALLOW count is 0 for severe
and nonsevere errors; it is unlimited for warnings. The allowable range
for errors is 0 through 32767. FUP keeps an error count internally for
both severe and nonsevere errors.
num ERRORS
sets the nonsevere error count to num.
num WARNINGS
sets the number of warnings that FUP allows while executing FUP
commands.
ABENDS [ ON | OFF ]
specifies whether FUP should terminate abnormally (ABEND) when it
encounters an error and the allowed count for the error was exceeded.
The default for ABENDS depends on:
• The default when FUP is not used interactively is ON.
• If FUP is used interactively (the IN file is a terminal), the ABENDS
option is set to OFF.
• If ABENDS is specified without ON or OFF, the ABENDS option is set
to ON.
• Errors from which the FUP process cannot recover always cause FUP
to terminate abnormally.
Note. Table 2-1 shows FUP responses to warnings and error types when
ALLOW ABENDS is ON or OFF.
DISPLAYBITS bitcount
lets the COPY, INFO, and SHOW commands display 8-bit character sets. The
default is 7-bit characters.
bitcount
is the number of bits used to determine a printable character for the FUP
display commands (COPY, INFO, and SHOW). Use 8 for 8-bit character
sets or 7 for 7-bit character sets. The default value is 7.
ECHO CONFIGURE
specifies that FUP should display the options currently configured to the
home terminal (or list file) when a FUP command that can be configured is
executed.
If no options are configured for the current command, nothing is displayed.
For example, FUP displays the current CONFIGURE options for a
DUP[LICATE] process when a DUP[LICATE] command is executed.
All options configured for a specific command are displayed, including any
that are explicitly overridden on the command.
ECHO OBEY
specifies that FUP should display the FUP commands that are executed in
a command file to the home terminal (or list file). If you specify the ECHO
OBEY option in a command file (including the FUPCSTM or FUPLOCL
files), it causes an echo to start on the next command in the file.
You can specify CONFIGURE and OBEY on the same command. The ECHO
option without any parameters implies both CONFIGURE and OBEY. The
default is ECHO OBEY and NO ECHO CONFIGURE.
[ NO ] DISPLAYALLNUM
specifies the way you want to display the key values.
If DISPLAYALLNUM is not configured, the key values are displayed as
characters if they are all printable, else they are displayed as a mixture of
characters and numbers if there is any non-printable character.
If DISPLAYALLNUM is configured, the key values are displayed as characters
if they are all printable, else they are displayed as all numbers if there is any
non-printable character.
IOTIMEOUT time
specifies the time (in seconds) FUP should use for I/O timeouts with any
DUP[LICATE] commands. Use this option only at sites where configurations
can cause a DUP[LICATE] process to time out. The default values are two
minutes for disk transfers and ten minutes for optical transfers.
Note. A value of 0 for time specifies the FUP defaults, and -1 specifies no timeouts.
No other negative values are permitted.
NETBLOCKSIZE size
specifies the size of the blocks a DUP[LICATE] process uses to transfer data
when the SOURCE or TARGET files are not on the nodes where FUP is
currently running. The size is expressed in units of 1024 bytes. You must
specify the size using numbers from 4 through 28. For example, 28 specifies
28 KB block transfers. FUP rounds numbers down to the nearest 4KB units.
By default, FUP uses 28 KB transfers when the files are not local, and 56 KB
transfers for local files.
[ NO ] PROMPT [ PURGE ]
specifies if you want to be prompted prior to any PURGE processes. The NO
PROMPT PURGE option directs FUP to purge files without issuing any
prompts. The PROMPT PURGE option directs FUP to issue prompts before
purging files—unless an “!” was entered with the PURGE command. PROMPT
without any parameters implies purge. The default is PROMPT.
REPORTWIDTH width
specifies the maximum length (in columns) that FUP uses to format output for
commands (if applicable). width can be from 80 through 132. Although this
option replaces the FUP REPORTWIDTH command, it has the same syntax
and meaning.
RESTARTUPDATE time
specifies the time duration (in seconds) between updates to the ZZRESTRT
file on any DUP[LICATE] process with the RESTARTABLE option. By default,
the interval is 30 seconds for disk transfers and 10 seconds for optical
transfers. The time you specify must be a positive number; 0 specifies the FUP
defaults.
Because each update to the RESTART file requires a synchronized update to
the TARGET file, smaller values can slow down a DUP[LICATE] process.
STATONLY
suppresses the implicit INFO DETAIL display if INFO STAT is specified.
CONFIG[URE] Guidelines
• The ALLOW, DISPLAYBITS, and REPORTWIDTH commands from the previous
product versions of FUP are now CONFIGURE commands. However, for
compatibility, they are also supported as separate commands.
• Any values specified in the CONFIGURE command are enabled until they are
reset (with a RESET CONFIGURE command) or until the FUP session ends. To
save configuration values, use the SHOW command. You can set configuration
values from the FUPCSTM or FUPLOCL files or from a file with an OBEY
command.
• FUP uses the current configuration options wherever applicable. You can set or
change the configuration options in the FUPLOCL file or the FUPCSTM file or do
so interactively with the CONFIGURE or RESET commands.
• If an option is specified more than once, FUP uses the last one specified.
• Configuration options have precedence over the standard FUP defaults, but explicit
options at the command line can override them—unless the configuration option
does not have a named default.
To cancel configuration options that were set with the CONFIGURE command, you
must use the RESET command.
For example, if you specify CONFIGURE COPY UNSTRUCTURED, the
UNSTRUCTURED value applies to all COPY commands until it is reset. You
cannot override the UNSTRUCTURED option in a COPY command because it
does not have a named default. To cancel it, you must enter RESET COPY
UNSTRUCTURED.
However, the explicit SOURCEDATE option in a DUP A, B, SOURCEDATE
command can override a SAVEALL option used in a CONFIGURE DUP SAVEALL
command. SOURCEDATE and SAVEALL are in the same family of options, and
the explicit option in the DUP command overrides the option in the CONFIGURE
command.
• Because CONFIGURE remembers options that contain file or volume names, such
as DUP ALTFILE filename or LOAD PARTOF volume, you must be careful. The
file names often apply only to specific parts of a file and cannot be overridden
because they have no named defaults.
• Although using CONFIGURE with options that have no named defaults (including
the options with file names) is more difficult than using options with named
defaults, it can be useful on consecutive FUP commands that apply to the same
circumstance.
• You can use the ALLOW config-command to force FUP to execute a command
on each file within a file set when some of the commands might generate errors.
The allowed error counts do not apply to command parsing errors or errors that
FUP classifies as calamities (errors from which FUP cannot recover).
• To set defaults for COPY, INFO (DETAIL), and SHOW commands, use
CONFIG[URE] DISPLAYBITS:
° When bitcount is 7, COPY displays the character for any byte with an octal
value from %40 through %177. A period (.) appears for an octal value greater
than %177. When bitcount is 8, COPY displays the character for any byte
with an octal value of %40 or greater.
° When bitcount is 7, the ALTKEY key specifier of INFO, DETAIL, and SHOW
displays for any character with an octal value greater than %177. When
bitcount is 8, the ALTKEY key specifier is displayed for all characters with
an octal value greater than %40.
° You can switch between the 7-bit and 8-bit display modes during an interactive
FUP session. To reset the display mode, enter another DISPLAYBITS
command at the FUP prompt with the desired mode.
• To set defaults for COPY (with a DUMP option), FILES, SUBVOLS, and
LISTLOCKS commands, use CONFIG[URE] REPORTWIDTH:
° If the OUT file (or OUT device) record length is less than REPORTWIDTH,
FUP breaks the output records into multiple records (or lines) to fit the file or
device record length.
° The output format for all other FUP commands that output information is 80
columns—regardless of the value specified for REPORTWIDTH.
• You can specify translation table defaults for COPY, LOAD, and
BUILDKEYRECORDS commands:
° If you use the XLATE option but an XLATE table was not specified in the
CONFIGURE command, an error occurs.
CONFIG[URE] Examples
• To use CONFIGURE to set the default options for a DUP[LICATE] process:
-CONFIGURE DUP PURGE, SAVEID
-DUP A, B
FILES DUPLICATED: 1
The first command (CONFIGURE DUP PURGE, SAVEID) declares the
DUP[LICATE] configure options. The second command (DUP A, B) uses the
PURGE and SAVEID options implicitly. Then FUP displays the number of files that
are duplicated (1).
Note. For information on displaying file contents with the COPY command, see COPY: Display
Form on page 2-55.
You need to understand when to use the COPY command, DUP[LICATE], and LOAD
commands:
• Use the COPY command to change file attributes, to copy files to or from nondisk
devices, or to display records in a file (optionally in different formats).
• Use the DUP[LICATE] command to create identical copies of disk files.
• Use the LOAD command to create a structured disk file from scratch. It is much
faster than the COPY command.
Use the Copy Form of the COPY command to make a record-by-record copy from one
file to another.
copy-option is:
control-option
in-option
out-option
display-option
control-option is:
COUNT num-records
FIRST { ordinal-record-num }
{ KEY { record-spec | key-value } }
{ KEY64 { record-spec } }
{ key-specifier ALTKEY key-value }
[ NO ] TITLE
UNSTR[UCTURED]
UPSHIFT
in-option is:
BLOCKIN in-block-length
[ NO ] COMPACT
EBCDICIN
RECIN in-record-length
REELS num-reels
[ NO ] REWINDIN
SHARE
SKIPIN num-eofs
TRIM [ trim-character ]
[ NO ] UNLOADIN
VARIN
out-option is:
BLOCKOUT out-block-length
DENSITYOUT density
EBCDICOUT
FOLD
PAD [ pad-character ]
RECOUT out-record-length
[ NO ] REWINDOUT
SKIPOUT num-eofs
TAPEMODE mode
[ NO ] UNLOADOUT
VAROUT
XLATE [ translation-table-name ]
XLATEIN [ translation-table-name ]
XLATEOUT [ translation-table-name ]
display-option is:
O[CTAL]
D[ECIMAL]
H[EX]
BYTE[S]
A[SCII]
NO HEAD
[ NO ] TITLE
in-filename
is the name of the file that is the source of the copy. This file can be a disk file, a
nondisk device, a process, or a tape DEFINE name. The file can also be any type
of file that FUP handles, including structured files, EDIT files (up to 99,999,000
lines), or SPOOLER (file code 129) files. You cannot use wild-card characters in
in-filename or specify qualified-fileset for it.
If you omit in-filename, the IN file enabled for FUP is used. For example, this
situation occurs when you use FUP interactively, causing the home terminal to be
the IN file.
Note. For information on relative files, see COPY: Copy Form Guidelines on page 2-51.
out-filename
is the name of the file that is the destination of the copy. This file can be a nondisk
device, a process, an existing disk file, or a tape or spool DEFINE name. The file
can also be any type of file that FUP handles, including structured files, EDIT files,
or SPOOLER (file code 129) files. You cannot use wild-card characters in
out-filename or specify a qualified-fileset for it. For EDIT files, the
maximum number of output lines is 99,999,000.
If you omit out-filename, the OUT file enabled for FUP is used. For example,
this situation occurs when you use FUP interactively, causing the home terminal to
be the OUT file.
Although existing data in out-filename is never overwritten during a COPY
operation, the placement of the records being copied depends on the file type
specified for the destination file.
Note. For more information on the placement of the records, see COPY: Copy Form
Guidelines on page 2-51.
control-option
controls the method used for copying.
COUNT num-records
is the number of records to copy.
• If you omit COUNT and do not use the FIRST option, FUP copies all
records from the first record through the end-of-file (EOF).
• If you omit COUNT and use FIRST, FUP copies all records from the file
indicated by FIRST through the EOF.
FIRST { ordinal-record-num }
{ KEY { record-spec | key-value } }
{ KEY64 { record-spec } }
{ key-specifier ALTKEY key-value }
names the starting record of the input file for the copy. If you omit FIRST, the
copy starts with the first record of the input file.
ordinal-record-num
is the number of records (from the beginning of the file) to be skipped. The
first record in a file is record zero. If you specify this option for an
unstructured disk file, the copy begins at:
ordinal-record-num * in-record-length
Note. The actual reading begins with the first record in the source file.
For example, specify a key value as the ASCII string “T905”, followed by a
word containing the integer value zero, and a word containing the integer
value nine:
[ "T905", 0, 0, 0, 9 ]
KEY64 { record-spec }
specifies the 64-bit key value for the starting record of an entry-sequenced
file. FUP begins reading the input file at the record you name as KEY64.
key-specifier
is a one-character or two-character string (located inside quotation marks)
specifying the alternate key to use for positioning purposes.
ALTKEY key-value
specifies the alternate key of the starting record for a key-sequenced disk
file. FUP begins reading the input file at the specified record and obtains
subsequent rows in alternate key order. Specify key-value for key-
sequenced files as described in KEY { record-spec | key-value } on
page 2-39.
[ NO ] TITLE
directs FUP to write a title line that includes the name of the file, the time of the
COPY process, and the last modification time of the file. The title line is the first
line of the OUT file, followed by a blank line. These two lines become part of
the OUT file and are included in the RECORDS TRANSFERRED count. The
default is NO TITLE.
UNSTR[UCTURED]
causes an unstructured open on a file. You can use this option for any
unstructured disk file that is not in EDIT file format (or any Enscribe file
structure) and for any structured file where you want FUP to ignore its file
structure. This option is for disk files only.
You must use this option on partitioned files to copy partitions individually.
UPSHIFT
converts lowercase alphabetic characters to uppercase.
in-option
controls the handling of the copied input file. The in-option parameter is not
normally used with structured disk files.
BLOCKIN in-block-length
specifies the number of bytes in an input block. This number can be a value in
the range 1 through 32,767 indicating the actual number of bytes requested in
a single physical read operation.
Input records are blocked if in-block-length exceeds
in-record-length. Records of the specified length are extracted from the
input block until the number of bytes extracted equals in-block-length or
until the last input record is read.
The read count for the records in a block (with the exception of the last record
in the block) is equal to in-record-length. If in-block-length is not an
exact multiple of in-record-length, the last record extracted from a full
block is a short record with a read count equal to the number of bytes
extracted.
[ NO ] COMPACT
controls whether zero-length records are ignored when reading the file. This
applies only to any IN relative files. The default is COMPACT (empty records
are ignored), and records that follow an empty record in the OUT file are
renumbered.
If you specify NO COMPACT, COPY transfers empty records from the IN
relative files.
If you select the COMPACT option and the source is a relative file that contains
empty records, this message appears:
source file : EMPTY RECORD FOUND AND NOT TRANSFERRED
This message indicates that the target file has fewer records than the source
file. It is issued only once, when the first empty record is encountered.
EBCDICIN
treats the file specified for in-filename as though it contains EBCDIC
characters and translates the characters to their ASCII equivalents.
All the characters converted between EBCDIC and ASCII retain their own
image—with these exceptions:
EBCDIC ASCII
Logical OR Exclamation point
Cent sign Left square bracket
Exclamation point Right square bracket
Logical NOT sign Circumflex
RECIN in-record-length
specifies the maximum number of bytes in an input record.
If you include the RECIN option (and in-filename is a tape DEFINE), the
RECLEN attribute of the DEFINE must either be unspecified or match the
RECIN in-record-length.
If you include the RECIN option, the actual number of bytes in each input
record (the read count) depends on whether you also specify the TRIM option:
• If you do not specify the TRIM option, the read count is the actual number
of bytes in the input record. Although the read count for unstructured files
that are not in EDIT file format is exactly in-record-length bytes, the
last record of the file might be less. The read count is the number of bytes
actually read for all other files.
• If you specify the TRIM option, every trailing trim-character is deleted
from the input record. The read count includes only the significant
characters (those not trimmed).
If you omit the RECIN option, FUP determines the in-record-length value:
• If in-filename is a tape DEFINE that has RECLEN specified, USE set to
IN (or not specified), and LABELS set for labeled tape processing,
in-record-length is set to the RECLEN value.
• If you specify in-block-length using the FUP BLOCKIN option or the
BLOCKLEN attribute of a tape DEFINE (and it is less than or equal to
4096), that value is used for in-record-length. If you specify an
in-block-length greater than 4096, the value of in-record-length
is 4096.
REELS num-reels
indicates the use of multiple reels and sets the number of reels that make up
in-filename (for unlabeled magnetic tape only). You cannot specify this
option if in-filename is a tape DEFINE with LABELS set for labeled tape
processing. You must use the appropriate tape DEFINE attributes to process
multi-reel labeled tapes.
Note. For D-series and G-series RVUs, see the Guardian User’s Guide for more
information. This parameter does not apply to H-series RVUs because reel-to-reel
tapes are not supported.
Specify num-reels as an integer in the range 2 through 255. The tape is read
until num-reel occurrences. FUP always requests num-reels tapes
because it can recognize the end of a multiple reel (unlabeled) tape file only by
using the num-reels specification. At the end of each reel (except the last
one, depending on the REWINDIN and UNLOADIN parameters), the tape is
rewound and unloaded, and you are prompted for the next reel.
The FUP process expects two consecutive EOF marks, and it cannot
accommodate multireel unlabeled tapes that are written on IBM systems (or
any other systems) that do not adhere to the two consecutive EOF mark
format.
For FUP to distinguish between the end of an intermediate volume and the
EOF, the value of num-reels must be correct.
If you include the REELS option in the COPY command and in-filename is
a tape DEFINE, FUP displays this error message:
ERROR - REELS PARAMETER NOT ALLOWED
FOR LABELED TAPES
If you omit REELS, in-filename data transfer terminates when FUP
encounters a single EOF mark.
Labeled-tape handling for multiple reels follows IBM or ANSI standards, which
do not have the limitations imposed by FUP.
[ NO ] REWINDIN
specifies that the tape is rewound (or not rewound) when the EOF is read from
the tape (for magnetic tape only). If you specify NO REWINDIN, the tape
SHARE
opens in-filename with a shared exclusion mode (for disk files only). Using
SHARE lets you copy a file even if it is currently opened by another process
(unless it is open with exclusive exclusion mode). If you omit SHARE (and
in-filename is a disk file), the file is opened with protected exclusion mode.
SKIPIN num-eofs
moves the tape past num-eofs end-of-file (EOF) marks before the data
transfer begins (for magnetic tape only). You cannot specify this option if
in-filename is a tape DEFINE with LABELS set for labeled tape processing.
You must use the appropriate tape DEFINE attribute to skip files on a labeled
tape.
If you include the SKIPIN option (and in-filename is a tape DEFINE), FUP
displays this error message:
ERROR - SKIPIN PARAMETER NOT ALLOWED FOR
LABELED TAPES
Specify num-eofs as an integer in the range -255 through 255.
• If you specify a positive value, the tape is forwarded past num-eofs EOF
marks and positioned immediately after the last EOF mark passed.
• If you specify a negative value, the tape is rewound over num-eofs EOF
marks (-1 multiplied by num-eofs) and then moved forward and
positioned immediately ahead of the last EOF mark passed.
• If you specify a value of zero, the SKIPIN option is ignored.
• If you omit the SKIPIN option, the tape remains at its current position, and
the data transfer begins with the next physical record on tape.
TRIM [ trim-character ]
deletes any trailing characters matching trim-character. Specify
trim-character as either a single ASCII character in quotation marks:
"c"
Or specify it as an integer in the range 0 through 255 that specifies a byte
value:
{ 0:255 }
[ NO ] UNLOADIN
specifies whether the tape is unloaded after it is rewound (for magnetic tape
only). The default is UNLOADIN. (The tape is unloaded after it is rewound.)
This option also applies to labeled tapes.
VARIN
reads variable-length, blocked records. These records can be produced using
the COPY command option VAROUT on page 2-49.
Each variable-length, blocked record begins with a word that contains the
length of the record, and the read count equals the value of that length
indicator. You cannot use the TRIM option with VARIN, and this option works
only with tapes written by FUP.
out-option
controls the handling of the output file. out-option is not normally used with
structured disk files.
BLOCKOUT out-block-length
sets the number of bytes in an output block. Specify out-block-length as
an integer in the range 1 through 32,767. This value is the maximum number of
bytes to be written in a single physical operation. If a value greater than 32,767
is specified, FUP issues this error message and then aborts:
ERROR - TAPE DEFINE BLOCKLEN VALUE TOO LARGE
If you specify BLOCKOUT and out-filename is not a tape DEFINE:
• If the output-block length is greater than the output-record length, output-
record blocking occurs. The block is filled with out-record-length
records until it contains out-block-length bytes or the last output
record is encountered.
• If out-block-length is not a multiple of out-record-length, the last
record in a full block is truncated.
• If the write count for a record is less than out-record-length, the
output record is padded in the output block with trailing nulls.
• The actual number of bytes written in a physical operation is
out-block-length for all blocks (except the last). If the last block is not
full, the actual number of bytes written is equal to the number of records in
the last block multiplied by out-record-length.
If you specify the BLOCKOUT option and out-filename is a tape DEFINE
with these attributes, you must not specify the BLOCKLEN attribute of the
DEFINE, or the BLOCKLEN attribute must match out-block-length:
• LABELS set for labeled tape processing
DENSITYOUT density
indicates the recording density for the output tape file. This option applies only
to tape drives that support multiple densities.
Specify density as:
Density Recording Density
GCR or 6250 6250 bpi (bits per inch)
PE or 1600 1600 bpi
NRZI or 800 800 bpi (not used on 5130)
EBCDICOUT
specifies that output characters be translated to their EBCDIC equivalents. If
you omit EBCDICOUT, FUP does not translate output.
All characters converted between ASCII and EBCDIC retain their own image—
with these exceptions:
ASCII EBCDIC
Exclamation point Logical OR
Left square bracket Cent sign
Right square bracket Exclamation point
Circumflex Logical NOT sign
Note. A warning occurs if you specify EBCDICOUT and an XLATE option (and FUP
uses the XLATE option). For more information about the XLATE option, see
CONFIG[URE] on page 2-26.
FOLD
divides input records that are longer than out-record-length into as many
out-record-length records as needed to copy the entire input record.
If the last record written because of a FOLD is shorter than
out-record-length (and you specify PAD), padding occurs. If you omit
FOLD, truncation might occur.
PAD [ pad-character ]
specifies that records containing fewer than out-record-length bytes are
padded with the pad-character (up to the specified record length). Specify
pad-character as one of:
• A single ASCII character inside quotation marks:
"c"
• An integer in the range 0 through 255 that specifies a byte value:
{ 0:255 }
RECOUT out-record-length
sets the maximum number of bytes in an output record.
If you specify the RECOUT option, the actual number of bytes written for each
output record (the write count) depends on whether you also specify the PAD
option:
• If you do not specify PAD, the write count is either the read count or
out-record-length (whichever is less).
• If you specify PAD, the write count is out-record-length. If the number
of input bytes is less than out-record-length, the record is padded
with trailing pad characters.
[ NO ] REWINDOUT
specifies whether the tape is rewound when the COPY command finishes. If
you specify NO REWINDOUT, the tape remains positioned without rewinding.
The default is REWINDOUT. (The tape is rewound.) This option also applies to
labeled tapes.
SKIPOUT num-eofs
moves the tape past num-eofs end-of-file (EOF) marks before the data
transfer begins (for magnetic tape only). You cannot specify this option if
out-filename is a tape DEFINE with LABELS set for labeled tape
processing. You must use the appropriate tape DEFINE attribute to skip files
on a labeled tape.
TAPEMODE mode
indicates the mode in which the tape is written (for streaming drives only). The
mode can be either STARTSTOP or STREAM. If you do not specify
TAPEMODE, the default STARTSTOP mode is used for the tape operation.
STARTSTOP
selects the start-stop mode for the tape operation. STARTSTOP is the
default.
STREAM
selects the streaming mode for the tape COPY operation.
You must also specify the largest block size possible to receive the
maximum performance for streaming mode. For example, to use streaming
mode, specify BLOCKOUT 28,000.
[ NO ] UNLOADOUT
specifies whether the tape is unloaded after rewinding (for magnetic tape only).
The default is UNLOADOUT. (The tape is unloaded after it is rewound.) This
option also applies to labeled tapes.
VAROUT
writes variable-length, blocked records.
Each variable-length record is preceded by a one-word record-length indicator.
The record-length word contains the record length in bytes. Although records
can contain an odd number of bytes, the record-length word is always aligned
on a word boundary.
The write count and the record-length word of a record are equal—even if the
record is truncated. Truncation occurs if the record is either longer than
RECOUT or longer than BLOCKOUT subtracted by two. (The record-length
word requires two extra bytes.)
If the next record with its record-length word does not fit in the current block,
VAROUT terminates the current block and begins a new block because the
blocks cannot be spanned.
VAROUT terminates a block by writing a record-length word of –1 (%177777)
to indicate that no more valid records are in the block and then pads the
remainder of the physical block. VAROUT cannot write the record-length word
of –1 when the previous record ends on the block boundary or when
out-block-length is odd and only one byte remains in the block.
Although null records (records with a length of zero) are supported, VAROUT
does not allow the PAD and FOLD options.
This sample block has three records (MYSELF, COMPUTERS, and INC.) and
uses a BLOCKOUT length of %30 to illustrate VAROUT:
%000006 <- Record-length word for record 1
M Y
S E Record 1
L F
%000011 <- Record-length word for record 2
C O
M P
U T Record 2
E R
S
%177777 <- Record-length word for end of block
p p <- Padding
The third record (INC.) could not be written in the sample block because its
record-length word requires six bytes beginning on a word boundary. The
VAROUT option terminates the block (because only four bytes remain) and
writes record three to the next block.
XLATE [ xlate-table-name ]
XLATEIN [ xlate-table-name ]
XLATEOUT [ xlate-table-name ]
specifies a translation table for translating records. You can specify XLATE,
XLATEIN, or XLATEOUT and optionally includes a translation table name
(xlate-table-name). If a translation table name is not provided, FUP uses
the one defined in the most recent CONFIG[URE] XLATE command. An error
occurs if xlate-table-name is omitted and no translation tables are defined
or if the translation table name is not a defined translation table.
If XLATE or XLATEIN are specified, FUP uses the in direction of the translation
table. If XLATEOUT is specified, FUP uses the out direction of the translation
table. Every character in the IN file is translated using the specified table
before it is written to the OUT file.
Note. A warning occurs if you specify EBCDICOUT and an XLATE option (and FUP
uses the XLATE option). For more information on the XLATE option, see
CONFIG[URE] on page 2-26.
display-option
specifies the display format of the file for the Display Form of the COPY command.
Note. For descriptions of the display option and the Display Form of the COPY command,
see COPY: Display Form on page 2-55.
• A COBOL program can read multireel tape files created by FUP COPY.
• FUP COPY uses sequential block buffering when accessing in-filename. All
record locks are ignored.
• When you enter a COPY command with no options, FUP opens in-filename
with the read-only access, protected exclusion mode, unless in-filename is a
terminal. If in-filename is a terminal, FUP opens it with shared exclusion mode.
If you include the SHARE option in the command (and in-filename is a disk
file), in-filename is opened with the shared exclusion mode.
• FUP opens out-filename with the exclusive exclusion mode unless
out-filename is a terminal.
A physical read of in-filename reads one logical record. A logical record is 132
bytes for unstructured files or files in the EDIT file format. A logical record is the
length you specify for structured disk files and nondisk devices.
• A physical write to out-filename writes one logical record. The actual number of
bytes written is the number of bytes read.
• Although you can specify a block size of as many as 32,767 bytes for the
BLOCKIN and BLOCKOUT options, some peripheral devices have smaller
maximum block sizes that must not be exceeded when you use the COPY
command.
• If you copy a file that contains data records and zero-length records (empty
records) to a relative output file, all the records are written, including the zero-
length records, unless the input file is also a relative file.
If the input file is a relative file, the zero-length records are skipped unless you
specify NO COMPACT. To transfer zero-length records from a relative file, include
NO COMPACT in the COPY command.
For example, when you copy a relative file containing a combination of eight data
records and two zero-length records, the record count of the output file is eight
records instead of ten. So you lose zero-length records if you copy data with zero-
length records in and out of a relative file.
• The [NO] COMPACT option affects copy operations for relative IN files only. If you
specify NO COMPACT when you are copying another type of file, this message
appears:
WARNING - COMPACT OPTION IGNORED FOR NONRELATIVE FILES
• If the AUDIT option is set for out-filename, the copy request fails with file-
system error 75 (requesting process has no current process transaction identifier).
• Be careful when using the PAD and TRIM options in a FUP COPY or LOAD
operation. If it contains trim-character or pad-character, data can be
altered or lost.
If you pad each record in a data file with zeros to a standard size in bytes, store the
records in another file, then trim the trailing zeros when you FUP COPY or LOAD
the stored records. Any original data ending with zero is also trimmed.
To avoid this problem, use a pad-character or trim-character that is not
contained in your data.
• If you enter the DENSITYOUT parameter for a tape drive that is not a model 5106
or 5130, the COPY command continues without setting the tape density, and this
message appears on the home terminal of the FUP process:
WARNING -filename: DRIVE DOES NOT SUPPORT DENSITY
SELECTION
• If you enter the TAPEMODE parameter to a tape drive that does not support the
setting, this warning message appears on your terminal, and the COPY operation
continues without trying to set the tape mode:
WARNING - filename: DRIVE DOES NOT SUPPORT TAPEMODE
SELECTION
• The COPY command cannot copy SQL files that are not SQL object files. You
must use SQLCI COPY instead.
Note. For examples of using the COPY command with tape DEFINEs, see the Guardian
User’s Guide .
• To copy the first 500 records of MYFILE to YOURFILE (two unstructured files in the
current default subvolume), where input records are 80 bytes and the data is
copied in 800-byte physical reads:
-COPY MYFILE, YOURFILE, COUNT 500, RECIN 80, BLOCKIN 800
• To copy records from RELFILE (a relative IN file) to FILEB (FUP transfers zero-
length records and records containing data):
-COPY RELFILE, FILEB, NO COMPACT
• To output the contents of the EDIT file MYTEXT to the printer $LASER:
10> FUP / OUT $S.#LASER, NOWAIT/ COPY MYTEXT
• To copy a series of files onto magnetic tape, use a FUP COPY command for each
file and include the NO REWINDOUT option in each command except the last one:
-COPY FILE1, $TAPE, NO REWINDOUT
-COPY FILE2, $TAPE, NO REWINDOUT
-COPY FILE3, $TAPE
• To return the files copied in the previous example from tape to disk:
-COPY $TAPE, NFILE1, NO REWINDIN
-COPY $TAPE, NFILE2, NO REWINDIN
-COPY $TAPE, NFILE3
• To copy two records from the key-sequenced FILEX to FILEZ with a primary key
value specifying the starting record for FILEX:
-COPY FILEX, FILEZ, COUNT 2, FIRST KEY [ "U","P",89 ]
• To copy all the files that have changed since a specified date (01/01/01) to a
backup tape:
-COPY $MOD.*.* WHERE MODTIME > 1 JAN 2001, $TAPE
• To read records from a file (PERSONAL), translate them using a specific
translation table (MY_ENCRYPT), and copy them to another file (SECRET):
-COPY personal,secret,XLATE MY_ENCRYPT
• To give the file you are copying a title and display four lines of the file (including the
title) on the terminal screen:
-COPY FILEZ, TITLE, COUNT 4
After you enter the command, the terminal displays:
$BASE.SAMPLE.FILEZ 12MAY01 DATA MODIF: 26APR2001 14:19
This is a sample file to show how the TITLE option within FUP
COPY works.
4 RECORDS TRANSFERRED
Note. For information about making a record-by-record copy from one file to another with the
COPY command, see COPY: Copy Form on page 2-36.
To display files containing 8-bit characters, use the DISPLAYBITS option (from the
CONFIGURE command) with COPY.
Note. For more information about the DISPLAYBITS option and files with 8-bit characters, see
CONFIG[URE] on page 2-26.
To display the contents of an Enscribe or SPOOLER file, use the Display Form of the
COPY command. This subsection describes only display-option syntax. For a
complete COPY syntax description, see COPY: Copy Form on page 2-36.
display-option is:
O[CTAL]
D[ECIMAL]
H[EX]
BYTE
A[SCII]
NO HEAD
[ NO ] TITLE
display-option
specifies the display format for the file. If you omit display-option, no
formatting or conversion occurs, and each record is displayed as it appears in
in-filename.
OCTAL or O
displays in octal and ASCII format.
DECIMAL or D
displays in decimal and ASCII format.
HEX or H
displays in hexadecimal and ASCII format.
BYTE
displays in byte and ASCII format. The two bytes of each word are converted
separately.
• If you do not specify BYTE, a word is treated as a single value and
converted accordingly.
• If you specify BYTE (but not OCTAL, DECIMAL, or HEX), the display is in
byte-octal format.
ASCII or A
displays in ASCII format. This option is ignored when combined with any
preceding display-option.
NO HEAD
omits the heading preceding each record displayed.
[ NO ] TITLE
directs FUP to write a title line that includes the name of the file, the time of the
COPY process, and the last modification time of the file.
mm/dd/yy is the current date (displayed for the first record only).
hh:mm is the current time (displayed for the first record only).
offset is the offset from the beginning of the word 0 record.
word 0...word n is a block of n contiguous words of filename. If
out-record-length is less than 100, n = 7. If
out-record-length is in the range 100 through 120, n = 9;
otherwise, n = 11.
ASCII-format is the ASCII representation of the line to the left. Nonprintable
characters are represented by a period (.).
If you specify only ASCII (or A), the display is in ASCII format, and
offset is given in decimal representation.
If you specify more than one number base (OCTAL, DECIMAL, or HEX), each line is
displayed in the specified format and in ascending order of base values (O = 8,
D = 10, and H = 16).
CREATE
Creates a disk file with the current file-creation attributes defined by the SET
command. To override the current file-creation attributes defined by the SET command,
include create-param in the CREATE command.
filename
is the name of the file to be created. A partial file name is expanded using the
current default names for system, volume, and subvolume.
create-param
overrides the current file-creation attribute setting that corresponds to this creation.
The create-param used in a CREATE command does not change the file-
creation attributes defined by the SET command.
Note. For descriptions of the create-param options, see SET on page 2-166.
CREATE Guidelines
• Extent size is rounded up at creation as necessary so it can hold an integral
number of blocks.
• For files on disk drives in a disk drive enclosure, there are additional
considerations. See EXT { extent-size } { ( pri-extent-size , sec-extent-size ) } on
page 2-168.
• During the file-creation process, DP2 rounds up the extent size (to 2 pages or
4,096 bytes) because the extent size of DP2 files must always be an integral
multiple of the BUFFERSIZE (for unstructured files) or of the BLOCK size (for
structured files).
To create an unstructured DP2 file with one-page extents, you must specify a
BUFFERSIZE of 2048 bytes with either the FUP SET or FUP CREATE command.
• To take advantage of the REFRESH, DCOMPRESS, and ICOMPRESS options of
the CREATE command, you must create primary and alternate-key files separately.
The CREATE command passes these attributes to the primary file only when it is
created.
• If you enter a CREATE, AUDIT command and the volume where the primary
partition, secondary partitions, or alternate-key files reside is not audited, file-
system error 80 (invalid operation on audited file or nonaudited disk volume) is
returned. If you create an alternate-key file with the NO UPDATE option (and the
key file is not on an audited volume), the error does not occur.
CREATE Examples
• To create a non-partitioned unstructured file on a disk drive in a disk drive
enclosure:
-SET TYPE U
-SET EXT (2,20)
-SET REC 80
-CREATE TEST
Note. DP will round the extent sizes up to multiples of 14 so the actual extent information
will be ext (14,28), not ext (2,20) as specified.
• To create a key-sequenced file named MYFILE in the current default volume and
subvolume with the file-creation attributes that are specified by the SET
commands:
-SET TYPE K
-SET REC 80
-SET KEYLEN 8
-CREATE MYFILE
• If you include create-param in the CREATE command, the values you specify
override the current file-creation attributes. (For more information, see SET on
page 2-166.) To override the current attributes and create a relative file with a
record length of 10, define an alternate key on the first five bytes of the record and
have the alternate key reside in the alternate-key file SECFILE:
-CREATE MYFILE, TYPE R, REC 10, ALTKEY ("AA", FILE 0, &
-KEYLEN 5) , ALTFILE (0, SECFILE)
• To create a partitioned alternate key-sequenced file from the existing alternate key-
sequenced file, add a partition to the existing file:
1. Ensure the file is closed:
-LISTOPENS FILE
2. Set file-creation attributes to match those of the existing file:
-SET LIKE $VOLnn.Subvol.first-alternate-keyfile
3. Add the partitioning specifics:
-SET PART (1,$VOLnn, primary EXT, secondary EXT, "PR
altkeyvalue1"
-SET PART (2,$VOLnn, primary EXT, secondary EXT, "PR
altkeyvalue2"
4. Purge the existing alternate key file:
-PURGE $VOLnn.Subvol.first-alternate-keyfile
5. Create the new partitioned alternate key file:
-CREATE $VOLnn.Subvol.partitioned-alternate-keyfile
6. Load the newly created partitioned alternate key file:
-LOADALTFILE 0,$VOL.Subvol.filename, loadaltfile-command-
options
7. Restore file-creation attributes to their default settings:
-RESET
8. For each alternate key file you want to partition, repeat Steps 2 through 7.
• To create a partitioned alternate key file that is not key-sequenced from an existing
alternate key file, use the ALTER command:
-ALTER filename,PART ( sec-partition-num , [\ node.]$ volume[
, pri-extent-size [ , sec-extent-size ] ] )
• To create a partitioned alternate key-sequenced file from scratch, use the SET:
> FUP
-set type k
-set code 1001
-set ext (32,8)
-set rec 54
-set block 4096
-set keylen 2
-set altkey ("LO",keyoff 42,keylen 4)
DEALLOCATE
Deallocates any file extents beyond the one that includes the end-of-file (EOF) address
of the specified disk files. This command applies only to Enscribe files.
fileset-list
is a list of files whose unused extents (beyond the EOF extent) are to be
deallocated. If a file has the CLEARONPURGE option set, the data is physically
deleted from the deallocated extents. You can specify qualified-fileset for
this fileset-list.
PARTONLY
specifies the deallocation of any unused extents for all the primary or secondary
partitions of partitioned files in fileset-list. If you reference a primary partition
name, only the extents are deallocated from the primary partition. If you omit
PARTONLY, extents are deallocated only for entire partitioned files, the primary
partition must be specified in fileset-list, and it is an error to reference a
secondary partition in fileset-list. PARTONLY has no effect on
nonpartitioned files.
DEALLOCATE Guidelines
• DP2 does not let unused extents of audit-trail files be deallocated. Audit-trail files
are identified by file code 134. Any attempt to deallocate unused extents of audit-
trail files are rejected with file-system error 2 (operation not allowed on this type of
file). To let FUP continue after receiving such errors, use the FUP ALLOW
ERRORS command.
• DEALLOCATE works only with SQL files that are SQL object files. On other SQL
files, you must use SQLCI ALTER instead.
• If you specify the CLEARONPURGE option (with the SECURE command) for a
file, a subsequent DEALLOCATE command physically clears the data from the
deallocated extents of the file.
• To deallocate volume directory extents, you must use this file name syntax for
fileset-list:
$volume.SYS00.DIRECTRY
DEALLOCATE Example
To deallocate the unused extents past the EOF of MYFILE:
-DEALLOCATE MYFILE
DISPLAYBITS
Lets the COPY, INFO, and SHOW commands display 8-bit characters.
The DISPLAYBITS command became an option of the CONFIG[URE] command with
the D30 product version of FUP. However, for compatibility purposes, any FUP product
versions prior to the D30 RVU continue to recognize the DISPLAYBITS option as
command syntax. For more information, see CONFIG[URE] on page 2-26.
DUP[LICATE]
Copies disk files. This command applies only to Enscribe files.
You need to understand when to use the COPY, DUP[LICATE], and LOAD commands:
COPY To change file attributes or copy files to or from nondisk devices
DUP[LICATE] To create identical copies of disk files
LOAD To create a structured disk file from scratch (much faster than COPY)
rename-option is:
from-fileset-list
is a file-set list specifying the files to duplicate. FUP duplicates files one at a time.
You can specify qualified-fileset for this from-fileset-list, but any
qualifiers must occur before the to-fileset specification.
• Partial file names are expanded using the current default node, volume, and
subvolume where necessary.
• If you specify the RESTARTABLE option, you can specify only a single file in
from-fileset-list.
• If you use the wild-card or qualified-fileset option to specify the name of
a file in from-fileset-list, you must specify the file name part of
to-fileset with the (*) wild-card character.
• To duplicate a Safeguard protected file, you must have read-access to the file.
Note. For more information, see DUP[LICATE] Guidelines for Safeguard Files on
page 2-69.
to-fileset
is a file set specifying the destination of the duplication. You cannot specify
qualified-fileset for to-fileset.
• If you use a wild-card character or qualified-fileset option to specify the
name of a file in from-fileset-list, you must specify the file name part of
to-fileset with the (*) wild-card character.
• If you specify the file name part of to-fileset as an asterisk (*), each output
file is given the disk file name of its corresponding input file.
• If you specify the subvolume of to-fileset as an asterisk (*), each output
file is given the subvolume name of its corresponding input file.
• The AUDIT options of the files in to-fileset are reset regardless of the
state of that option for the files in from-fileset-list.
RESTARTABLE [ restart-filename ]
specifies that FUP create restart-filename (as an unstructured disk file with
file code 855). FUP maintains information in this file that describes the progress of
the DUP operation. If a failure occurs, the RESTART command can use this
information when continuing the operation.
restart-filename
is the name of the unstructured (informational) disk file FUP creates for the
RESTARTABLE DUP operation. The name must not be the same as any file
name in the specified subvolume, the source file name, or the destination file
name.
If you specify only the file name part for restart-filename, the name is
expanded using the current default names for volume and subvolume.
The restart file is not created until after the destination file is successfully created. If
restart file-name is the same as either a source file or a destination file or is
the same as any other file name in the subvolume, FUP issues an error 10 (file
already exists). However, if the restart file name conflicts with an existing file name
and the source file is empty, FUP creates a destination file but does not create a
restart file and does not issue an error 10.
The restart file is purged when a DUP operation is completed successfully. If a
RESTARTABLE DUP[LICATE] process fails, use the FUP RESTART command to
complete the DUP operation.
The restart file can never reside on an optical disk volume.
If you do not specify restart-filename, FUP creates a file named ZZRSTART
on your current subvolume.
If multiple RESTARTABLE DUP[LICATE] processes are running concurrently, each
RESTARTABLE DUP[LICATE] process must use a different restart-filename.
rename-option
renames the secondary-partition volume and alternate-key files when they are
duplicated. FUP creates the destination file with new names for the secondary-
partition volumes and alternate-key files.
If you do not specify rename-option, FUP creates alternate-key files and
partitions with names specified by the primary partition. This option applies only
when FUP creates the destination file. It does not apply if the OLD option is
enabled.
EXT
specifies a change to the extent sizes in the duplicate file, including the primary
(pri-extent-size) and secondary (sec-extent-size) extents. This lets you
change the physical aspect of how the file is allocated on disk.
KEEP
does not duplicate any files that exist in both to-fileset and
from-fileset-list. If a file specified in to-fileset already exists, the
corresponding file in from-fileset-list is not duplicated. FUP lists the names
of the files that are not duplicated. Checking is by name only. No file attributes are
checked.
FUP uses only one of these options: KEEP, NEW, OLD, or PURGE. If more than
one is specified, FUP uses the last one specified.
NEW
specifies that no file in to-fileset exists. For each file in
from-fileset-list, FUP creates a new file with characteristics identical to the
corresponding input file. The default mode is NEW.
FUP uses only one of these options: KEEP, NEW, OLD, or PURGE. If more than
one is specified, FUP uses the last one specified.
OLD
specifies that each file in to-fileset must already exist. The characteristics of
each file in to-fileset must match the characteristics of the corresponding file
in from-fileset-list in these areas (if applicable): file type, record size, data-
block size, data and index compression mode, and key length and key offset.
The extent size of the files in to-fileset might differ. The files you specify for
to-fileset must be large enough to contain the from-fileset data. Existing
data in to-fileset is overwritten.
FUP uses only one of these options: KEEP, NEW, OLD, or PURGE. If more than
one is specified, FUP uses the last one specified.
PURGE
purges any existing file in to-fileset that has the same name as a file in the
from-fileset. If a file currently exists with the same name as a destination file,
FUP purges the current contents of the file. FUP creates a new file, with identical
characteristics, for each file in the from-fileset-list.
FUP uses only one of these options: KEEP, NEW, OLD, or PURGE. If more than
one is specified, FUP uses the last one specified.
PARTONLY
specifies that if from-fileset-list defines any primary or secondary partitions
of a partitioned file, only those partitions (and any nonpartitioned files) are to be
duplicated. Entire partitioned files are not duplicated.
Secondary partitions defined by from-fileset-list are not duplicated unless
you specify PARTONLY.
If you omit PARTONLY and define a primary partition in from-fileset-list,
FUP duplicates the entire partitioned file (all partitions). If PARTONLY is not
specified, it is an error to specify a secondary partition in from-fileset.
PHYSVOL [ physvol ]
specifies the physical volume where a file should be created. This option overrides
any SMF parameters after the CREATE command creates a file on the virtual disk.
The specified physvol must be included in the storage pool associated with the
SMF virtual disk process.
SAVEALL
transfers the owner ID, security, and last modified timestamp of the files in
from-fileset-list to to-fileset unchanged. If you omit SAVEALL,
SAVEID, and SOURCEDATE, FUP sets the owner ID and default security of the
user who is executing the DUP command (the user ID that you used for the current
session) and sets the timestamp to the time the DUP[LICATE] process completes.
The SAVEALL option is equivalent to the SAVEID and SOURCEDATE options.
FUP uses only one of these options: SAVEALL, SAVEID, or SOURCEDATE. If
more than one is specified, FUP uses the last one specified.
Note. For more information about the SAVEALL option, see DUP[LICATE] General
Guidelines on page 2-67.
SAVEID
transfers the owner ID and security of the files named in from-fileset-list to
to-fileset with no changes. If you omit SAVEID and SAVEALL, FUP sets the
owner ID and default security to that of the user executing the DUP[LICATE]
command (the user ID that you used for the current session).
FUP uses only one of these options: SAVEALL, SAVEID, or SOURCEDATE. If
more than one is specified, FUP uses the last one specified.
SOURCEDATE
transfers the last modified timestamp of the files in from-fileset-list to
to-fileset unchanged. If you omit SOURCEDATE and SAVEALL, FUP sets the
timestamp to the time the DUP[LICATE] process completes.
FUP uses only one of these options: SAVEALL, SAVEID, or SOURCEDATE. If
more than one is specified, FUP uses the last one specified.
Note. For information about file access and exclusion modes, see the Enscribe
Programmer’s Guide.
• The FUP DUP[LICATE] command cannot duplicate SQL files except for SQL
object files. You must use SQLCI DUP instead.
• The DUP[LICATE] command duplicates a corrupt file (in its corrupt state) and
displays this warning message:
filename: FILE IS CORRUPT
• Using the DUP[LICATE] command with the EXT option does not change
MAXEXTENTS. Before you use the DUP[LICATE] command, issue an ALTER
MAXEXTENTS command to the source file if necessary to make the new file fit.
• If you use the DUP command to duplicate a primary or secondary partition file with
the OLD option, you must also specify the PARTONLY option. For example:
-DUP from-partition,to-partition,OLD,PARTONLY
If you do not specify PARTONLY with OLD for a partition, FUP displays this error
message:
ERROR - NOT ALLOWED FOR PARTITIONED FILES
• If you use the DUP command to duplicate a file that is audited by TMF, the new file
is not an audited file. To restore audited status to a file, use an ALTER AUDIT
command.
• If you use the OLD parameter, you can use the DUP[LICATE] command and
change the extent sizes. Before you use the DUP[LICATE] command, you must
create the OLD file with the new extent sizes.
• The SAVEID and SAVEALL options:
° Preserve the licensed attribute (for more information, see LICENSE (Super ID)
on page 2-116) only if the process accessor ID of the current FUP is the super
ID, and the target file resides on the node where FUP is running. If the target
file is remote, no warning message appears.
° Preserve the state of Trust flag (for more information, see TRUST on
page 2-191) only if the process accessor ID of the current FUP is the super ID
(255,255).
• You can use the DUP command on queue files. For example, to increase the size
of the queue files:
-DUP oldqueuefile, newqueuefile [EXT (pri-extent-size, sec-
extent-size)] [SAVEALL]
You cannot use DUP to increase MAXEXTENTS. To alter the MAXEXTENTS of a
file, use the FUP ALTER command.
• The maximum number of files in fileset-list that can be duplicated at a time is
2,147,483,647.
DUP[LICATE] Examples
• To duplicate all the files in the current default subvolume (and in the MYSVOL
subvolume) and place copies of the files (using the same file names) in the
NEWSVOL subvolume:
-DUP (MYSVOL.*, *), NEWSVOL.*
• To let the DUP operation on the source file (FILE1) be restarted if the initial DUP
operation does not complete successfully:
-DUP FILE1, FILE2, RESTARTABLE
If the initial DUP operation does not complete successfully, restart the previous
command:
-RESTART
• To duplicate a file in a different volume and rename the first alternate-key file (file
0) of the destination file $VOL2.RECD.ALTFILE:
-DUP $VOL1.RECD.DATA, $VOL2.*.* , ALTFILE (0, $VOL2.RECD.&
-ALTFILE)
• To duplicate a partitioned file in a different node and rename the secondary
partition of the destination file \TWO.$VOL2.SUB1.PARTFILE:
-DUP $VOL1.SUB1.PARTFILE, \TWO.$VOL1.*.*, &
-PART (1, \TWO.$VOL2)
When you duplicate a partitioned file (in a different node or volume), to duplicate a
secondary partition from a volume that does not exist on the destination node, you
must include the PART option. To duplicate the secondary partitions, use these
commands (on separate command lines):
-DUP $VOL3.SUB1.PARTFILE, \TWO.$VOL2.*.*, PARTONLY
-ALTER \TWO.$VOL1.SUB1.PARTFILE, PART (1, \TWO.$VOL2)
The DUP command duplicates only the secondary partition of PARTFILE (to
$VOL2 on node \TWO) because the PARTONLY option is specified.
ALTER changes the primary partition for the file on \TWO to point to the secondary
partition that was duplicated using PARTONLY in the first command.
• To duplicate all files on the current subvolume that end in FILE:
-DUP *FILE, NEWSVOL.*
• To duplicate all the EDIT files (that begin with an S) from one subvolume to another
and purge any old copies (if they exist):
-DUP S* WHERE FILECODE=101, NEWSVOL.*, PURGE
• To duplicate the files that have changed (since a specified date) in all the
subvolumes to a backup volume:
-DUP $MILK.*.* WHERE MODTIME>1JAN2001, $BACKUP.*.*
• To duplicate a file, but specify a change to its primary and secondary extent sizes:
-DUP SPECIALK, SUGARPOP, EXT (8,4)
EXIT
Stops the current FUP process and returns to the command interpreter.
E[XIT]
EXIT Guidelines
• To run the EXIT command, you can enter E or EXIT.
• The FUP process terminates when FUP reads the end-of-file (EOF) mark of the
input file that you specified in your command to run FUP. You do not have to end a
FUP command file with an EXIT command because of the EOF mark.
• Simultaneously pressing the CTRL and Y keys at the terminal is the same as an
EOF. If you press CTRL-Y at the FUP prompt, FUP terminates after it displays:
EOF!
EXIT Example
To terminate FUP and return control of the terminal to the command interpreter:
-EXIT
10>
FC
Modifies and reexecutes a specific command.
-num
displays a command previous to the current command. For example, to modify and
execute the third command prior to the current command, specify -3.
num
is the number of a command line. For example, to modify and execute the second
command of the current FUP session, specify the number 2.
string
is the first characters of a previous command. For example, to display the most
recent DUP command that starts with a volume name, enter FC DUP $.
quoted
is a string enclosed in either single or double quotation marks. FUP searches every
character in the command buffer—not just the first characters—until it finds the
string. For example, to display the most recent command that referenced the
system \KAUAI, enter FC “\KAUAI”.
To edit the command, use the space bar and the backspace key to position the cursor
under the text that you want to change. Do not use the arrow keys to move the cursor.
The FC command accepts these three command-editing characters:
R replacement-string
replaces characters in the command line (beginning with the character above the
R) with replacement-string. replacement-string is terminated by // or
Return.
I insertion-string
inserts insertion-string into the command line in front of the character above
the I. insertion-string is terminated by // or Return.
D
deletes the character above the D. Repeat to delete more characters.
FC Guidelines
• If you do not use parameters, FC displays the last command line entered.
• When you finish editing the line or have no changes to make, press RETURN to
execute the modified command.
• Use the HISTORY command to obtain line numbers.
• You must begin your correction explicitly with one of the subcommands if the first
character of the change is I, D, or R. Type the D or R under the character to be
deleted or replaced. Type the I under the character that follows the insert position.
• If you enter a string without a preceding subcommand, R is assumed.
• The subcommand begins its operation at the character positioned directly above it.
FC displays the modified line and prompts for another subcommand.
• Specify more than one subcommand per line by separating the subcommands with
a double slash (//).
• To abort the FC command and not execute the modified command, press the
BREAK key or the CTRL and Y keys—or enter a double slash (//) in columns 1 and
2 and immediately press RETURN.
Note. For more information about the FC command, see the Guardian User’s Guide .
FC Examples
• To change the DETAIL option in the first command to the STATISTICS option:
-FC
-INFO MYFILE, DETAIL
. ddSTAT
-INFO MYFILE, STAT
.
• Spaces typed after the I or R subcommand are part of the text to insert or replace.
To make more than one change per line by ending the text string with two slashes
(//) and spacing over:
-FC
-DUP FILE1, FILE2
. i3// r14
-DUP FILE13, FILE14
.
Commands Related to FC
COMMAND Function Page
HISTORY Displays previous FUP commands 2-80
! Reexecutes a previous FUP command without modification 2-4
? Displays a previous FUP command 2-5
FILENAMES
Lists the names of files that match the specified wild-card option. The FILENAMES
command is similar to the FILES command, but the FILES command lists all the files in
the specified subvolumes.
OUT listfile
names an existing disk file or a device to receive the listing output from the
FILENAMES command. You can use either a standard file name or a spool
DEFINE name as the OUT file name for a FILENAMES command. If listfile is
an existing file, FUP appends the output to the file.
Note. For more information about listfile, see Specifying Files on page 1-8.
fileset
specifies the file set to be listed. The fileset parameter can use the wild-card
option in both the subvolume name and the file name. If fileset is not specified,
it defaults to all files in the subvolume. You can specify qualified-fileset for
fileset.
FILENAMES Example
To display the names of all the files containing OLT in all the subvolumes of the current
volumes that start with FUPKIR:
-FILENAMES FUPKIR*.*OLT*
$BASE.FUPKIR
NOLT NOLTE OLTR ROLT XOLT XOLT2
$BASE.FUPKIRK
NOLT NOLT2 OLTE OLTR ROLT
FILES
Displays all file names associated with one or more subvolumes. The FILES command
is similar to the FILENAMES command, which displays subsets of files within a
subvolume.
subvolset:
OUT listfile
names an existing file or device to receive the output of the FILES command. You
can use either a standard file name or a spool DEFINE name as the OUT
listfile for the FILES command.
If you omit the OUT option, output is displayed on the terminal. If listfile is an
existing file, FUP appends the output to that file.
Note. For more information about listfile, see Specifying Files on page 1-8.
subvolset
names a subvolume or set of subvolumes whose file names are to be listed. If you
omit subvolset, FUP lists the files that reside in the current default volume and
subvolume. You can use a wild-card character for subvolume, but you cannot
specify qualified-fileset for subvolset. (To specify qualified-
fileset, use the FILENAMES command.)
If you omit \node, FUP lists the designated file names on the current default node.
If you omit volume, FUP lists the designated file names on the current default
volume. If you include volume, you must also include subvolume or an asterisk
(*).
If you use an asterisk (*) in place of subvolume, FUP lists all the file names on the
designated volume.
On a system running the D-series, if you omit both subvolume and the asterisk
(*), you must also omit volume. FUP lists the file names in the current default
subvolume on the current default volume.
If you include both volume and subvolume, subvolset must be a contiguous
string with $volume separated from the subvolume by a period (.).
FILES Guidelines
• If you request information for all the subvolumes on a volume, FUP displays the file
names in each subvolume (by object name within the subvolume name).
• The FILES command applies to all types of Enscribe and SQL files. File names are
displayed for unstructured and structured Enscribe files and for all types of SQL
files (tables, indexes, views, catalog tables, the indexes on SQL catalog tables,
and SQL object program files).
FILES Examples
• To list the names of the files in your current default subvolume:
-FILES
• To list the names of the files in each subvolume of the current default volume:
-FILES *
• To list the names of all the files in each subvolume on the volume $VOL2:
-FILES $VOL2.*
• To list the names of all files in the subvolume (JIMMY) of the volume ($ACES) on
the node (\TENNIS):
-FILES \TENNIS.$ACES.JIMMY
• This example assumes that the current default node is \SYS1, the current default
volume is $VOL1, and the current default subvolume is SUBVOLA. The default
subvolume also contains an Enscribe file and an SQL table, view, and index.
To list the names of all four files in the previous examples, use any of these
commands:
-FILES
-FILES SUBVOLA
-FILES $VOL1.SUBVOLA
-FILES \SYS1.$VOL1.SUBVOLA
GIVE
Changes the owner of a file. This command applies only to Enscribe files. Only the
current owner of the file (or the super ID, (255,255)) can execute the GIVE command
for a file.
GIVE fileset-list ,
{groupnum , usernum | groupname.username }
[ , PARTONLY ]
fileset-list
is a list of files whose ownership is to be given to another user. You can use wild-
card characters, and you can specify qualified-fileset for fileset-list.
groupnum , usernum
is the group and user numbers of the user who is to be given ownership of the
files.
groupname.username
is the group and user names of the user who is to be given ownership of the files.
PARTONLY
specifies (for partitioned files) that only partitions included in fileset-list are to
be given. If you omit PARTONLY, only entire partitioned files whose primary
partitions reside in fileset-list are given. PARTONLY has no effect on
nonpartitioned files.
GIVE Guidelines
• If you try to use the GIVE command to change the ownership of a file protected by
Safeguard at the disk-file level and are not the super-ID (255,255), you receive file-
system error 199 (disk file is Safeguard protected). Instead, you must use the
Safeguard command interpreter (SAFECOM).
• If you try to change ownership of a Safeguard-protected file at the subvolume level
but you do not have owner permission, you receive file-system error 48 (security
violation).
• If you try to use the GIVE command to change the ownership of a file that is not
Safeguard protected but do not have purge permission on the file or are not the
super ID (255, 255), you receive file-system error 48 (security violation).
• You cannot use the GIVE command on a file that is currently open with exclusive
exclusion mode.
• If a file is given to another user while the file is open for a different process, the
access rights of the process that has it open are not affected.
• If you give a program file whose PROGID bit is set, the GIVE command clears that
bit. To set the PROGID bit, use the FUP SECURE command. If the bit is set, the
accessor ID is set to the ID of the program file when the program is run.
• Files that you give to another user remain in their original subvolume. To move a
copy of a file to another subvolume, use FUP DUP. After you give a file to another
user, you might not be able to duplicate it because you are no longer the owner.
• GIVE can only give SQL files that are SQL object files. For other SQL files, you
must use SQLCI ALTER, CREATE, and SQLCI SECURE instead.
• If you do not own all the files in a subvolume that you specify in fileset-list,
first use the ALLOW option (within the CONFIGURE command), to set the number
of allowable errors high enough to complete the GIVE operation.
GIVE Examples
• To give ownership of all files in the current default subvolume to the user with user
ID 8,1:
-GIVE *, 8,1
• To give the files PROG1, PROG2, and LIB in the subvolume $WORK.ORG to the
user whose user ID is MANUALS.MARTIN:
-GIVE ($WORK.ORG.PROG*, $WORK.ORG.LIB), MANUALS.MARTIN
HELP
Lists the syntax of the FUP commands.
OUT listfile
names a file to receive the output of the HELP command. If you omit this option,
the output is sent to the OUT listfile that is enabled for the current FUP
session—usually your home terminal. You can use either a standard file name or a
spool DEFINE name as the OUT listfile for the HELP command. If listfile
is an existing file, FUP appends the output to the file.
Note. For more information about listfile, see Specifying Files on page 1-8.
command
is the name of a FUP command whose syntax you want to see. If you abbreviate
the command name, FUP displays help for the first command that matches the
characters entered.
ALL
lists the names of all FUP commands. The default is ALL.
ALL,SYNTAX
lists the syntax for all the FUP commands.
NEWS
provides a one-line description for each significant new FUP feature in the last
several RVUs. This informal information source does not describe syntax and does
not necessarily include every new feature. The newest edition of this manual
contains comprehensive details and syntax for new commands or features.
HELP Examples
• To display the names of all FUP commands, enter HELP ALL (or HELP):
-HELP ALL
• To write the syntax for all the FUP commands to the file MYHELP:
-HELP /OUT MYHELP/ ALL,SYNTAX
HISTORY
Displays your previous FUP commands.
OUT listfile
names an existing disk file or a device to receive the listing output from the
HISTORY command. You can use either a standard file name or a spool DEFINE
name as the OUT file name for a HISTORY command. If listfile is an existing
file, FUP appends the output to the file.
Note. For more information about listfile, see Specifying Files on page 1-8.
num
is the number of previous commands to display.
HISTORY Guidelines
• If you omit num, FUP displays the last 10 commands.
• If num is greater than the number of commands in the history buffer, FUP displays
all the commands in the buffer.
• The HISTORY command display shows line numbers for each command. You can
use line numbers in the FC, !, or ? commands. Line numbers are not displayed
anywhere else in FUP.
• The HISTORY command buffer can hold from 50 through 200 commands,
depending on the size of the commands. After the buffer becomes full, the oldest
command is discarded, as necessary, for each new command. Discarded
commands are not available from the HISTORY command, the ! command, or the
? command.
HISTORY Example
To display the last four commands entered:
-HISTORY 4
8:INFO, DETAIL
9:CREATE NEWFILE
10:DUP OLDFILE, NEWFILE
11:HISTORY 4
INFO
Displays disk file characteristics of Enscribe files; SQL/MP and SQL/MX tables,
indexes, or views; direct and SMF virtual disk files; and OSS files.
FUP INFO supports SQL/MX ANSI names. FUP converts each ANSI name to the
corresponding list of Guardian file names and then performs INFO on each of these
files.
Note. FUP support for fully qualified ANSI names for the INFO command is applicable on
H06.04 and subsequent RVUs.
FUP short INFO for a TABLE or INDEX displays the information about all partitions of
that particular table or index.
FUP short INFO for a TABLE PARTITION or INDEX PARTITION displays the
information about the partition of the table or the index.
FUP long INFO for a TABLE or INDEX displays the information about a single partition
as all the partitions contain the same information in the long form.
OUT listfile
names an existing disk file or device to receive the listing output of the FUP INFO
command. The OUT listfile defaults to your home terminal. You can use either
fileset-list
is a list of disk files for which the file characteristics are displayed. The
fileset-list can include Enscribe files, OSS files, and all types of SQL/MP and
SQL/MX files (tables, indexes, views, catalog tables, and indexes on SQL/MP and
SQL/MX catalog tables). If you omit fileset-list, the INFO command displays
characteristics for all files in the current subvolume. You can specify
qualified-fileset for fileset-list.
To display information for temporary files only, you must explicitly specify # in the
file name. The file name can contain a wild card, as in INFO #*. You can use the
asterisk (*) wild-card character alone or with other characters. For example, to
display information for all temporary files that end in 1:
INFO #*1
The command INFO *.* does not display temporary files.
ansiname-list
ansiname-list = ‘ansiname’ [, ‘ansiname’ ]...
identifies SQL/MX ANSI name tables, indexes, partitions of tables and indexes,
and any combination of these objects. A single quote ( ' ) is required to precede
and delimit each ansiname. The ANSI names syntax is in accordance with Unified
Syntax Proposal. The syntax is:
SQL-name
is used to name base SQL objects (such as tables or indexes) in addition to
their SQL containers: catalogs and schemas. The names (called 3-part names)
for SQL base objects such as tables, indexes, or modules are composed of
three SQL identifiers separated by two dot characters (for example,
CAT.SCH.T).
SQL-identifier
is a name used by SQL/MX to identify tables, views, columns, and other SQL
entities. SQL identifiers can be either regular or delimited and can contain up to
258 characters in external form, or equivalently up to 128 characters in internal
format. Regular identifiers begin with a letter (A through Z or a through z), but
can also contain digits (0 through 9), or underscore characters (_).
Regular identifiers used to name a SQL/MX module (the basic object part) can
start with the ^ character or contain the ^ character.
Note. The information regarding SQL/MX module provided above is for reference
purpose only. FUP commands do not support the MODULE keyword.
DETAIL
gives detailed information on file characteristics (including SMF information). Use
the DISPLAYBITS option (from the CONFIGURE command) with INFO,DETAIL
when the file contains alternate keys containing 8-bit characters.
Note. For more information about the DISPLAYBITS option, see CONFIG[URE] on
page 2-26.
EXTENTS
provides a listing of extent allocation by file (except SQL/MP and SQL/MX views).
If specified, views are skipped.
STAT[ISTICS]
provides all the DETAIL information and statistical data on blocks and records for
Enscribe-structured files, SQL/MP and SQL/MX tables, and indexes. Statistics
information does not appear for unstructured Enscribe files, SQL/MP and SQL/MX
program files, views, or shadow labels.
Note. The DETAIL information is not provided if the CONFIG STATONLY option is
specified. For more information, see CONFIG[URE] on page 2-26.
PARTONLY
limits the information to any partitions that you specify explicitly in
fileset-list. For example, if you specify only a secondary partition of an
SQL table, statistical information about the primary partition (or any other
secondary partition) does not appear.
PARTIAL num
specifies the percentage of the file used (read) to generate the statistics. You
specify the percentage (num) of the file (from 0 through 100) that FUP reads
and analyzes.
If PARTIAL is not specified (or if num is 0 or 100), FUP reads all the file.
Because FUP reads the file in increments of 56 KB, the actual percentage
used might be higher than the percentage requested.
USER
restricts the display to files in fileset-list owned by the user identified by
groupnum,usernum or by groupname.username. If you include USER but omit
groupnum,usernum and groupname.username, the display is restricted to the
files in fileset-list that you own.
Note. Super-group users (255, n) do not have automatic access to OSS files.
INFO Guidelines
• If a transaction is still open, INFO specifies files that were opened and closed
during the transaction as open. The files still have outstanding locks against them.
The LISTOPENS command does not specify these files as open because the files
do not have any openers.
• When a listed file is purged during the execution of the INFO command, INFO
displays Error 11 (File not in directory).
• To perform a FUP INFO,STAT on volume directories, use this list-file syntax:
$volume.SYS00.DIRECTRY
For example, the command FUP INFO $DATA01.SYS00.DIRECTRY,STAT returns
statistics on the volume $DATA01.
• FUP cannot show Safeguard protection for files protected at the volume or
subvolume level. Additional discrepancies between Safeguard protection and what
FUP displays are evident when files are copied to or from systems where the user
has default Safeguard protection on only one system.
• Although the Safeguard product might not be currently running, FUP INFO always
displays **** for the security vector of files individually protected by Safeguard.
• If you use INFO with the STAT option and FUP detects errors while generating the
statistics, you receive an error (ERR 59). The error can occur if the file is being
File Utility Program (FUP) Reference Manual—523323-012
2-84
FUP Commands INFO Listing Format
name
is the disk file name of the file whose characteristics are being displayed.
A question mark (?) after the file name might appear for an OSS file entry for which
FUP cannot read the last modification time.
open-state
is the open state of the file. It is displayed as any of:
null The file is not open, failed, or broken.
C The file is corrupt. A corrupt file is a file whose contents are in question. DUP
and LOAD mark the destination file as corrupt while these operations are
performed. If the operation does not complete normally, the file is marked
corrupt and should be purged.
O The file is open, or a TMF transaction is active on the file.
? The file is crash-open. That is, it was open when a total system failure occurred
or when the volume where it resides became unavailable.
R The file cannot be opened. Media recovery is needed (undo, redo, or
rollforward).
B The file is open but received an I/O or consistency check failure and needs
media recovery.
The states can occur in many combinations. For example:
OB The file is open but has an I/O or consistency check failure and needs a media
recovery at some point.
?B The file is crash-open and broken.
code
is the file code. Software development has reserved file codes 100 through 999 for
its own use. For a list of the file codes that are currently reserved, see Table 2-2,
System File Code Definitions, on page 2-88.
CODE 0 (zero) is the default code for user-created files. It appears as a blank in
the CODE column of the FUP INFO listing.
OSS designates OSS files.
Letters and symbols that appear after the code indicate:
A TMF audits the file.
L The file is licensed. For more information, see LICENSE (Super ID) on page 2-116.
P The PROGID attribute of the file is on. For more information, see SECURE on
page 2-162.
+ The file is a Format 2 file.
eof
is the number of bytes contained in the file.
mod
is the date that the file was last written. If the file was modified today, the date is
blank, and only the time of day is given. Otherwise, the year, month, day, and time
are given. The year field is displayed with four digits.
The word QUESTIONABLE might appear for an OSS file entry for which FUP
cannot read the last modification time.
owner
is the identification number of the file owner:
group-num , user-num
The super ID (255,255) is given as -1.
sec
is the security level assigned to the file (rwep):
r Read
w Write
e Execute
p Purge
Values for rwep are:
**** Safeguard protected (file mode only)
- Local super ID only
O Owner only (local)
G Member of owner's group (local)
A Any user (local)
U Member of owner's user class—owner only (local or remote)
C Member of owner's community—member of owner's group (local or remote)
N Any user (local or remote)
*SQL SQL/MX object
Note. For an OSS file, a 10-character OSS security vector appears in the RWEP column. (The
vector extends into the TYPE column.) For more information, see the Open System Services
User’s Guide.
Note. If an OSS file has a POSIX ACL protection, FUP INFO displays a plus sign (+) after the
permissions. However, if FUP INFO is executed remotely from a system without ACL support,
“+” will not be printed for files with optional ACL entries. This feature is supported only on
systems running G06.29 and later G-series RVUs and H06.08 and later H-series RVUs.
type
is one or more of:
null Unstructured
R Relative file structure
E Entry-sequenced file structure
K Key-sequenced file structure
tA File has alternate key
Pt File is partitioned
XPt File is an extra partition
Ta SQL/MP or SQL/MX table
In SQL/MP or SQL/MX index
PVi SQL/MP or SQL/MX PView
SVi SQL/MP or SQL/MX SView
Pg A code 100 file with SQL/MP or SQL/MX compiled objects
where t can be one or more of R, E, K, P, or X.
rec
is the logical record length of the Enscribe file in bytes. For unstructured files, this
field is blank.
bl
is the block length of the Enscribe file in kilobytes. For unstructured files, this field
is blank.
Software development has reserved file codes, 47 through 22222 for its own use.
Table 2-2 lists these codes and their corresponding definitions.
The four asterisks in the RWEP field indicate that the file NEWFILE is Safeguard
protected.
Example 2-4. Short INFO for SQL/MX Table Using ANSI Names
FUP INFO 'TABLE CAT_ANSINAME01.SCH_ANSINAME01.TAB1 PARTITION
(PART1,PART2)'
CODE EOF LAST MODIF OWNER RWEP TYPE REC BL
$DATA05.ZSDHKPKT
CGJC4500 550A+ 12288 27Oct2005 23:34 -1 *SQL PK Ta 12 4
$DATA1.ZSDHKPKT
J1C73500 550A+ 12288 17Oct2005 7:32 -1 *SQL XPK Ta 12 4
Example 2-5. Short INFO for OSS Files With POSIX ACL
$SYSTEM SYSTEM 1> fup info
\OCTOPUS.$OSS.ZYQ00001.Z0000010
CODE EOF LAST MODIF OWNER RWEP TYPE REC BL
$OSS.ZYQ00001
Z0000010 OSS 0 13:48 -1 -rw-rw-rw-+
Example 2-6. DETAIL Format for SQL Tables and Indexes and for Enscribe and
OSS Files
filename date-and-time
1- object-type
2- CATALOG catalog-name
VERSION number
BASE TABLE base-table-name
3- PHYSICAL FILENAME
VIRTUAL FILENAME
ANSI NAME ansi-name
RESOURCE FORK resource-fork-location
SYSTEM METADATA system-metadata-location
4- TYPE file-type
5- FORMAT format-code
6- CODE file-code
7- EXT ( pri-num PAGES, sec-num PAGES,
MAXEXTENTS max-extents)
8- REC record-length
PACKED REC packed-record-length
RECLENGTH max-record-length
BLOCK block-length
9- IBLOCK block-length
KEY ( key-descriptor )
SYSKEY
LOCKLENGTH lock-length
DCOMPRESS, ICOMPRESS
10- {INDEX } (key-spec, FILE alt-fnum, file-name,
{ALTKEY} key-descriptor
{ UNIQUE | NO UNIQUE }
, {UPDATE | NO UPDATE}, NULL null-value) .
11- PART ( part-num , $volume , pri-ext PAGES,
sec-ext PAGES, MAXEXTENTS max-ext,)
firstkey-value .
12- ODDUNSTR
REFRESH
AUDIT
BUFFERSIZE
BUFFERED
AUDITCOMPRESS
VERIFIEDWRITES
SERIALWRITES
13- OWNER group-id,owner-id
SECURITY (RWEP) : rwep, PROGID, CLEARONPURGE, LICENSE,
TRUSTtrust-flag,
(SUPPRESSED: rwep)
NOPURGEUNTIL: expire-time
14- SECONDARY PARTITION
15- DATA MODIF: modif, open-state
CREATION DATE: create-time
REDEFINITION DATE: redefinition-time
LAST OPEN: last-open-time
16- EOF eof(percent-used % USED)
17- FILE LABEL: num-bytes(percent-used % USED)
Note. If you do not have access privileges to a file and you issue the INFO DETAIL command,
UNAVAILABLE is displayed as the pathname.
In Example 2-6, DETAIL Format for SQL Tables and Indexes and for Enscribe and
OSS Files, on page 2-99, the headers and variables are:
1. object-type indicates whether the file is an SQL/MP or SQL/MX base table,
catalog table, index, or catalog index; Enscribe file; or Enscribe file containing an
SQL/MP or SQL/MX object program.
• INVALID indicates that an SQL/MP or SQL/MX object program is not valid and
might need to be SQL/MP or SQL/MX compiled.
• SHADOW LABEL indicates that the file is a shadow label. This file label exists
temporarily after an SQL/MP or SQL/MX object is dropped and until the
transaction is committed.
2. CATALOG identifies the catalog in which the object is defined. VERSION is the
SQL/MX software version. (For more information about SQL/MX versioning, see
the SQL/MX Database and Application Migration Guide.) If the file is an index,
BASE TABLE is the underlying table.
3. PHYSICAL FILENAME indicates a logical file by its logical name. VIRTUAL
FILENAME indicates a logical file by its physical name.
4. TYPE indicates the file organization:
K Key sequenced
E Entry sequenced
R Relative
U Unstructured
5. FORMAT is the new file’s format designator, which can have these values:
Format
Designator Indicates the File Should Be a...
1 Format 1 file as described in Handling File Formats on page 1-22
2 Format 2 file as described in Handling File Formats on page 1-22
If you omit the FORMAT option, the system decides the file format based on other
file attributes.
6. CODE is the file code. File codes are displayed for SQL tables and indexes and for
Enscribe files. The default file code of 0 is not displayed.
File codes in the range 100 through 999 refer to specific types of files and are
reserved by HP. For a description of these codes, see Table 2-2, System File Code
Definitions, on page 2-88.
Letters that follow the file code have specific meanings:
A TMF audits the file.
L The file is licensed by the super ID (255,255).
P The PROGID security attribute of the file is on.
7. Lists the sizes of the primary (pri-num) and secondary (sec-num) extents, and
the maximum number of extents that can be allocated.
8. The items in this section do not appear for unstructured files:
• REC indicates the maximum exploded record length for objects.
• PACKED REC indicates the maximum packed record length for objects.
• RECLENGTH indicates the maximum record length for relative tables.
• BLOCK indicates the length of a block.
9. Describes the primary key of a key-sequenced file or other structured file type:
• IBLOCK is the length of an index block of an Enscribe file.
• KEY key-descriptor is one or more sets of these items (the number of sets
is determined by the number of columns in the key):
COLUMN col-num, OFFSET key-offset, TYPE col-type
LENGTH key-length, {ASC }
{DESC}
° COLUMN number indicates the position of the key column in the row. If the
row contains a system-defined primary key, the primary key is column 0.
Otherwise, the first column defined for the table is column 0.
° OFFSET indicates the zero-relative byte address of the key column in the
exploded record.
° TYPE indicates the data type of the column. The data types supported by
SQL/MX are a superset of the data types supported by SQL/MP.
Note. The TYPE and LENGTH fields are not displayed for SQL/MX objects if SQL/MX
returns default values (zeroes) for these fields.
13. OWNER is the user ID of the file’s owner. This section also displays the security
string of the file, which indicates whether the PROGID, TRUST, and
CLEARONPURGE attributes are set, whether the LICENSE attribute is set, and
when you can purge the file.
TRUST trust-flag controls whether direct I/O access to user buffers is
permitted when this process is running.
NOPURGEUNTIL: timestamp (if included) indicates the expiration date set for a
file. This is the date after which you can successfully purge the file.
(SUPPRESSED: rwep) indicates the underlying security of a file protected at the
file level by Safeguard. This indicates the security the file would have if Safeguard
security were removed. For more information, see the Safeguard Reference
Manual.
14. SECONDARY PARTITION indicates the file is a secondary partition of an Enscribe
file.
15. Lists dates and times of file activity. DATA MODIF indicates when the data in the
file was last modified and one of the open states (if applicable). CREATION DATE
indicates when the file was created. REDEFINITION TIME indicates when a
change to the SQL table or index caused an SQL object program to be recompiled.
LAST OPEN indicates when the file was last open.
The modification date can be older than the file-creation date if the file was created
by duplicating it with the FUP DUP command (with the SAVEALL or
SOURCEDATE option).
The open states are:
BLANK The file is not open, failed, or broken.
BROKEN The file is open but received an I/O or consistency check failure
and needs media recovery.
CORRUPT The file is corrupt. (The contents of the file are in question.) DUP
and LOAD mark the destination files as corrupt while these
operations are being performed. If the operation does not complete
normally, the file is marked corrupt and should be purged.
DEFINITION The data or definition of the object is invalid.
INVALID
LABEL The file is in crash-label state. This state applies only to SQL
QUESTIONABLE views. A file is in the crash-label state if a file label operation was
taking place at the time of a total system failure or if the disk on
which it is located becomes unavailable.
OPEN The file is open, or a TMF transaction is active on the file.
QUESTIONABLE The file is in crash-open state. Either the file was open when a total
system failure occurred, or the volume where the file resides
became unavailable while the file was open.
REDO NEEDED The file cannot be opened, and media recovery (redo) is needed.
UNDO NEEDED The file cannot be opened, and media recovery (undo) is needed.
Note. For more information about media recovery, see the TMF Operations and Recovery
Guide.
16. For unstructured files, EOF is the end-of-file pointer containing the relative byte
address of the byte—following the last significant data byte.
For structured files, EOF is the relative byte address of the first byte of the next
available block.
If all extents were allocated, the percent-used parameter is the amount of
available file space currently used based on available space.
17. FILE LABEL is the number of bytes currently used for the file label and the
percentage of the maximum file label it uses. If this is close to 100 percent full, the
file cannot add any new extents.
18. EXTENTS ALLOCATED is the number of extents currently allocated for the file.
19. Indicates the number of index levels used for index blocks (for key-sequenced
files).
Example 2-7 shows the format used by the FUP INFO command (with the DETAIL
option) for SQL views:
The DETAIL listing format for SQL/MP views contains the same type of information as
the other DETAIL listing format. The BASE TABLE field indicates the name of the
underlying table (for protection views), and the PART and REDEFINITION DATE fields
also appear for protection views only. If the partition is on the current node, the node
name does not appear. The LABEL QUESTIONABLE and DEFINITION INVALID fields
are open states.
The DETAIL listing format for SQL/MX objects contains the same type of information
as for SQL/MP objects except:
• object-type differentiates between different SQL/MX objects including ANSI
tables, ANSI indexes, and metadata tables.
• new-option displays the location of the resource fork for an ANSI table, index, or
metadata table.
• The file format is always 2.
• The security vector is ‘*SQL.’
The DETAIL listing format for SQL/MX objects also contains:
• The ANSI name associated with the object.
Note. For examples of listings for SQL files, see the SQL/MP Reference Manual.
• To show the FUP INFO, DETAIL listing for TEMPE (a DP2 entry-sequenced file):
-INFO TEMPE, DETAIL
$VOL1.SVOL.TEMPE 15 Apr 2001, 21:03
ENSCRIBE
TYPE E
FORMAT 1
EXT (1 PAGES, 1 PAGES)
REC 80
BLOCK 4096
MAXEXTENTS 16
OWNER 1,40
SECURITY (RWEP): CUCU
DATA MODIF: 14 Apr 2001, 16:59
CREATION DATE: 10 Apr 1997, 16:00
LAST OPEN: 14 Apr 2001, 18:00
EOF 0 (0.0% USED)
FILE LABEL: 214 (5.2 % USED)
EXTENTS ALLOCATED: 0
• To show the FUP INFO, DETAIL listing for PARTFILE (a key-sequenced,
partitioned file with alternate keys):
-INFO PARTFILE, DETAIL
$VOL1.SVOL.PARTFILE 11 Nov 2000, 14:16
ENSCRIBE
TYPE K
FORMAT 1
EXT ( 1 PAGES, 1 PAGES )
REC 80
BLOCK 4096
IBLOCK 4096
KEYLEN 10
KEYOFF 0
ALTKEY ( "ab", FILE 0, KEYOFF 10, KEYLEN 10 )
ALTKEY ( "cd", FILE 1, KEYOFF 20, KEYLEN 10 )
ALTFILE ( 0, $VOL1.SVOL.AK1 )
ALTFILE ( 1, $VOL1.SVOL.AK2 )
PART ( 1, $VOL2, 1, 1, "AA" )
OWNER 8,1
• To show the FUP INFO, DETAIL listing for a logical file by its physical name:
-INFO $HEAT.ZYS00000.Z00057I0, DETAIL
$HEAT.ZYS00000.A00057I0 3 Jul 2000, 16:50
ENSCRIBE
VIRTUAL FILENAME: \SMSDEV.$BALL.CAROLS.FILE
TYPE U
FORMAT 1
EXT ( 2 PAGES, 2 PAGES )
MAXEXTENTS 16
BUFFERSIZE 4096
OWNER -1
SECURITY (RWEP): NUNU
DATA MODIF: 3 Jul 2000, 16:48
CREATION DATE: 3 Jul 2000, 14:48
LAST OPEN: NEVER OPENED
FILE LABEL: 314 (7.7% USED)
EOF: 0 (0.0% USED)
EXTENTS ALLOCATED: 0
• File attributes are set using the SET command in this example. The SHOW
command displays the results of the attributes assigned by the previous SET
commands:
-SET TYPE K
-SET KEYLEN 2
-SET ALTKEY ("AA",FILE 0,KEYLEN 2,KEYOFF 0,INSERTIONORDER)
-SET ALTFILE (0, ALT0)
-SHOW TYPE K
FORMAT 1
EXT (1 PAGES, 1 PAGES)
REC 80
BLOCK 4096
IBLOCK 4096
KEYLEN 2
KEYOFF 0
ALTKEY ("AA", FILE 0, KEYOFF 0, KEYLEN 2, INSERTIONORDER)
ALTFILE ( 0, $DATAA.DCDTEST.ALT0 )
ALTCREATE
MAXEXTENTS 16
-
Next, issue a CREATE KEY command. (FUP responds by displaying information
on the keys just created.) Then issue an INFO KEY, DETAIL command:
-CREATE KEY
CREATED - $DATAA.DCDTEST.KEY
CREATED - $DATAA.DCDTEST.ALT0
-INFO KEY, DETAIL
$DATAA.DCDTEST.KEY 17 Dec 2000, 16:45
ENSCRIBE
TYPE K
FORMAT 1
EXT (2 PAGES, 2 PAGES)
REC 80
BLOCK 4096
IBLOCK 4096
KEYLEN 2
KEYOFF 0
ALTKEY ("AA",FILE 0,KEYOFF 0,KEYLEN 2,INSERTIONORDER)
ALTFILE ( 0, $DATAA.DCDTEST.ALT0 )
MAXEXTENTS 16
OWNER 1, 164
SECURITY (RWEP): CUCU
DATA MODIF: 17 Dec 2000, 16:45
CREATION DATE: 17 Dec 1997, 16:45
LAST OPEN: NEVER OPENED
EOF 0 (0.0% USED)
FILE LABEL: 248 (6.1 % USED)
EXTENTS ALLOCATED: 0
INDEX LEVELS: 0
• To request detailed information for a Format 2 partitioned SQL/MP table:
-INFO T2, DETAIL
$DATA00.TEST.T2 19 May 2003, 17:08
SQL BASE TABLE
CATALOG $DATA00.TEST
VERSION 350
TYPE K
FORMAT 1
EXT ( 16 PAGES, 64 PAGES, MAXEXTENTS 160 )
REC 46
PACKED REC 46
BLOCK 4096
KEY ( COLUMN 0, OFFSET 0, LENGTH 4, ASC )
PART ( 0, $DATA00, 16 PAGES, 64 PAGES, MAXEXTENTS 160,
FORMAT 1, -2147483648 )
PART ( 1, $D72GB1, 16 PAGES, 64 PAGES, MAXEXTENTS 160,
FORMAT 1, 40 )
AUDIT
BUFFERED
AUDITCOMPRESS
OWNER -1
SECURITY (RWEP): NUNU
DATA MODIF: 19 May 2003, 15:25
CREATION DATE: 19 May 2003, 15:25
REDEFINITION DATE: 19 May 2003, 15:24
LAST OPEN: NEVER OPENED
EOF: 0 (0.0% USED)
EXTENTS ALLOCATED: 0
INDEX LEVELS: 0
PARTITION ARRAY FORMAT2ENABLED
• To request detailed information about ALT0:
-INFO ALT0, DETAIL
$DATAA.DCDTEST.ALT0 17 Dec 2000, 16:46
ENSCRIBE
TYPE K
FORMAT 1
EXT (2 PAGES, 2 PAGES )
REC 14
BLOCK 4096
IBLOCK 4096
KEYLEN 14
KEYOFF 0
MAXEXTENTS 16
OWNER 1,164
SECURITY (RWEP): CUCU
DATA MODIF: 17 Dec 2000, 16:45
CREATION DATE: 17 Dec 1997, 16:45
LAST OPEN: NEVER OPENED
EOF 0 (0.0% USED)
FILE LABEL: 214 (5.2 % USED)
EXTENTS ALLOCATED: 0
INDEX LEVELS: 0
• To request detailed information about all subvolumes that begin with ER:
- INFO ER*.*, DETAIL
$GUEST.ERIC.TACLCSTM 13 Aug 2000, 14:04
ENSCRIBE
TYPE U
CODE 101
FORMAT 1
EXT ( 4 PAGES, 16 PAGES )
MAXEXTENTS 16
BUFFERSIZE 4096
OWNER 96,27
SECURITY (RWEP): NUNU
DATA MODIF: 22 Jun 1999, 10:25
CREATION DATE: 22 Jun 1997, 10:25
LAST OPEN: 7 May 2000, 15:32
EOF 80 (0.0% USED)
FILE LABEL: 214 (5.2 % USED)
EXTENTS ALLOCATED: 1
$GUEST.ERD.A 13 Aug 2000, 14:04
ENSCRIBE
TYPE U
CODE 101
FORMAT 1
EXT ( 2 PAGES, 2 PAGES )
MAXEXTENTS 16
BUFFERSIZE 4096
OWNER 64,5
SECURITY (RWEP): AAAA
DATA MODIF: 27 Jun 2000, 11:04
CREATION DATE: 27 Jun 1997, 11:03
LAST OPEN: 25 Jul 2000, 18:37
EOF 40 (0.1% USED)
• To request detailed information for a partitioned SQL/MX table using ANSI names:
FUP INFO 'TABLE CAT_ANSINAME01.SCH_ANSINAME01.TAB1',DETAIL
$DATA05.ZSDHKPKT.CGJC4500 16 Nov 2005, 0:38
SQL ANSI TABLE
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
RESOURCE FORK \DRP42.$DATA05.ZSDHKPKT.CGJC4501
SYSTEM METADATA \DRP42.$DATA05.ZSD0
VERSION 1200
TYPE K
FORMAT 2
CODE 550
EXT ( 16 PAGES, 64 PAGES, MAXEXTENTS 160 )
PACKED REC 12
BLOCK 4096
KEY ( COLUMN 0, ASC )
INDEX ( 0, \DRP42.$DATA05.ZSDHKPKT.DK8CK600 )
PART ( 0, \DRP42.$DATA05.ZSDHKPKT.CGJC4500 )
PART ( 1, \DRP42.$DATA1.ZSDHKPKT.J1C73500 )
PART ( 2, \DRP42.$DATA2.ZSDHKPKT.KD973500 )
PART ( 3, \DRP42.$DATA3.ZSDHKPKT.L9G93500 )
PART ( 4, \DRP42.$DATA4.ZSDHKPKT.NQW83500 )
PART ( 5, \DRP42.$DATA05.ZSDHKPKT.P1G73500 )
PART ( 6, \DRP42.$DATA1.ZSDHKPKT.PJ183500 )
PART ( 7, \DRP42.$DATA2.ZSDHKPKT.Q6KB4500 )
PART ( 8, \DRP42.$DATA3.ZSDHKPKT.SVS83500 )
PART ( 9, \DRP42.$DATA4.ZSDHKPKT.WTK93500 )
AUDIT
BUFFERED
AUDITCOMPRESS
OWNER -1
SECURITY (RWEP): *SQL
DATA MODIF: 27 Oct 2005, 23:34
CREATION DATE: 16 Oct 2005, 20:31
REDEFINITION DATE: 16 Oct 2005, 20:31
LAST OPEN: 16 Nov 2005, 0:33
EOF: 12288 (0.1% USED)
EXTENTS ALLOCATED: 1
INDEX LEVELS: 1
PARTITION ARRAY FORMAT2ENABLED
• To request detailed information for an OSS File with POSIX ACL:
$SYSTEM SYSTEM 2> fup info \OCTOPUS.$OSS.ZYQ00001.Z0000010,
detail
$SYSTEM SYSTEM 2..
$OSS.ZYQ00001.Z0000010 11 May 2006, 13:59 OSS
PATH: /aclutils/file1
OWNER -1
SECURITY: -rw-rw-rw-+
CREATION DATE: 11 May 2006, 13:48
ACCESS TIME: 11 May 2006, 13:48
EOF: 0
level
indicates the tree level of the entry. Values for level are:
DATA Indicates that the entry is for the data level. Only this level is shown for relative
and entry-sequenced files.
A One or greater indicates an index level, and one (1) is the lowest. Index levels
are shown only for key-sequenced files.
t-blocks
is the total number of blocks in use at the indicated level.
t-recs
is the total number of records at the indicated level. At the DATA level, t-recs is
the total number of data records in the file.
a-recs
is the average number of records for each block at the indicated level.
a-slack
is the average number of unused bytes for each block at the indicated level.
a-%-slack
is the average percentage of unused bytes for each block at the indicated level.
name
is shown only if the file has extra partitions. It is the volume name of the partition
associated with the entry.
FREE t-blocks
is the total number of unused blocks in the file between the beginning of the file
and the current EOF (for key-sequenced files).
FREE t-recs
is the total number of empty records in the file between the beginning of the file
and the current EOF location (for relative files).
BITMAP t-blocks
is the number of bitmap blocks (for DP2 relative and key-sequenced files only).
2 1 14 14.0 3760 92
1 1 75 75.0 2054 50
DATA 166 4840 29.2 1363 33
FREE 0
BITMAP 1
• To display the FUP INFO, STATISTICS, PARTIAL listing for a partitioned SQL/MX
table:
-INFO WWVL3T00, STAT, PARTIAL 30
(DETAIL option listing displays first, followed by this)
***** PARTIAL STATISTICS: 30% OF FILE *****
AVG # AVG AVG %
LEVEL BLOCKS RECS RECS SLACK SLACK PART
1 1 1 1.0 4036 99
$D1103.ZSDL2BDF.WWVL3T00
DATA 1 19 19.0 3170 77
FREE 0
BITMAP 1
1 1 3 3.0 4012 98
$D1103.ZSDL2BDF.J4XL3T00
DATA 3 179 59.7 1299 32
FREE 0
BITMAP 1
• To display the FUP INFO, STATISTICS, PARTONLY listing for a partitioned
SQL/MX table using ANSI name:
-INFO 'TABLE
CAT_ANSINAME01.SCH_ANSINAME01.TAB1',STAT,PARTONLY
$DATA05.ZSDHKPKT.CGJC4500 16 Nov 2005, 0:38
SQL ANSI TABLE
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
RESOURCE FORK \DRP42.$DATA05.ZSDHKPKT.CGJC4501
SYSTEM METADATA \DRP42.$DATA05.ZSD0
VERSION 1200
TYPE K
FORMAT 2
CODE 550
EXT ( 16 PAGES, 64 PAGES, MAXEXTENTS 160 )
PACKED REC 12
BLOCK 4096
KEY ( COLUMN 0, ASC )
INDEX ( 0, \DRP42.$DATA05.ZSDHKPKT.DK8CK600 )
PART ( 0, \DRP42.$DATA05.ZSDHKPKT.CGJC4500 )
PART ( 1, \DRP42.$DATA1.ZSDHKPKT.J1C73500 )
PART ( 2, \DRP42.$DATA2.ZSDHKPKT.KD973500 )
PART ( 3, \DRP42.$DATA3.ZSDHKPKT.L9G93500 )
PART ( 4, \DRP42.$DATA4.ZSDHKPKT.NQW83500 )
PART ( 5, \DRP42.$DATA05.ZSDHKPKT.P1G73500 )
PART ( 6, \DRP42.$DATA1.ZSDHKPKT.PJ183500 )
PART ( 7, \DRP42.$DATA2.ZSDHKPKT.Q6KB4500 )
PART ( 8, \DRP42.$DATA3.ZSDHKPKT.SVS83500 )
PART ( 9, \DRP42.$DATA4.ZSDHKPKT.WTK93500 )
AUDIT
BUFFERED
AUDITCOMPRESS
OWNER -1
SECURITY (RWEP): *SQL
DATA MODIF: 27 Oct 2005, 23:34
CREATION DATE: 16 Oct 2005, 20:31
REDEFINITION DATE: 16 Oct 2005, 20:31
LAST OPEN: 16 Nov 2005, 0:38
EOF: 12288 (0.1% USED)
EXTENTS ALLOCATED: 1
PARTITION ARRAY FORMAT2ENABLED
TOTAL TOTAL AVG # AVG AVG %
LEVEL BLOCKS RECS RECS SLACK SLACK
1 1 1 1.0 4036 99
DATA 1 29 29.0 3580 87
FREE 0
BITMAP 1
0 1 239 $VOL2
1 1 293
2 1 294
3 1 297
4 1 307
• To show the FUP INFO, EXTENTS listing for a partitioned SQL/MX table:
-INFO J4XL3T00, EXTENTS
$D1103.ZSDL2BDF.J4XL3T00 27 Jan 2004, 13:39
EXTENT # OF PAGES STARTING PAGE PART
0 16 176539 \SURYA.$D1103.ZSDL2BDF.WWVL3T00
0 16 167595 \SURYA.$D1103.ZSDL2BDF.J4XL3T00
• To show FUP INFO, EXTENTS listing for a partitioned SQL/MX table using ANSI
names:
-INFO 'TABLE CAT_ANSINAME01.SCH_ANSINAME01.TAB1',EXTENTS
$DATA05.ZSDHKPKT.CGJC4500 16 Nov 2005, 0:38
EXTENT # OF PAGES STARTING PAGE PART
0 16 2020385 \DRP42.$DATA05.ZSDHKPKT.CGJC4500
0 16 5728688 \DRP42.$DATA1.ZSDHKPKT.J1C73500
0 16 336363 \DRP42.$DATA2.ZSDHKPKT.KD973500
0 16 15263976 \DRP42.$DATA3.ZSDHKPKT.L9G93500
0 16 2356668 \DRP42.$DATA4.ZSDHKPKT.NQW83500
0 16 2020401 \DRP42.$DATA05.ZSDHKPKT.P1G73500
0 16 5728736 \DRP42.$DATA1.ZSDHKPKT.PJ183500
0 16 364775 \DRP42.$DATA2.ZSDHKPKT.Q6KB4500
0 16 15263928 \DRP42.$DATA3.ZSDHKPKT.SVS83500
0 16 2356636 \DRP42.$DATA4.ZSDHKPKT.WTK93500
Note. For more information about licensing programs, see the Guardian User’s Guide .
Only the super ID can run a privileged program that is not licensed, run license
privileged programs, or use the REVOKE command to revoke the license of a
privileged program. To license files protected by the Safeguard product, use the
Safeguard command interpreter (SAFECOM).
LICENSE fileset-list
fileset-list
is a list of files to be licensed for use by nonprivileged users. You can use wild-card
characters and specify qualified-fileset for fileset-list.
LISTLOCKS
Displays information on all locks (granted or waiting) for specified Guardian file sets or
SQL/MX ANSI names.
FUP converts each ANSI name to the corresponding list of Guardian file names, then
performs LISTLOCKS on each of these files, and displays information of the locked
files only.
Note. FUP support for fully qualified ANSI names for the LISTLOCKS command is applicable
on H06.04 and subsequent RVUs.
By default, locks on all partitions of a partitioned Enscribe file are displayed. Use
LISTLOCKS to clarify lock situations. This command does not provide instantaneous
views of the locks.
OUT listfile
names a file or device to receive the listing output of the LISTLOCKS command.
You can use either a standard file name or a spool DEFINE name as the OUT
listfile. If listfile is an existing file, FUP appends output to it.
fileset-list
is a set of files for which to list existing locks (including SQL files and SQL object
files). You can use wild-card characters and can specify qualified-fileset.
ansiname-list
base-mx-object-names :: = base-mx-object-name |
( base-mx-object-name [, base-mx-object-name …] )
SQL-name
is used to name SQL base objects (such as tables or indexes) in addition to
their SQL containers: catalogs and schemas. The names (called 3-part names)
for SQL base objects such as tables, indexes, or modules are composed of
three SQL identifiers separated by two dot characters (for example,
CAT.SCH.T).
SQL-identifier
is a name used by SQL/MX to identify tables, views, columns, and other SQL
entities. SQL identifiers can be either regular or delimited and can contain up to
258 characters in external form, or equivalently up to 128 characters in internal
format. Regular identifiers begin with a letter (A through Z or a through z), but
can also contain digits (0 through 9), or underscore characters (_).
Regular identifiers used to name a SQL/MX module (the basic object part) can
start with the ^ character or contain the ^ character.
Note. The information regarding SQL/MX module provided above is for reference
purpose only. FUP commands do not support the MODULE keyword.
Note. The keywords, SYSTEM and USER, help distinguish user data from metadata.
The SYSTEM keyword can be used together only with the keyword CATALOG or
SCHEMA, to indicate the system metadata contained inside a catalog or schema. The
USER keyword can only be used with the CATALOG keyword to indicate the user
metadata contained inside a catalog. The set of tables defined by USER and those
defined by SYSTEM are mutually exclusive. The CATALOG or SCHEMA keyword
without the SYSTEM keyword, or the CATALOG keyword without the USER keyword,
indicates both the user data and the metadata.
The SCHEMA USER keyword is not supported and FUP returns an error if the parsed
ANSI name is of this type.
GRANTED
specifies to list only currently granted locks. Locks in a waiting state are not shown.
DETAIL
specifies that the internal LOCK STATE is displayed.
PARTONLY
specifies to list only the locks against the specified partition.
F G \FOXII.$:3:52:17484380
LOCK TYPE
is the type of lock:
F File
R Record
RG Generic record
STATE
is the state of the lock:
G Granted
I Internally generated intent lock
W Waiting
REQUESTER ID
is either a named or unnamed process (or a transaction ID).
KEY LEN
is the key length.
KEY/RECORD ADDRESS
is a key value for each locked record according to the type of file:
• For key-sequenced files, the key value is displayed. This value can wrap
around to the next line.
• For unstructured files, the relative byte address is displayed.
• For entry-sequenced files, the record address is displayed.
• For relative record files, the record number is displayed. The key field cannot
be longer than the generic lock length for generic key locks.
LOCK STATE
specifies the size and range of the lock.
LK^IS Intent shared. This lock is acquired for the table only. You can upgrade it to
LK^S when escalation to a table lock is required.
LK^IX Intent exclusive. This lock is acquired for the table only. You can upgrade it
to LK^X when escalation to a table lock is required.
LK^R Range check. This lock assures that the range is not protected by another
lock before an insert. It is always released after it is granted.
LK^US Unique shared. This lock is acquired to protect a single row (no range
protection).
LK^S Shared. This lock protects an entire table if granted for the table. Otherwise
it protects the row that is locked and the range between the locked row and
the row that precedes it.
LK^D Delete. This lock is acquired for the row after a deleted row. Use it to prevent
scans from skipping uncommitted deletes.
LKDUS Delete. This lock is acquired for the row after a deleted row. Use it to prevent
scans from skipping uncommitted deletes.
LKDS Delete shared. This lock is a composite of LK^D and LK^S. It is a shared
lock for the row and a delete of one or more of the rows that precede the
locked row.
LKSIX Shared, intent exclusive. This lock is acquired for a table when an LK^IX
table lock exists and a user duration LK^S table lock is required for a scan.
LK^UX Unique, exclusive. This lock is acquired to protect a single row (no range
protection).
LKDUX Delete, unique exclusive. This lock is a composite of LK^D and LK^UX. It
describes a unique exclusive lock for the row and a delete of one or more
rows before the locked row.
LK^X Exclusive. This lock is acquired to provide exclusive protection for the row
that is locked and for the range between the locked row and the row before
it.
LK^DP2 DP2 key-link (either left or right key linklock). This is not a lock but is
maintained in the same manner. A key-link is an entity that helps DP2
process efficiency and is in the output of most utilities that display lock
information.
LISTLOCKS Guidelines
• Use the REPORTWIDTH option (from the CONFIGURE command) to set the
maximum length (in columns) for a subsequent LISTLOCKS listing.
• The LISTLOCKS display for SQL files sometimes shows meaningless values in
one row; for example, a row of the number 255 (or all zeros).
LISTLOCKS Example
• To display all locks on files contained in subvolume $DATA.SUBVOL that end with
FILE:
-LISTLOCKS $DATA.SUBVOL.*FILE
FUP displays the locks listed for two files (KSFILE and ENTFILE):
\MYSYS.$DATA.SUBVOL.KSFILE
LOCK REQUESTER KEY
TYPE STATE ID LEN KEY/RECORD ADDRESS
R G \MYSYS.02,132 10 "AT1200-659"
R G \MYSYS.$MYPROG 5 "Z0072"
R W \MYSYS.01,044 5 "Z0072"
\MYSYS.$DATA.SUBVOL.ENTFILE
LOCK REQUESTER KEY
TYPE STATE ID LEN KEY/RECORD ADDRESS
R G \MYSYS.$MYPROG 477184
R G \MYSYS(162):15:002314 2502656
R W \MYSYS.01,044 11264000
F WI \MYSYS.$MYPROG
The locks in this example are grouped by locked resource. A locked resource occurs
when the locks on a file are grouped together, and the locks on a record in that file are
also grouped together.
When the file-set list is a single full volume (for example, $volume.*.*), a volume
request is made to the Guardian file system to retrieve the lock information. The lock
information is retrieved quicker (but unsorted), and FUP displays it in unsorted order.
• To display locks on SQL/MX index using ANSI names:
FUP LISTLOCKS 'INDEX CAT_ANSINAME01.SCH_ANSINAME01.IND1
PARTITION (IPART1)',DETAIL
$DATA05.ZSDHKPKT.DK8CK600
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.IND1
LOCK REQUESTER KEY
TYPE STATE ID LEN KEY/RECORD ADDRESS
F GI \DRP42(2).2.7784507
LOCK STATE = LK^IX
R G \DRP42(2).2.7784507 8 ?0 ?0 ?0 ?12 ?0 ?0 ?0 ?11
LOCK STATE = LK^UX
LISTOPENS
Lists the processes that have files open from the files specified in the command and
provides other related information.
LISTOPENS supports SQL/MX ANSI names. FUP converts each ANSI name to the
corresponding list of Guardian file names, then performs LISTOPENS on each of these
files, and displays information of the opened files only.
Note. FUP support for fully qualified ANSI names for the LISTOPENS command is applicable
on H06.04 and subsequent RVUs.
OUT listfile
names a file or device to receive the listing output of the LISTOPENS command.
You can use either a Guardian file name or a spool DEFINE name as the OUT
listfile. If listfile is an existing file, FUP appends output to that file.
fileset-list
is a list of files or devices for which opens are to be displayed—including Enscribe
files and all types of SQL/MP or SQL/MX files (except SQL/MP shorthand views).
You can use wild-card characters. You cannot specify qualified-fileset for
fileset-list. The fileset-list parameter can include device names in any
of these forms:
$device
$device.#name
[$device].#number
name
is a device name made up of alphabetic characters, wild-card characters (* or
?), or a combination of both.
number
is a temporary file name made up of numbers, wild-card characters (* or ?), or
a combination of both. If $device is not specified, it defaults to the current
volume.
ansiname-list
ansiname-list = ‘ansiname’ [ , ‘ansiname’ ]...
identifies SQL/MX ANSI Name catalogs, schemas, tables, indexes, partitions of
tables and indexes, and any combination of these objects. A single quote ( ' ) is
required to precede and delimit each ansiname. The ANSI names syntax is in
accordance with Unified Syntax Proposal. The syntax is:
Note. For more description about ansiname-list, see ansiname-list on page 2-118.
SCRATCH scratch-filename
names a file or volume to be used for temporary storage during the sorting phase.
If you omit this option, LISTOPENS uses a temporary file on the default volume.
s
is the network node number of the node running the process that has the specified
file open in the command.
c,p
is the processor number and process number of the process.
-p or -b
indicates that this is the primary or backup process (respectively) of a process pair.
a
is the access mode:
R Read
R/W Read and Write
E Execute
-e
is the exclusion mode:
S Shared
E Exclusive
P Protected
g,u
is the group ID, user ID of the process accessor ID.
sd
is the sync or receive depth specified by the process when the file was opened.
term
is the name of the home terminal of the process:
[\node.]$term
prog-name
is the program file name of the program that has the specified file open as it
appears for a user process:
$volume.subvolume.file-id
It appears for a system process as:
$SYSTEM.SYSnn.OSIMAGE
$SYSTEM.SYSnn is the subvolume containing the operating system image that is
currently in use (and nn is a two-digit octal integer.)
Note. For more information on automatic communication-path error recovery for disk files, see
the Guardian Programmer’s Guide. For a complete syntax description of the FILE_OPEN_ and
OPEN procedures, see the Guardian Procedure Calls Reference Manual.
LISTOPENS Guidelines
• To get information on temporary files, you must explicitly specify temporary file
names using the # character.
• When you apply LISTOPENS to an SQL/MP protection view, it displays the
processes that opened the view and the processes that opened the table view
depends on.
• You cannot apply LISTOPENS to an SQL/MP shorthand view. If you include a
shorthand view in fileset-list, LISTOPENS skips that view and issues a
warning message.
• An open SQL base table might not be reported by LISTOPENS as open even if a
FUP INFO command reports it is open. This situation occurs if the SQL base table
is opened indirectly by a protection view. You can issue a FUP LISTOPENS
against all protection views on the SQL base table of interest. To list the protection
views for a base table, use the command SQLCI DISPLAY USE OF tablename.
• The only device processes that currently give LISTOPENS information are disks,
terminals, and X.25 lines.
• Fields that do not contain valid data for particular files in fileset-list are blank
or zero filled.
• The information displayed in the USERID, MODE, and SD field for X.25 lines is:
° The X.25 process does not use (or keep) the accessor ID, causing it to always
return 000,000 for the accessor ID. This information appears in the USERID
field for any process that has an X.25 data communication line open.
° X.25 lines are always opened for read and write access. The display always
shows R/W in the MODE field for any process that has an X.25 data
communication line open.
° An X.25 process does not use sync depth, causing the display to always show
0 in the SD field for any process with an X.25 data communication line open.
• If the device (controlling process) does not support a LISTOPENS request, FUP
LISTOPENS displays:
WARNING - dev name: WILL NOT RETURN OPEN INFORMATION: ERR 2
• LISTOPENS does not show a file as open if it was opened and closed during a
transaction (although the transaction itself is still open). This situation occurs
because the file has no openers. The INFO command does show as open files that
have been opened and closed during a transaction (if the transaction itself is still
open). This situation occurs because the file still has outstanding locks against it.
• LISTOPENS displays the message Nonexistent process if the process that is
opening a file stops during the execution of the LISTOPENS command.
• The user running FUP must have remote access to any system, which is implicitly
referenced by the ANSI name used in the FUP command. For example, if the ANSI
name is 'CATALOG*', the user must have access to any node on which any visible
catalogs reside, and so on. If the ANSI name is 'TABLE C.S.T', the user must have
access to any node on which partitions of table C.S.T reside, and so on. Users who
want to limit the scope of the command to all SQL/MX objects on the local machine
can use a Guardian wildcard of the form: $*.ZSD*.*.
• The above explanation can be used to explain both an error 8551 from ANSI
names or error 48 from the file system, depending on the command that was used.
• LISTOPENS supports SQL/MX objects, CATALOG, SCHEMA, TABLE, INDEX, and
PARTITIONS.
LISTOPENS Examples
• To display a list of all processes that currently have open the file MYFILE (a file in
the current default volume and subvolume):
-LISTOPENS MYFILE
• To list all the opens of terminal $TERM:
5>FUP LISTOPENS $TERM
• To list all the opens of X.25 device $DEVICE.#NAME:
6>FUP LISTOPENS $DEVICE.#NAME
• To list all temporary files on the current volume beginning with 4:
-LISTOPENS #4*
• To display a list of all processes that currently have opened the SQL/MX objects:
FUP LISTOPENS 'SCHEMA CAT_ANSINAME01.SCH_ANSINAME01'
$DATA05.ZSDHKPKT.CGJC4500
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
PID MODE USERID SD MYTERM PROGRAM FILE NAME
042,02,0432 R -S 255,255 15 $ZTNT.#PTUJKRM
$SYSTEM.SYSTEM.MXCI
$DATA05.ZSDHKPKT.P1G73500
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
PID MODE USERID SD MYTERM PROGRAM FILE NAME
042,02,0432 R -S 255,255 15 $ZTNT.#PTUJKRM
$SYSTEM.SYSTEM.MXCI
$DATA1.ZSDHKPKT.J1C73500
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
PID MODE USERID SD MYTERM PROGRAM FILE NAME
042,02,0432 R -S 255,255 15 $ZTNT.#PTUJKRM
$SYSTEM.SYSTEM.MXCI
$DATA1.ZSDHKPKT.PJ183500
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
PID MODE USERID SD MYTERM PROGRAM FILE NAME
042,02,0432 R -S 255,255 15 $ZTNT.#PTUJKRM
$SYSTEM.SYSTEM.MXCI
$DATA2.ZSDHKPKT.KD973500
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
PID MODE USERID SD MYTERM PROGRAM FILE NAME
042,02,0432 R -S 255,255 15 $ZTNT.#PTUJKRM
$SYSTEM.SYSTEM.MXCI
$DATA2.ZSDHKPKT.Q6KB4500
ANSI NAME CAT_ANSINAME01.SCH_ANSINAME01.TAB1
PID MODE USERID SD MYTERM PROGRAM FILE NAME
LOAD
Loads data into a structured disk file without affecting any associated alternate-key
files. Data in the file being loaded is overwritten. This command applies only to
Enscribe files.
To load any alternate-key files, use the command LOADALTFILE on page 2-137 after
you complete a LOAD command.
load-option is:
EMPTYOK
FIRST { ordinal-record-num }
{ KEY { record-spec | key-value } }
{ key-specifier ALTKEY key-value }
PAD [ pad-character ]
in-option
key-seq-option
in-option is:
BLOCKIN in-block-length
[ NO ] COMPACT
EBCDICIN
RECIN in-record-length
REELS num-reels
[ NO ] REWINDIN
SHARE
SKIPIN num-eofs
TRIM [trim-character ]
[ NO ] UNLOADIN
VARIN
XLATE [ translation-table-name ]
XLATEIN [ translation-table-name ]
XLATEOUT [ translation-table-name ]
key-seq-option is:
MAX num-records
PARTOF $volume
SCRATCH scratch-filename
SORTED
DSLACK percentage
ISLACK percentage
SLACK percentage
You need to understand when to use the COPY, DUP[LICATE], and LOAD commands:
COPY To change file attributes or copy files to or from nondisk devices
DUP[LICATE] To create identical copies of disk files
LOAD To create a structured disk file from scratch (much faster than COPY)
in-filename
names the file containing the records to be loaded. This file can be a disk file, a
nondisk device, a process, a tape DEFINE name, an EDIT file, or a SPOOLER
(code 129) file. You cannot use wild-card characters in in-filename or specify
qualified-fileset.
destination-filename
specifies an existing disk file in which the records from in-filename are to be
loaded. You cannot use wild-card characters in destination-filename or
specify qualified-fileset for it. Any data already in
destination-filename is overwritten by the LOAD process.
EMPTYOK
accepts an empty file for in-filename. If the IN file is empty (and you do not
include the EMPTYOK option), the LOAD command terminates, and this message
appears:
ERROR - EMPTY SOURCE FILE
If you include EMPTYOK (and LOAD encounters an empty IN file), this message
appears:
RECORDS LOADED: 0
FIRST { ordinal-record-num }
{ KEY { record-spec | key-value } }
{ key-specifier ALTKEY key-value }
names the starting record of the input file for the copy. If you omit FIRST, the copy
starts with the first record of the input file.
ordinal-record-num
is the number of records (from the beginning of the file) that are to be skipped.
The first record in a file is record zero. If you specify this option for an
unstructured disk file, the copy begins at:
ordinal-record-num * in-record-length
Note. The actual reading begins with the first record in the source file.
For example, specify a key value as the ASCII string “T905”, followed by a
word containing the integer value zero, and a word containing the integer
value nine:
[ "T905", 0, 0, 0, 9 ]
key-specifier
is a one-character or two-character string (specified inside quotation marks)
specifying the alternate key to be used for positioning purposes.
ALTKEY key-value
specifies the alternate key of the starting record for disk files. FUP begins
reading the input file at the specified record. Specify key-value for key-
sequenced files according to the description of key-value in KEY { record-
spec | key-value }.
PAD [ pad-character ]
specifies that records containing fewer than in-record-length bytes are
padded with pad-character up to the record length specified in the file label.
Specify pad-character as a single ASCII character inside quotation marks:
"c"
or as an integer in the range 0 through 255, specifying a byte value:
{ 0:255 }
in-option
specifies the format and control of in-filename. The value of in-option is any
one of:
BLOCKIN in-block-length
[ NO ] COMPACT
EBCDICIN
RECIN in-record-length
REELS num-reels
[ NO ] REWINDIN
SHARE
SKIPIN num-eofs
TRIM [ trim-character ]
[ NO ] UNLOADIN
VARIN
XLATE [ translation-table-name ]
XLATEIN [ translation-table-name ]
XLATEOUT [ translation-table-name ]
Note. For a complete description of these options, see in-option on page 2-40.
key-seq-option
is an option for specifying the loading of key-sequenced files.
MAX num-records
is the number of records in in-filename. Specify num-records as a whole
number in the range 0 through (512000000 * 2048) -1. This value does not
need to be exact, but it should be equal to or greater than the actual number of
records in in-filename. FUP uses num-records to determine the size of
the scratch file used by the SORT process. The default num-records is
10,000.
If you specify SORTED, you can ignore this option.
PARTOF $volume
loads only the partition named in destination-filename. This option is for
key-sequenced, partitioned files only. The $volume is the volume containing
the primary partition of the destination file.
Note. For more information about the PARTOF option, see LOAD Guidelines on
page 2-134.
SCRATCH scratch-filename
names a file or volume to be used for temporary storage during the sorting
phase. If you omit this option, FUP uses a scratch file on the default volume.
If you specify SORTED, you can ignore this option.
SORTED
specifies that the records in in-filename are in the key-field order of the
destination file, causing FUP to not sort the in-filename records. This option
is for key-sequenced destination files only.
If you omit this option, FUP sorts in-filename records before loading
destination-filename.
These three options specify the minimum percentage of space required by index
blocks and in data blocks for future insertions. If space is not available when an
insertion is made, a block split occurs.
DSLACK percentage
sets the minimum percentage of slack space in data blocks. Specify
percentage as a value in the range 0 through 99. If you omit this option, FUP
uses the SLACK percentage value.
ISLACK percentage
sets the minimum percentage of slack space for index blocks. Specify
percentage as a value in the range 0 through 99. If you omit this option, FUP
uses the SLACK percentage value.
SLACK percentage
sets the minimum percentage of slack space in both index and data blocks.
Specify percentage as a value in the range 0 through 99. If you omit this
option, FUP does not provide slack space (SLACK 0).
LOAD Guidelines
• The input records of key-sequenced files can be in sorted or unsorted order. If you
do not specify SORTED, FUP invokes a SORT process to sort the records before
loading the destination file.
• The FUP LOAD command reads the source file directly using large buffers unless
you specify SHARE, in which case the Enscribe file system performs the reads.
• You can use a SORT DEFINE with the LOAD command. You must define it before
starting FUP. For more information, see the FastSort Manual.
• For key-sequenced files, you can specify the percentage of slack space you want
to keep for future insertions.
• If you specify the PARTOF option when loading a file that is not partitioned, you
receive one of these error messages:
° File-system error 11 (file not in directory) if the primary partition specified with
PARTOF does not exist
Note. For more information about error messages, see Section 3, FUP Messages.
° The range of keys for the different partitions is stored in the primary partition of
a partitioned file.
° If you try to load a secondary partition but you did not specify the PARTOF
option, you receive an error message.
° The FUP process loads all partitions if you do not specify PARTOF, and
destination-filename is the primary partition.
° To load only the primary partition, specify the name of the primary partition as
destination-filename, and specify the name of the primary volume for
the PARTOF option.
• To sort the in-filename records, you must have both disk space for the sort
scratch file and for destination-filename during the sorting phase.
• Use the PAD and TRIM options carefully in a FUP LOAD or COPY operation.
An example of this situation occurs if you pad each record in a data file with zeros
to a standard size in bytes and then store the records in another file. If you trim the
trailing zeros when you execute a FUP LOAD or COPY of the stored records, any
original data that ends with a zero is trimmed. To avoid this problem, use a value
for pad-character or trim-character that is not contained in your data.
• The NO COMPACT option affects only relative files. If you include NO COMPACT
in a LOAD command to load data from a nonrelative file, this message appears:
WARNING - COMPACT OPTION IGNORED FOR NONRELATIVE FILES
• When you select the COMPACT option, and the source is a relative file that
contains empty records, this message appears:
source file : EMPTY RECORD FOUND AND NOT TRANSFERRED
This message indicates that the target file has fewer records than the source file. It
is issued only once, when the first empty record is encountered.
• If the input file for the FUP LOAD operation is a relative file that contains zero-
length (empty) records (and you did not specify the NO COMPACT option), FUP
ignores the zero-length records.
This compacts records that are not zero length at the beginning of the file, and the
records lose their relativity to position and key value. To avoid this problem, always
include the NO COMPACT option to preserve the record position in relative files.
• When alternate-key records are not built because the full-alternate key does not
exist within the primary record, this message appears:
nnn RECORDS CONTAIN INCOMPLETE ALTERNATE KEY FIELDS
(ALTERNATE KEY RECORDS NOT GENERATED)
• LOAD cannot load SQL files. You must use SQLCI LOAD instead.
• You cannot use the FUP LOAD command on queue files. Using the LOAD
command can cause significant problems. FUP returns an error stating that this
action on the queue file is not permitted.
LOAD Examples
• To load data from a relative file (RELFILE) to DFILE, causing zero-length records
and data records to be transferred:
-LOAD RELFILE, DFILE, NO COMPACT
• This example loads data from the tape device ($TAPE) into the key-sequenced
disk file (KSFILE). Any records to be loaded from $TAPE are already sorted in the
order of the key fields of KSFILE. The minimum percentage of slack space to
remain in data blocks is 10 percent:
-LOAD $TAPE, KSFILE, SORTED, DSLACK 10
• To load records from a file (OLDMAST), translate them using a translation table
(MY_ENCRYPT), and write them to another file (SAVEMAST):
-LOAD oldmast,savemast, XLATE my_encrypt
LOADALTFILE
Generates (from a primary file) the alternate-key records for a designated alternate-key
file and then loads the records into the file. You can also specify the amount of slack
space reserved for future insertions. This command applies only to Enscribe files. The
LOAD command does not work on alternate-key files.
key-seq-option is:
MAX num-records
SCRATCH scratch-filename
DSLACK percentage
ISLACK percentage
SLACK percentage
key-file-number
selects the alternate-key file to load. Specify key-file-number as an integer in
the range 0 through 255 to indicate an alternate-key file of the primary file. The
alternate-key file must already exist. You can display the key-file number with the
FUP INFO command on the primary file. This number might be different from the
one used if you created the primary file with FUP.
primary-filename
names the existing primary file whose alternate-key records are to be generated
and loaded into the file indicated by key-file-number. Partial file names are
expanded using the current default node, volume, and subvolume names (if
necessary).
key-seq-option
specifies options for loading the alternate-key file.
MAX num-records
specifies the number of records to be read from primary-filename. The
default is 10,000,000. Specify num-records as a whole number in the range
0 through (512000000 * 2048) - 1. To determine the size of the scratch file that
is used by the SORT process, FUP multiplies num-records by the number of
alternate keys associated with key-file-number. The num-records value
should be equal to or greater than the actual number of records in the primary
file.
SCRATCH scratch-filename
specifies a file or volume to use for temporary storage while sorting. If you omit
this option, FUP uses a scratch file on the default volume.
DSLACK percentage
ISLACK percentage
SLACK percentage
specify the minimum percentage of space to reserve in the index and data
blocks for future insertions. If space is not available when an insertion is made,
a block split occurs.
Note. For a complete description of these options, see LOAD on page 2-130.
LOADALTFILE Guidelines
• To perform a LOADALTFILE operation, you must have read and write access to
the alternate-key files and read access to the primary file.
• LOADALTFILE ignores any NO UPDATE specifications.
LOADALTFILE honors any NULL specification defined for a key field, but an
alternate-key record is not generated for a field consisting entirely of null
characters (if such a null character was defined).
• A sort operation is performed when you execute LOADALTFILE. The primary file is
read sequentially according to its primary-key field. The alternate-key records are
generated and written to the SORT process for each record read from the primary
file. The sorted records are read from the SORT process and loaded into the
indicated alternate-key file after the sort is complete. Disk space for the sort
scratch file must exist during the sorting phase.
• You can use a SORT DEFINE with the LOADALTFILE command. You must define
it before you start FUP. For more information, see the FastSort Manual.
• If the attributes of the alternate-key file are incorrect (for example, insufficient
record length, key offset not zero, or not a key-sequenced file), LOADALTFILE fails
and displays:
ERROR - ALT FILE IS INCOMPATIBLE WITH ALT KEYS
° The full length of the alternate-key field is not contained in a specific primary
record.
° A null value was specified for the key, and the field contains only the null value.
• LOADALTFILE cannot work with SQL files. You must use SQLCI LOAD.
LOADALTFILE Example
This command generates alternate-key records from the primary file (KSFILE) and
loads the records into the alternate-key file with key-file-number 0. A minimum of 10
percent slack space remains in the data blocks of the alternate-key file:
-LOADALTFILE 0, KSFILE, DSLACK 10
OBEY
Reads commands from the specified file and executes them.
After FUP reads an EOF in the command file, it returns you to the FUP command
prompt. If the command file causes a change in status (by executing VOLUME or
CONFIGURE commands), the new status remains enabled after the completion of the
OBEY command.
OBEY filename
filename
is the name of the file containing the commands that you want to execute. The
default volume and subvolume for the file (and all files in the command file) are
derived using the standard FUP rules for defaulting any files enabled during the
execution of the command.
OBEY Guidelines
• Command files must contain ASCII text with valid FUP commands. Command files
are usually EDIT files but can be any other file type that FUP reads.
OBEY Example
To read commands from a specified file (ALLSUBS) and execute them:
1> FUP OBEY ALLSUBS
FUP is started using the OBEY command to execute FUP commands in the specified
file (ALLSUBS). This example writes to the terminal because there is no OUT file.
Control of the terminal returns to TACL after FUP executes the last command in the
command file.
PURGE
Deletes a single disk file, a set of files, many sets of files, or an entire subvolume of
files. This command applies only to Enscribe files.
!
in either or both of the displayed positions, indicates to purge the files without
prompting for permission.
If the CONFIGURE NO PROMPT PURGE option is enabled, ! is assumed
implicitly, and prompting does not occur.
If you are not running FUP interactively but are entering FUP commands through
an IN file or another process, the ! is required. If you omit !, you get a syntax
error, and FUP does not purge any files.
If you are running FUP interactively (entering FUP commands at the command
interpreter or FUP prompt) and you omit !, FUP prompts you for permission to
purge the fileset-list:
• If fileset-list is a single file (or a list of single files), the FUP INFO listing
is displayed for each file and followed by the PURGE? prompt. Type Y or y to
purge the file. If you type any other response, the file is not purged.
• If fileset-list includes an entire volume or subvolume (or if you use the
wild-card option to specify the subvolume or file ID), you are prompted for
permission to purge each file set as it is encountered in fileset-list:
Type your answer from the four choices displayed and press RETURN:
° If you type Y, y, or yes, the file set is purged with no more prompting.
° If you type N, n, or none, the file set is skipped and PURGE continues with
the remainder of the fileset-list.
° If you type S, s, or select, the FUP INFO listing is displayed for each file
in the file set, and it includes the PURGE? prompt. Type Y(es) or N(o) for
each file to specify whether to purge it.
° If you type F, f, or files, a FILES display appears for the file set, followed
by a prompt for the entire file set with the same four choices.
° If you press CTRL-Y, PURGE terminates, and the FUP prompt reappears.
° If you enter any other response, FUP ignores your response and
redisplays the file set prompt.
Note. For different responses to the file-set prompt, see PURGE Examples on
page 2-143.
fileset-list
names a file, a set of files, or many sets of files to purge. Partial file names are
expanded using the current default node, volume, and subvolume. You can use
wild-card characters and specify qualified-fileset for fileset-list.
If you are including the LISTALL or NO LISTALL option, you must place
parentheses around fileset-list. If you omit the parentheses, FUP treats
LISTALL or NO LISTALL as file names. After FUP executes PURGE, it displays the
total number of files it purged.
[NO] LISTALL
specifies whether to list the names of all files as they are purged. The listing is:
$volume.subvolume.file-id PURGED.
LISTALL is the default in interactive mode. The default for noninteractive mode is
NO LISTALL.
File Utility Program (FUP) Reference Manual—523323-012
2-141
FUP Commands PURGE Guidelines
If you include this option, you must enclose fileset-list in parentheses even if
the list contains only one file. If you omit the parentheses, FUP treats LISTALL or
NO LISTALL as file names.
If you omit the parentheses and specify NO LISTALL, FUP returns a required
delimiter is missing error message. If you omit the parentheses and specify
LISTALL, FUP searches for a LISTALL file in the current default subvolume. If no
file with that name exists, it returns file-system error 11 (file not in directory).
PARTONLY
purges only the specified partition for a partitioned file.
If you include this option, you must enclose fileset-list in parentheses even if
the list contains only one file. If you omit the parentheses, FUP treats PARTONLY
as a file name.
This option is not valid for any Enscribe object. If you use PARTONLY for Enscribe
objects, PURGE terminates abnormally (ABENDs) with an error.
PURGE Guidelines
• If you try to purge a file that has transaction-mode record or file locks pending and
it is audited by TMF, the purge request fails with file-system error 12 (file in use).
Even if the processes that opened the file no longer exists, the file-system error still
occurs.
• PURGE cannot dispose of SQL files that are not SQL object files. Instead, you
must use SQLCI PURGE.
• For objects compiled by SQL, PURGE displays error 197 (an SQL error has
occurred).
• You can purge a file only if it is not in use (that is, it cannot be open, running, or
undergoing a backup process). You also must have purge access or be logged on
as the super ID (255,255).
• If you try to purge a file before its expiration date (NOPURGEUNTIL attribute), an
error occurs. To alter the expiration date, use the ALTER command.
• If you purge a file that has the CLEARONPURGE option set (for more information,
see SECURE on page 2-162), the disk process physically deletes the file data from
the disk (overwrites it with blank data) and then deletes the file name from the
directory.
• In the LISTALL mode, PURGE lists each file name immediately after deleting it. If
you press the BREAK key while FUP is running, the PURGE command terminates
and does not delete any remaining files in fileset-list.
PURGE Examples
• To purge the file WKLYRPRT in the current default subvolume without prompting
for permission:
-PURGE WKLYRPRT !
$MYVOL.RECDS.WKLYRPRT PURGED.
1 FILE PURGED
• To purge an entire file set without prompting for permission by including the !
option:
-PURGE $VOL1.SVOL.*FILE !
$VOL1.SVOL.MYFILE PURGED.
$VOL1.SVOL.NEWFILE PURGED.
$VOL1.SVOL.OLDFILE PURGED.
3 FILES PURGED
-
• To purge an entire file set without being prompted (and without getting a list of the
files purged), include the ! and NO LISTALL options:
• To purge all files other than EDIT files that start with the letter M or later:
• To display the FUP INFO listing for each file in the specified file set and prompt for
permission to purge each file:
• To purge a file set and be prompted for permission (for the entire file set and for
individual files), omit the ! option and specify the subvolume to purge.
These examples show what happens when you choose each of the four possible
responses to the file set prompt. Each example starts with this command:
-PURGE $A.B.FILE*
DO YOU WISH TO PURGE THE ENTIRE FILESET $A.B ?
( Y[ES], N[ONE], S[ELECT], F[ILES] )?
If you type S (or press Return), the FUP INFO listing appears for each file in the file
set, followed by the PURGE? prompt:
If you type F, the contents of the file set are listed, and the original prompt appears
again:
PURGEDATA
Removes all data from a file. This command applies only to Enscribe files.
PURGEDATA does not physically purge data. It purges data logically by setting the
end-of-file (EOF) pointer to zero—the relative position to the beginning of the file. The
file still exists, its file name can be displayed (with the FILES command), and the
extents remain allocated (until you issue a DEALLOCATE command for the file).
fileset-list
is a list of files from which data is to be purged. Partial file names are expanded
using the current default node, volume, and subvolume (if necessary). You can use
wild-card characters and can specify qualified-fileset for fileset-list.
PARTONLY
purges data in any primary or secondary extents of partitioned files that reside in
fileset-list. If you omit PARTONLY, data is purged from all partitions of
partitioned files—but only if the primary partitions of the files reside in
fileset-list.
PURGEDATA Guidelines
• To use PURGEDATA and purge data from a file, you must have write access to
that file or be a super-group user (255, n).
• If you try to purge data from an audited file, TMF must be running, and the disk
containing the file must be enabled for TMF.
• For a PURGEDATA operation, the FUP process tries to open the specified file with
exclusive access. If the file cannot be opened with exclusive access, the
PURGEDATA command fails; for example, if the file is already open.
• The CLEARONPURGE option set with the FUP SECURE command has no effect
on PURGEDATA. After the PURGEDATA command is executed, the data is
physically present on the disk (but inaccessible) until it is overwritten by new data
or deallocated with the DEALLOCATE command.
• PURGEDATA cannot purge SQL files that are not SQL object files. Instead, you
must use SQLCI PURGEDATA.
PURGEDATA Example
To logically purge the data from MYFILE in the current default subvolume:
-PURGEDATA MYFILE
RELOAD
Physically reorganizes a key-sequenced file or SQL object (table or index only) while
allowing shared read and write access to the file or object.
FUP RELOAD supports key-sequenced file reorganization of volume directories, TMF-
audited files and SQL tables, and nonaudited files and SQL tables. A reload operation
improves access time and use of space for a key-sequenced file or SQL object that
has undergone many insertions, deletions, and updates with length changes.
FUP RELOAD supports SQL/MX ANSI names. FUP converts each ANSI name to the
corresponding list of Guardian file name and then performs RELOAD on these files.
Note. FUP support for fully qualified ANSI names for the RELOAD command is applicable on
H06.04 and subsequent RVUs.
OUT listfile
names a file or device to receive the output of the RELOAD command. You can
use either a standard file name or a spool DEFINE name as the OUT listfile. If
listfile is an existing file, FUP appends output to that file.
Note. For more information about listfile, see Specifying Files on page 1-8.
filename
is the name of the key-sequenced file or SQL object (table or index only) to be
reorganized. It can also be a secondary partition of a partitioned file. For more
information, see PARTOF $volume on page 2-149. You cannot use wild-card
characters or specify qualified-fileset for filename.
ansiname
identifies SQL/MX ANSI name table partition or an index partition. A single quote
( ' ) is required to precede and delimit ansiname.
SQL-name
is used to name base SQL base objects (such as tables or indexes) in addition
to their SQL containers: catalogs and schemas. The names (called 3-part
names) for SQL base objects such as tables, indexes, or modules are
composed of three SQL identifiers separated by two dot characters (for
example, CAT.SCH.T).
SQL-identifier
is a name used by SQL/MX to identify tables, views, columns, and other SQL
entities. SQL identifiers can be either regular or delimited and can contain up to
258 characters in external form, or equivalently up to 128 characters in internal
format. Regular identifiers begin with a letter (A through Z or a through z), but
can also contain digits (0 through 9), or underscore characters (_).
Regular identifiers used to name a SQL/MX module (the basic object part) can
start with the ^ character or contain the ^ character.
[ NO ] DEALLOCATE
the default (DEALLOCATE) lets you compact a file and delete all space beyond the
new EOF.
If you select NO DEALLOCATE, the FUP process does not deallocate any extent
beyond the new EOF that is set by RELOAD.
NEW
specifies to perform a new reload operation on filename even if there is a
suspended reload for filename. If NEW is not specified (and a previous reload for
filename was suspended or stopped), FUP restarts the previous operation from
the point where it stopped.
NEW does not override a reload operation in progress. If NEW is specified and a
reload for filename is in progress, FUP displays an error message.
PARTOF $volume
specifies the volume where the primary partition resides if filename is an
Enscribe secondary partition. If you specify a secondary partition but do not specify
the PARTOF option, the system returns an error message.
The PARTOF option is not required for an SQL object. If you specify this option for
an SQL object, FUP ignores it.
RATE percentage
specifies that the reload operation should spend percentage of its time executing
the reorganization and the rest of its time delaying. Specify percentage as an
integer in the range 1 through 100. The default value is 100.
If you are restarting a reload operation and do not specify a new rate value, FUP
uses the value from the previous reload.
DSLACK percentage
sets the minimum percentage of slack space in data blocks. Specify percentage
as a value in the range 0 through 99. If you omit this option, FUP uses the SLACK
percentage value.
ISLACK percentage
sets the minimum percentage of slack space in index blocks. Specify percentage
as a value in the range 0 through 99. If you omit this option, FUP uses the SLACK
percentage value.
SLACK percentage
sets the minimum percentage of slack space in both index and data blocks. Specify
percentage as a value in the range 0 through 99. If you omit the SLACK
parameter (and the DSLACK and ISLACK parameters), FUP uses a default value
of zero for Enscribe files and 15 for SQL objects. If you are restarting a reload that
has been suspended, FUP uses the slack values from the previous reload
operation.
The SLACK value overrides any DSLACK and ISLACK values. If you specify an
ISLACK or DSLACK value (and then specify a SLACK value), FUP uses the
SLACK value and ignores the DSLACK or ISLACK value.
SHARE
allows a RELOAD operation on a file to occur concurrently with update operations
being executed by other applications or allows a user to open a file after a
RELOAD operation has been initiated if the user does not have write permission
for the file.
Note. These steps can be performed only on the ORSERV object, which has symbols
information.
TACL>TAL ;suppress
*INT PROC CHECK^RELOAD^FLAG; BEGIN RETURN 1; END;
*|EOF
TACL>BIND
*ADD * FROM orserv-path
*REPLACE * FROM OBJECT
*BUILD ORSERV!
*|EOF
TACL>PURGE OBJECT
If the newly built ORSERV object is used for a RELOAD operation, the SHARE
command works even without write access.
If RELOAD is performed with the SHARE value, the default DEALLOCATE is
ignored, and a warning is returned.
If RELOAD is performed with the SHARE value and suspended, it still has the
SHARE attribute when it is restarted.
If RELOAD is performed with the SHARE value and the requisite DP2 is not
present, error 49 is returned.
RECLAIM
The RECLAIM option does not perform a RELOAD, or re-organize the file. It
reclaims the unclaimed free space for a SQL object caused by a Data Definition
Language (DDL) move partition boundary or one-way split operation.
This option works only with the required DP2 option installed in the system. A
suspended RELOAD with the RECLAIM option automatically takes the RECLAIM
value when RELOAD is restarted.
The RECLAIM option is not valid for any Enscribe object. If RELOAD is performed
with the RECLAIM value on any Enscribe object, error 2 appears.
This value is available only in G-series RVUs.
The free space is reclaimed on a subsequent RELOAD without the RECLAIM
option.
The RECLAIM option can be used to clear the F-flag on an SQL table that has
"UNRECLAIMED FREE SPACE" due to a DDL "move partition boundary" or "one-
way split" operation.
The RECLAIM option directs DP2 to mark data blocks vacated due to a DDL
"move partition boundary" or "one-way split" operation as FREE blocks.
The RECLAIM completes very quickly, but the EOF of the file may not be reduced
after a RECLAIM.
If the original goal of the partition split was to free space within the source partition
for future data insertion, the RECLAIM option is sufficient.
If the original goal of the partition split was to obtain maximum reduction of EOF,
omit the RECLAIM option which will compress the file data and eliminate FREE
blocks.
If the original goal of the partition split was to reduce disk space consumption by
the source partition and free up space for OTHER files on the disk, the RECLAIM
option must NOT be used. Use the DEALLOCATE option to return any freed
extents back to the disk volume's free space pool.
When a RELOAD with RECLAIM is in progress, a STATUS command for that file
will indicate RECLAIM IN PROGRESS and RECLAIM: YES
RELOAD Guidelines
• RELOAD must be used with a C30 or later product version of TMF. Using RELOAD
with earlier product versions of TMF might produce new information in the file that
is unrecoverable.
• RELOAD does not require files to be audited.
• If reload a volume directory during a peak production period, run RELOAD at a low
rate to reduce the degradation of system performance caused by the RELOAD.
• D-series software supports allocation (with the ALLOCATE command, for Enscribe
only) and deallocation of volume directory extents, with the [NO] DEALLOCATE
parameter.
This deallocates unused extents when you issue the RELOAD command against a
volume directory. If you want unused extents in a directory, use the ALLOCATE
command after each reload.
• When you issue a RELOAD command, FUP creates an online reload server
(ORSERV) process (on the node of the file) to perform the reload. If FUP cannot
create the ORSERV process, an error message appears.
• The ORSERV process created by a RELOAD command runs in the NOWAIT state.
While the reload operation is processing, FUP displays its hyphen (-) prompt so
you can perform other FUP operations.
• The reload operation on an audited file generates audit records that describe the
movement of data within the file. The audit record total from a file can triple the file
length (discounting free space).
• A reload operation can degrade system performance. To control the processor time
used by the reload, use the RATE parameter. Any percentage below 100 prevents
the reload from monopolizing the processor and its resources.
• To prevent system performance degradation, do one of:
° Increase the TMF audit-trail file configuration before you use FUP RELOAD.
RELOAD Example
-RELOAD PAYFILE, NEW, RATE 40
-RELOAD 'TABLE CAT_ANSINAME01.SCH_ANSINAME01."TABLE5" PARTITION
(PART1)'
RELOCATE
Moves files on SMF virtual disks from one physical volume to another within a storage
pool. This command is available only to the file owner or a super-group user (255,n).
logical-set
specifies a set of logical files where relocation is to occur. You can use wild-card
characters or specify a qualified file set. The logical-set can also include SQL
files.
physvol
specifies the name of a physical volume.
nnn
specifies the priority of a job (1 through 199). Changing this option lets you run
RELOCATE at a lower priority if you do not want it to interfere with other jobs.
SOURCEDATE
allows the preservation of the original timestamp of the DATAMODIF attribute of a
file. It works when the user is SUPER.SUPER or owner of the file. When relocating
files on a virtual drive without the SOURCEDATE option, the original time stamps
are not retained.
RELOCATE Guidelines
• If multiple files are specified in the logical-set, FUP relocates them one at a
time (waited).
• If a file in logical-set is not an SMF file or if physvol is not a member of the
SMF storage pool that the virtual disk is associated with for each file in logical-
set, it is an error.
• Errors encountered during this command are displayed by the normal FUP error-
handling mechanisms and terminate the command or not according to the normal
FUP ALLOW rules.
• If a file in logical-set is an SQL/MP base table with protection views, the
protection views are also moved. The protection views do not have to be in the file-
set list.
• If a file in the file-set list is an SQL/MP protection view, it is not moved unless the
base table it is defined on is also moved.
• If the PHYSVOL option is not specified, SMF chooses the best volume on the
virtual disks.
RELOCATE Example
To relocate the files starting with BLUE from $L.SMS to the physical volume $ABC:
RENAME
Changes the file name or subvolume name of a disk file or renames sets of files. This
command applies only to Enscribe files.
You can also use RENAME to reorganize your Enscribe files by renaming one or more
files in one or more subvolumes to a single subvolume. If the destination subvolume
does not already exist, the renaming process creates it. To transfer files from one
volume to another, you must use FUP DUP or FUP COPY—RENAME cannot do it.
old-fileset-list
specifies the files to be renamed. You can use wild-card characters and can
specify qualified-fileset for fileset-list.
For example, *SF finds all the files that end with SF. You can use ?SF to find all
files that start with one character and are followed by SF. The files in
old-fileset-list can reside on more than one subvolume.
new-fileset
specifies the new names of the files. You cannot specify qualified-fileset for
fileset, and you must specify the disk file-name portion of new-fileset as an
asterisk (*) to rename more than one file. If you do this, each file is given the disk-
file name of its corresponding input file, and the new subvolume name is taken
from new-fileset. The files in new-fileset must reside on one subvolume.
PARTONLY
specifies that only partitions included in old-fileset-list are to be renamed
(for partitioned files). If you omit PARTONLY, FUP renames all the partitions for
partitioned files if their primary partitions reside in old-fileset-list.
PARTONLY has no effect on files that are not partitioned.
RENAME Guidelines
• If you try to rename a file whose AUDIT attribute is set for auditing by TMF, the
rename request fails, and you receive file-system error 80 (invalid operation on
audited file or nonaudited disk volume).
• You cannot rename a file that is currently open with exclusive access.
• To rename a file, you must have purge access to the file or be the super ID
(255,255). To change a file’s contents, you must have write access to the file.
• RENAME cannot work with SQL/MP or SQL/MX files that are not object files.
• If you do not own all the files in old-fileset-list, use the ALLOW option
(from the CONFIGURE command) with SEVERE ERRORS set to a high number.
This lets the RENAME command finish. The command can rename only files that
you own.
RENAME Example
To change the subvolume name of all files in the current default subvolume (and all
files in the SVOL subvolume) to NEWSVOL, having all files retain their original file
names:
-RENAME (SVOL.*, *), NEWSVOL.*
REPORTWIDTH
Sets the maximum length (in columns) for FUP to format its output. This command
changes the normal (default) 132-character output format of these FUP commands to
a smaller output format:
• COPY with a DUMP option
• FILES
• LISTLOCKS
• SUBVOLS
The REPORTWIDTH command became an option of the CONFIG[URE] command
with the D30 product version of FUP. However, for compatibility, FUP product versions
starting at D30 continue to recognize the REPORTWIDTH option as command syntax.
For more information, see CONFIG[URE] on page 2-26.
RESET
Restores one or more file-creation attributes to the default settings. For a list of file-
creation attributes and their default values, see SET on page 2-166.
reset-opts is:
create-spec is:
ALTCREATE
ALTFILE [ key-file-number ]
ALTFILES
ALTKEY [ key-specifier ]
ALTKEYS
AUDIT
AUDITCOMPRESS
BLOCK
BUFFERED
CODE
COMPRESS
DCOMPRESS
EXT
FORMAT
ICOMPRESS
KEYLEN
KEYOFF
MAXEXTENTS
ODDUNSTR
PART [ partition-num ]
PARTONLY
PARTSREC
PHYSVOL
QUEUEFILE
REFRESH
SERIALWRITES
TYPEBUFFERSIZE
VERIFIEDWRITES
config-simple-opts
config-simple-option [ , config-simple-option ]
config-simple-option is:
ALLOW
DISPLAYBITS
DISPLAYALLNUM
ECHO [CONFIG[URE] | OBEY
IOTIMEOUT
NETBLOCKSIZE
PROMPT [ PURGE ]
REPORTWIDTH
RESTARTUPDATE
STATONLY
XLATE [ xlate-table-name ]
XLATEIN [ xlate-table-name ]
XLATEOUT [ xlate-table-name ]
COPY [ copy-option ]
DUP[LICATE] [ dup-option ]
LOAD [ load-option ]
LOADALTFILE [ loadaltfile-option ]
RELOAD [ reload-option ]
RESET Guidelines
• Each create-spec keyword in a RESET command resets the corresponding
creation attribute to its default setting. If you do not include create-spec in your
RESET command, FUP resets all creation attributes to their default settings.
• If you specify any of the plural keywords (ALTKEYS, ALTFILES, or PARTS), FUP
resets all the corresponding creation values (alternate keys, alternate-key files, or
secondary partitions).
Note. For the create-spec parameter default values, see SET on page 2-166. For
config-simple-opts, see CONFIG[URE] on page 2-26.
RESET Examples
• This example assumes you set the file-creation attributes to create a relative
structured DP2 file $COMPUTR.BOOKS.PASCAL (which is not displayed in this
example) with a record size of 10 bytes and an alternate-key file (named
$COMPUTR.BOOKS.BLAISE). The SHOW command displays:
SET LIKe $COMPUTR.BOOKS.BLAISE
TYPE R
EXT ( 1 PAGES, 1 PAGES )
FORMAT 1
REC 10
BLOCK 1024
ALTKEY ("aa", FILE 0, KEYOFF 0, KEYLEN 5)
ALTFILE (0, $COMPUTR.BOOKS.BLAISE)
ALTCREATE
MAXEXTENTS 16
Reset the record size file-creation attribute to the default value of 80 bytes:
-RESET REC
After you complete the reset, the FUP SHOW command includes this line in its
display:
REC 80
• This example uses a single RESET command with no create-spec to restore all
the defaults. This example includes the MAXEXTENTS and BUFFERSIZE DP2 file
attributes:
-RESET
-SHOW
TYPE U
EXT ( 1 PAGES, 1 PAGES )
FORMAT1
MAXEXTENTS 16
BUFFERSIZE 4096
• To reset all the DUP[LICATE] options for the CONFIG[URE] command:
-RESET CONFIGURE DUP
RESTART
Restarts a RESTARTABLE DUP operation at the point where it failed. The operation
continues from near the point where it failed.
RESTART [ restart-filename ]
restart-filename
is the name of an unstructured disk file created by a previous DUP operation with
the RESTARTABLE option specified. The file contains information describing the
progress of the operation.
If restart-filename is not specified, FUP searches your current subvolume for
a file named ZZRSTART.
RESTART Guidelines
• FUP (DUP) updates the restart file periodically. This causes a RESTART operation
to start from the last update, which is not necessarily from the last record written.
• The RESTART operation fails if any of these conditions exist:
RESTART Examples
• This example assumes that the DUP operation from the first command fails. The
RESTART command restarts the DUP operation.
-DUP FILE1, FILE2, RESTARTABLE
-RESTART
If a RESTARTABLE DUP operation fails and you issue a RESTART command,
FUP always echoes the original DUP command in uppercase.
A restart-filename is not included in the RESTARTABLE option of the DUP
command, so FUP searches for the ZZRSTART file on the current subvolume for
the information it needs to perform the RESTART operation.
• To include a restart-filename with the RESTARTABLE DUP command:
-DUP OLDFILE, NEWFILE, RESTARTABLE RSFILE
.
.
-RESTART RSFILE
If the DUP operation fails, the RESTART command must specify the name of the
RSFILE to request FUP to use the information contained in the unstructured file
(file code 855) RSFILE for the RESTART operation.
secure-option is:
CLEARONPURGE
PARTONLY
PROGID
fileset-list
is a list of files whose licenses or other attributes are to be revoked. Partial file
names are expanded using the current default node, volume, and subvolume (if
necessary). You can use wild-card characters and specify qualified-fileset
for fileset-list.
secure-option
is one of three options that can be set by a program or by a SECURE command.
For more information, see SECURE on page 2-162.
CLEARONPURGE
physically deletes all data within fileset-list from the disk (by overwriting
the file space with blank data) when the file is purged.
Including CLEARONPURGE in a REVOKE command revokes the
CLEARONPURGE option for the file. When you purge a file that does not have
CLEARONPURGE set, the disk space is logically deallocated. This option has
no effect on a PURGEDATA operation.
PARTONLY
specifies that only the designated partition is affected by this REVOKE
command (for partitioned files). If you omit PARTONLY, every partition is
PROGID
sets the process accessor ID to the owner ID of the program file when the
program file is run (for program files only). Including PROGID in a REVOKE
command revokes the PROGID option for the file. The process accessor ID is
set to the ID that corresponds to the creator of the process when the program
is run.
° To revoke an attribute of the primary partition only. Otherwise, FUP revokes the
attribute for all partitions of the file.
• REVOKE cannot work with SQL files other than SQL object files.
SECURE
Sets or changes the standard security attributes of a file. You must secure files
protected by the Safeguard product through its command interpreter (SAFECOM). This
command applies only to Enscribe files.
To execute the SECURE command, you must be the owner of the file you want to
secure, or you must log on as the super ID (255,255).
security is:
secure-option is:
fileset-list
is a list of files whose security attributes are to be set or changed. Partial file
names are expanded using the current default node, volume, and subvolume (if
necessary). You can use wild-card characters and specify qualified-fileset
for fileset-list.
security-string
sets new Guardian file security for the files in fileset-list. You can also
enclose security-string with quotation marks. Specify security-string as
a literal four-character string. You can include the ? character, but you must always
supply all four (rwep) security characters:
rwep
The four characters in security-string assign these new values for file
security:
rwep
r Read access allowed
w Write access allowed
e Execute access allowed
p Purge access allowed
security-num
is an integer encoding of the file security. You cannot specify the (?) character
attribute in security-num form. You can specify security-num in octal notation
(% indicates octal notation) as in:
%ijkkkk
i 1 if the PROGID option is to be set; 0 if it is not.
j 4 if CLEARONPURGE is to be set; 0 if it is not.
You can also specify the decimal equivalent of the octal number. See the PROGID
and CLEARONPURGE options in this syntax description. You cannot specify the ?
character attribute in security-num form.
The kkkk sets the values for read, write, execute, and purge, respectively. k can
be any of:
0 Any local user
1 Member of owner's group
2 Owner
4 Any local or remote user
5 Member of owner's community
6 Member of owner's user class
7 Local super ID (255,255) only
A standard user can secure a file (by using security-num) to allow only super ID
(255,255) access (%7777). After doing this, the user does not have access to the
file and must ask the super ID to change the security of the file.
Note. The SECURE command does not change the file security if you omit
security-string and security-num.
CLEARONPURGE
physically deletes all data in fileset-list from the disk (by overwriting the file
space with blank data) when the file is purged (or deallocated) with the
DEALLOCATE command. When you purge a file that does not have
CLEARONPURGE set, the disk space is logically deallocated, and the data is not
physically destroyed.
CLEARONPURGE does not affect the PURGEDATA command, but it does affect
the DEALLOCATE command. If the CLEARONPURGE option is specified in a
SECURE command for a file, a subsequent DEALLOCATE command physically
clears the data from the deallocated extents of the file.
PARTONLY
changes security only for the designated partition (for partitioned files). If you omit
PARTONLY, security for every partition of the file is affected if the primary partition
of the file is included in fileset-list. If the primary partition is not included in
fileset-list, only the referenced file partitions are affected.
PROGID
is the program ID for program files only. When the program is run, the PROGID
option sets the process accessor ID to the ID of the owner of the program file. This
option is not valid if you are on a remote system.
SECURE Guidelines
• To secure Safeguard protected files, unless you are a super ID (255,255), you
must use the Safeguard command interpreter (SAFECOM). If you try to use FUP
to secure a Safeguard protected file, you receive file-system error 199 (disk file is
Safeguard protected).
• The owner, group manager, or super ID (255,255) can secure a file. Remote
members of the owner’s user class can also secure files if they have read and
purge access to them. If you try to use FUP to secure a file without having read
and purge security access to it, you receive file-system error 48 (security violation).
• You cannot secure a file that is open with exclusive access.
• If a process has a file open when you secure it, the access rights of the process
are not affected until the process closes the file.
• FUP SECURE changes the security of existing files. When you create a new file,
its security is defined by your default Guardian file security. To change it, use the
DEFAULT command at the TACL prompt.
• FUP SECURE cannot secure SQL files other than SQL object files. To change the
security of other SQL files, use the SQLCI SECURE command.
• If you use SECURE against SQL-compiled objects, you receive error 197 (an SQL
error has occurred).
• To reset security attributes, use the REVOKE command.
• The security-string cannot begin with a hyphen (-). To set the read access to
the local super ID (255,255) only, specify a security-num (for example, %7jjj).
SECURE Examples
• To change the security for MYFILE (a file in the current default subvolume) to let
any local user read the file (but only the owner can write, execute, or purge the
file):
-SECURE MYFILE, "AOOO"
• To make the same change, but use the numeric notation to designate the security:
-SECURE MYFILE, %0222
• To change the security for MYFILE to let any local or remote user read the file, but
keep all the remaining security attributes:
-SECURE MYFILE, N???
• To secure MYPROG, permit only the owner to read, write, and purge it (although
any local user can execute it), and set the PROGID bit so the owner ID of
MYPROG is used as the process accessor ID when the program is run:
-SECURE MYPROG, OOAO, PROGID
• To set the network security for the file $OFFICE.BILLS.PAPER so any local or
remote user can read the file, only the local group members can execute the file,
and only the local owner can write or purge the file:
-SECURE $OFFICE.BILLS.PAPER, NOGO
• To resecure all the files owned by WRITE.KIRK with NNNN security:
-SECURE * WHERE OWNER=WRITE.KIRK, NNNN
SET
Changes one or more file-creation default attributes before you create files. You can
specify parameter values explicitly or set them to match those of an existing file.
To display the current file-creation attributes (the values for create-param), use the
SHOW command. To restore create-param values to the default settings, use the
RESET command.
[ NO ] AUDIT
CODE file-code
EXT { extent-size }
{ ( pri-extent-size , sec-extent-size ) }
FORMAT formatcode
LIKE filename
[ NO ] REFRESH
TYPE file-type
BLOCK data-block-length
REC record-length
[ NO ] AUDITCOMPRESS
[ NO ] BUFFERED
[ NO ] COMPRESS
[ NO ] DCOMPRESS
[ NO ] ICOMPRESS
KEYLEN key-length
KEYOFF key-offset
MAXEXTENTS maximum-extents
QUEUEFILE
[ NO ] SERIALWRITES
[ NO ] VERIFIEDWRITES
[ NO ] ALTCREATE
ALTFILE ( key-file-number , filename )
ALTKEY ( key-specifier { , altkey-param }... )
FILE key-file-number
[ NO ] INSERTIONORDER
KEYLEN key-length
KEYOFF key-offset
NO NULL
NULL null-value
[ NO ] UNIQUE
[ NO ] UPDATE
BUFFERSIZE unstructured-buffer-size
ODDUNSTR
PHYSVOL [ physvol ]
[ NO ] AUDIT
designates the file as audited or nonaudited by TMF. The default is NO AUDIT.
[ NO ] AUDITCOMPRESS
sets the mode of producing audit-checkpoint messages (using compressed or
entire messages) for audited files. The default is NO AUDITCOMPRESS.
[ NO ] BUFFERED
sets the mode of handling write requests to the file using buffered or write-through
cache. A buffered cache is specified with BUFFERED. A write-through cache is
specified with NO BUFFERED. The default is BUFFERED for audited files and NO
BUFFERED for nonaudited files.
Caution. If you use the buffered-cache option on a DP2 file that is not audited by TMF, a
system failure or disk-process takeover can cause the loss of buffered updates to the file. An
application program might not detect this loss (or handle the loss correctly) unless it is modified
to do so.
CODE file-code
sets the file code. Specify file-code as an integer in the range 0 through 65,535.
The default is zero. File codes 100 through 999 are reserved for use by HP.
Note. For a list of the reserved file codes, see Table 2-2 on page 2-88.
EXT { extent-size }
{ ( pri-extent-size , sec-extent-size ) }
sets the extent size—including primary and secondary extent sizes (if applicable).
Extents should be at least as large as the block size (for structured files). If they
are not, multiple extents are allocated every time a block is required. The extent
size must be an integral multiple of the unstructured buffer size (for unstructured
DP2 files).
The default value for extent-size and pri-extent-size is one page (2,048
bytes). If you specify a value of zero or do not specify a value, sec-extent-size
defaults to the pri-extent-size value. When creating files, you cannot define
primary or secondary extents as zero pages.
For unstructured files on a disk drive in a disk drive enclosure, the pri-extent-
size and the sec-extent-size must both be divisible by 14. DP2
automatically rounds the size up if the file is non-partitioned.
You can specify these values for extent-size, pri-extent-size, and
sec-extent-size:
0:maximum [ PAGE[S] ]
specifies the extent size in pages (2,048-byte units). Possible values of
maximum are:
Format 1 Format 2
0:65,535 [ PAGE[S] ] 0:512,000,000 [ PAGE[S] ]
Because the minimum extent size is one page (2,048 bytes), one page is
also allocated if you specify zero extents.
Note. If you specify an extent size over 65,535 pages, you must assign Format 2
to your files.
0:maximum BYTE[S]
specifies the extent size in bytes. Possible values of maximum are:
Format 1 Format 2
0:134,215,680 BYTE[S] 0:2,147,483,647 BYTE[S]
The FUP process rounds the extent size up to the next full page. If you
specify 2,047 bytes, FUP allocates one page. If you specify 2,049 bytes,
FUP allocates two pages.
0:maximum REC[S]
specifies the extent size based on the current settings for
record-length (REC), data-block-length (BLOCK),
index-block-length (IBLOCK), key-field lengths, and compression
settings. Possible values are:
Format 1 Format 2
0:134,215,680 REC[S] 0:2,147,483,647 REC[S]
The FUP process rounds the extent size up to the next full page.
0:maximum MEGABYTE[S]
specifies extent sizes in million-byte units. Possible values are:
Format 1 Format 2
0:134 MEGABYTE[S] 0:2,147 MEGABYTE[S]
The FUP process rounds the extent size up to the next full page.
FORMAT format-code
is the format designator for the new file. The designator can have these values:
0 The system decides the format of the file based on the values of other attributes,
such as block size. The system chooses Format 2 when the partition size is over
2 GB minus 1 MB. For files that are not key-sequenced, the total size is 4 GB or
more.
1 The file should be Format 1. For more information, see Handling File Formats on
page 1-22.
2 The file should be Format 2. For more information, see Handling File Formats on
page 1-22.
LIKE filename
sets file-creation attributes to match those of an existing file. FUP expands a partial
file name using the current default values for node, volume, and subvolume (if
necessary). If you specify a secondary partition of a partitioned file in filename,
FUP automatically sets the PARTONLY create-param.
File Utility Program (FUP) Reference Manual—523323-012
2-169
FUP Commands SET Parameters for All File Types
MAXEXTENTS maximum-extents
sets the maximum number of extents to be allocated (for nonpartitioned files or
key-sequenced partitioned files). Specify maximum-extents as an integer from
16 through n, where n is a maximum value determined by the amount of free
space in the file label. The FUP process rounds any value you set (from 1 through
15) up to 16. The absolute maximum is 978 extents, and the default is 16.
It is not always possible to allocate all the extents specified by
maximum-extents. The actual number of extents that can be allocated depends
on the amount of space in the file label. If there are alternate keys or partitions, the
maximum number of extents allowed is less than 978. If you specify
MAXEXTENTS, you must also consider the primary and secondary extent sizes to
avoid exceeding the maximum file size.
When the primary and secondary extent size (plus the specified MAXEXTENT
size) is larger than two gigabytes, a CREATE operation rejects the request with
file-system error 21 (illegal count specified).
[ NO ] REFRESH
causes the file label to be copied to disk whenever the file control block is marked
as dirty. This situation occurs if the end of file or a free-space block changes. If the
only change is updating the LAST MODIFIED field, the file label is not written to
disk. The default is NO REFRESH (the label is not copied to disk).
[ NO ] SERIALWRITES
sets the mode of writes to the mirror: serial or parallel. The default is NO
SERIALWRITES.
TYPE file-type
sets the file type. The default is U. Values for file-type are:
U or 0 Unstructured file
R or 1 Relative file
E or 2 Entry-sequenced file
K or 3 Key-sequenced file
[ NO ] VERIFIEDWRITES
sets the mode of file writes: verified or unverified. The default is NO
VERIFIEDWRITES.
BLOCK data-block-length
sets the data-block length. Specify data-block-length as an integer that is a
power of 2 from 512 bytes through 4096 bytes (512, 1024, 2048, 4096).
The default data-block-length is 4096 bytes. Blocks should not be larger than
extents. If this situation occurs, multiple extents are allocated every time a block is
required.
When you are setting file-creation attributes, you can SET BLOCK
data-block-length to any value within the valid range for this parameter, but
FUP rounds it up to a DP2 block size.
You might need to decrease the number of 1 KB cache blocks and increase the
number of 4 KB cache blocks (in the DP2 configuration) if there is significant use of
the default block size (4096 bytes).
During the file-creation process, DP2 rounds up the extent size (to 2 pages or
4,096 bytes) because the extent size of DP2 files must always be an integral
multiple of the BUFFERSIZE (for unstructured files) or of the BLOCK size (for
structured files).
REC record-length
sets the record length. Specify record-length as an integer in the range 1
through:
Maximum Value for Relative Maximum Value for
File Format and Entry-Sequenced Files Key-Sequenced Files
1 Blocksize - 24 Blocksize - 34
2 Blocksize - 48 Blocksize - 56
The default setting for record-length is 80 bytes, and the maximum record
length is reduced by one byte when data compression is used.
[ NO ] COMPRESS
sets or clears the states of key compression in both index and data blocks. The
default setting is NO COMPRESS. The key offset must be 0 for key compression
in data blocks, and the maximum record size is reduced by one byte.
[ NO ] DCOMPRESS
sets or clears key compression for data blocks. The default setting is NO
DCOMPRESS. The key offset must be 0 for key compression, and the maximum
record size is reduced by one byte.
[ NO ] ICOMPRESS
sets or clears key compression in index blocks. The default setting is NO
ICOMPRESS.
KEYLEN key-length
sets the primary-key length. Specify key-length as an integer in the range 1
through 255. To create key-sequenced file structures, you must specify KEYLEN,
or the creation attempt fails.
KEYOFF key-offset
sets the primary-key offset. Specify key-offset as an integer in the range 0
through 2034. The default setting for key-offset is 0.
QUEUEFILE
sets the queue file attributes. Queue files are a special type of key-sequenced file
with timestamps automatically added by DP2. Processes can queue and dequeue
records in a queue file. Queue files contain variable-length records that are
accessed by values in designated key fields. Unlike other key-sequenced files,
queue files have primary keys but cannot have alternate keys. The primary key for
a queue file includes an 8-byte timestamp. You can optionally add a user key. The
disk process inserts the timestamp when each record is inserted into the file and
maintains the timestamp during subsequent file operations.
[ , partial-key-value ] ] ] )
PART
sets secondary partition specifications for partitioned files. Specify each secondary
partition separately.
sec-partition-num , \node.$volume
names the volume where this secondary partition is to reside. Specify
sec-partition-num as an integer from 1 through 15 to designate the
secondary partition. Specify node and volume as the names of the node and
volume to contain the partition. The file name and the subvolume of the
primary partition are specified when the file is created.
Although FUP lets you specify any number in the range of 1 through 15 for
sec-partition-num, FUP changes it to a standard DP2 number that starts
at zero when the file is created.
Note. Example 2-6, DETAIL Format for SQL Tables and Indexes and for Enscribe and
OSS Files, on page 2-99 shows the DP2 number as listed in the INFO DETAIL
command.
pri-extent-size , sec-extent-size
sets the primary and secondary extent sizes. The default value for
pri-extent-size is one page (2,048 bytes). If you specify a value of zero or
do not specify a value, sec-extent-size defaults to the
pri-extent-size value. When altering files, you cannot define primary or
secondary extents as zero pages.
For partitioned unstructured files where one or more partitions reside on a disk
drive in a disk drive enclosure, DP2 has additional restrictions. You must
specify both the pri-extent-size and the sec-extent-size so that they
can be explicitly divisible by 14. DP2 does not automatically round the size up.
You can specify these values for pri-extent-size and
sec-extent-size:
0:maximum [ PAGE[S] ]
specifies the extent size in pages (2,048-byte units). Possible values are:
Format 1 Format 2
0:65,535 [ PAGE[S] ] 0:512,000,000 [ PAGE[S] ]
Because the minimum extent size is one page (2,048 bytes), one page is
also allocated if you specify zero extents.
0:maximum BYTE[S]
specifies the extent size in bytes. Possible values are:
Format 1 Format 2
0:134,215,680 BYTE[S] 0:2,147,483,647 BYTE[S]
The FUP process rounds the extent size up to the next full page. If you
specify 2,047 bytes, FUP allocates one page. If you specify 2,049 bytes,
FUP allocates two pages.
0:maximum REC[S]
specifies the extent size based on the current settings for
record-length (REC), data-block-length (BLOCK),
index-block-length (IBLOCK), key-field lengths, and compression
settings. Possible values are:
Format 1 Format 2
0:134,215,680 REC[S] 0:2,147,483,647 REC[S]
The FUP process rounds the extent size up to the next full page.
0:maximum MEGABYTE[S]
specifies extent sizes in million-byte units. Possible values are:
Format 1 Format 2
0:134 MEGABYTE[S] 0:2,147 MEGABYTE[S]
The FUP process rounds the extent size up to the next full page.
partial-key-value
specifies the lowest key value that can reside in this partition (for key-
sequenced files only). You must include partial-key-value for each
partition of a key-sequenced file. The partial-key value is 0 for the primary
partition.
Specify partial-key-value as any one (or a combination) of:
• A string of characters enclosed in quotation marks:
"c1c2...cn"
• A list of single characters (with quotation marks around each character)
separated by commas:
"c" , "c" , "c" ...
• Integers representing byte values from 0 through 255 enclosed in brackets:
[ 5, 8, 220 ]
To specify partial-key-value correctly:
• If the partial-key values are a string of alphabetic characters, enclose them
in quotation marks. For example, enter A, B, and C as a string of
characters:
"ABC"
• If the partial-key values include a string of integers only, each item is
separated by a comma, and the entire string must be enclosed in brackets:
[ 1,2,3 ]
[ NO ] PARTONLY
specifies whether subsequent file creations create all partitions of a partitioned file
(NO PARTONLY) or a single partition (PARTONLY). The default setting is NO
PARTONLY.
If you specify PARTONLY while a PART specification is enabled, any file you
create is designated as a primary partition. If a PART specification is not enabled,
any file you create is designated as a secondary partition. When you create the
file, you must specify the file name of the (primary or secondary) partition.
[ NO ] ALTCREATE
sets or clears the automatic alternate-key file creation. The alternate-key files are
created (if you specify ALTCREATE) when you create the primary file. The default
setting is ALTCREATE.
If you are deleting the last alternate-key specification, you must also delete the
corresponding alternate-file specification in the same SET command.
If you are deleting the last alternate-file specification, you must also delete the
corresponding alternate-key specification in the same SET command.
key-specifier
is a 2-byte value that identifies this alternate-key field. Specify
key-specifier as either a one-character or two-character string in quotation
marks:
"[c1]c2"
Or specify it as an integer from -32,768 through 32,767:
{ -32,768 : 32,767 }
You can use any characters for key-specifier except zero. If you omit c1,
then c1 is treated as a zero.
Note. If you add a new key specifier that references an undefined key-file number,
you must include the ALTFILE option to define the alternate-key file.
altkey-param
specifies attributes of the alternate-key file.
FILE key-file-number
sets the key-file number for key-specifier. Specify key-file-number
as an integer from 0 through 255. ALTFILE create-param relates this
number to an actual file. The default is zero.
[ NO ] INSERTIONORDER
specifies whether or not insertion-ordered alternate-key sequencing is to
be used. The default is NO INSERTIONORDER. The default specifies
alternate key records of files with duplicate key values ordered by their
primary key sequence and not their order of insertion.
An insertion-ordered alternate key cannot share an alternate-key file with
other keys of different lengths or with other keys that are not insertion-
ordered.
All nonunique alternate keys of a file must have the same duplicate-key
ordering attribute. A file with this specification must not have both insertion-
ordered alternate keys and standard (duplicate ordering by primary key)
nonunique alternate keys.
KEYLEN key-length
sets the key length for key-specifier. You must specify a KEYLEN to
create a key-sequenced file, or the creation attempt fails.
KEYOFF key-offset
sets the key offset for key-specifier. The default setting for
key-offset is zero.
Note. For information about null values, see the Enscribe Programmer’s Guide.
[ NO ] UNIQUE
specifies whether or not key-specifier is set as a unique key. The
default is NO UNIQUE.
[ NO ] UPDATE
specifies whether or not automatic updating is set for the alternate-key file
represented by key-specifier.
The NO UPDATE option prevents the file system from automatically
updating the specified alternate-key file when you write to the main file.
Although you usually keep alternate-key files synchronized with their main
files, you might leave files unsynchronized in rare circumstances. For
example, you could have two files pointing to the same alternate-key file
but only want updates from one of the two written to it.
The default is UPDATE.
BUFFERSIZE unstructured-buffer-size
is the internal buffer size to use when accessing the specified file (for unstructured
files only). To set the BUFFERSIZE file attribute, use this command (FUP SET) or
the FUP ALTER and FUP CREATE commands. Possible values for
unstructured-buffer-size are (in bytes):
FUP rounds the actual buffer size up to the nearest valid DP2 block size.
During the file-creation process, DP2 rounds up the extent size (to 2 pages or
4,096 bytes) because the extent size of DP2 files must always be an integral
multiple of the BUFFERSIZE (for unstructured files) or of the BLOCK size (for
structured files).
To create an unstructured DP2 file with one-page extents, you must specify a
BUFFERSIZE of 2048 bytes with either the FUP SET or FUP CREATE command.
ODDUNSTR
changes an even unstructured file to an odd unstructured file.
Unstructured Enscribe files can be even or odd. The FUP process rounds up any
odd byte count that you give to an even unstructured file (for reading, writing, or
positioning). This is the default for unstructured files.
FUP does not round up odd unstructured files. You always read, write, or position
at the byte count you give. To change an odd unstructured file to an even
unstructured file, copy the odd file into a new file that was created as even
unstructured.
For unstructured files created on 512 byte sector disk enclosure, the primary and the
secondary extent values are rounded to a mod-14 number by DP2, and the
MAXEXTENTS parameter may be adjusted accordingly.
An extent page is 2 KB. When a non-partitioned unstructured file is created within or
migrated to the 512 byte sector disks, the extent sizes are rounded up to a mod-14
boundary (14 blocks of 2 KB size). The actual space allocated on disk includes a 4 KB
check block.
For each 28 KB of data of unstructured files on 512 byte sector disks, one 4 KB check
block with the checksum data is allocated. For extent management, check blocks are
added to the file size and extents are aligned to include check blocks. For I/O
management, all unstructured I/Os are in units of 32 KB (28 KB user data and 4 KB
check block).
PHYSVOL [ physvol ]
specifies the physical volume where a file should be created. This option overrides
any SMF parameters after the CREATE command creates a file on the virtual disk.
The value of physvol specified must be included in the storage pool associated
with the SMF virtual disk process.
SET Guidelines
• The SET command sets the file-creation attributes only for the current session of
FUP. Each time you start a new FUP session (when you type FUP at the command
interpreter prompt), the attributes are reset to their default values. You must use
FUP in the interactive mode to use the SET and SHOW features.
• The REFRESH, DCOMPRESS, ICOMPRESS, and COMPRESS attributes are not
passed to alternate-key files when you create the primary-key file. To set these
attributes, include the NO ALTCREATE option and create the alternate-key files in
a separate operation.
• If you use the PHYSVOL option with the SET command, it applies to each
subsequent CREATE command until it is reset. An error occurs during a CREATE
command if the PHYSVOL specified does not belong to the volume specified in the
CREATE command.
• The extent size is rounded up when you create a file according to these rules:
° The extent size must be a multiple of the block size (for DP2 structured files).
° The extent size must be a multiple of the unstructured buffer size (for DP2
unstructured files).
Note. For information on extent-size rounding, see the Enscribe Programmer’s Guide.
Table 2-3 summarizes the upward rounding that can occur at file creation. Block and
buffer sizes are in bytes, and extent sizes are in pages. (The default extent size is one
page.) Although you can specify different sizes for primary and secondary extents,
equal sizes are displayed to simplify the table.
SET Examples
• To create a non-partitioned unstructured file on a disk drive in a disk drive
enclosure:
-SET TYPE U
-SET EXT (2,20)
-SET REC 80
-CREATE TEST
Note. DP rounds the extent sizes up to multiples of 14 so the actual extent information will
be ext (14,28), not ext (2,20) as specified.
• If you set the BUFFERSIZE for an unstructured DP2 file, FUP rounds the buffer
size up to the next DP2 block size. Suppose you set the buffer at 30 bytes:
-SET BUFFERSIZE 30
The SHOW command indicates the buffer size is 512 bytes:
-SHOW
TYPE U
FORMAT 1
EXT ( 1 PAGES, 1 PAGES )
MAXEXTENTS 16
BUFFERSIZE 512
• To set file-creation attributes for a key-sequenced file with 50-byte records and a
primary-key length of 36 bytes:
-SET TYPE K, REC 50, KEYLEN 36
• To create the file $L.SMS.BLUE on the physical volume $ABC, use the PHYSVOL
option of the SET command (which subsequently resets the PHYSVOL option):
-SET PHYSVOL $ABC
CREATE $L.SMS.BLUE
RESET PHYSVOL
SHOW
Displays the current settings of the file-creation attributes. Use the FUP SET and FUP
RESET commands to set and reset these attributes.
show-opts is:
create-spec is:
TYPE
CODE
FORMAT
EXT
REC
BLOCK
COMPRESS
DCOMPRESS
ICOMPRESS
KEYLEN
KEYOFF
ALTKEY [ key-specifier ]
ALTKEYS
ALTFILE [ key-file-number ]
ALTFILES
ALTCREATE
ODDUNSTR
REFRESH
AUDIT
PART [ partition-num ]
PARTS
PARTONLY
PHYSVOL
MAXEXTENTS
BUFFERSIZE
BUFFERED
AUDITCOMPRESS
VERIFIEDWRITES
SERIALWRITES
QUEUEFILE
configure-opts is:
config-simple-opts
config-simple-option [ , config-simple-option ]
config-simple-option is:
ALLOW
DISPLAYBITS
DISPLAYALLNUM
ECHO [CONFIG[URE] | OBEY
IOTIMEOUT
NETBLOCKSIZE
PROMPT [ PURGE ]
REPORTWIDTH
RESTARTUPDATE
STATONLY
XLATE [ xlate-table-name ]
XLATEIN [ xlate-table-name ]
XLATEOUT [ xlate-table-name ]
COPY [ copy-option ]
DUP[LICATE] [ dup-option ]
LOAD [ load-option ]
LOADALTFILE [ loadaltfile-option ]
RELOAD [ reload-option ]
OUT listfile
names an existing disk file or a device to receive the listing output from the SHOW
command. You can use either a standard file name or a spool DEFINE name as
the OUT listfile for a SHOW command. If listfile is an existing file, FUP
appends the output to the file.
Note. For more information about listfile, see Specifying Files on page 1-8.
create-spec
specifies which file attributes you want to show. If you do not specify any attributes,
FUP shows all the current settings applicable to the current file type.
Note. For create-spec parameter default values, see SET on page 2-166. For
config-simple-opts, see CONFIG[URE] on page 2-26.
AS COMMANDS
causes FUP to insert the word CONFIGURE in front of the CONFIG[URE] options. If
this option is present, it can only occur immediately after the CONFIG[URE]
keyword, and you must specify both words. If this display is written to a file, you
can use it to reestablish CONFIG[URE] options from a command file or FUPCSTM
file.
SHOW Guidelines
• You cannot use show-opts and configure-opts in the same command. You
must use separate SHOW commands.
• If you issue the SHOW command with a create-spec that is not currently set (or
is not applicable to the current value of TYPE), FUP returns only a prompt.
• The SHOW display might list an EXT size that differs by one page from what might
actually be created. This situation occurs because FUP rounds extent sizes up (if
necessary).
Note. For more information about the EXT parameter, see SET on page 2-166.
• Use the DISPLAYBITS option (from the CONFIGURE command) with SHOW for a
file that has alternate keys containing 8-bit characters in the partial key-value field.
For more information about the DISPLAYBITS option, see CONFIG[URE] on
page 2-26.
• You can set the extent size as the number of records in each extent. SHOW
displays the extent size in that form. For example, suppose you enter:
-SET EXT (100 RECS, 10)
The SHOW display then includes:
EXT ( 100 RECS, 10 PAGES )
• Use the SHOW command with the PHYSVOL option to display any SMF settings
for the file-creation attributes.
SHOW Examples
• This example assumes you enter FUP SHOW when the defaults are enabled for all
file-creation attributes, causing this information to be displayed:
TYPE U
FORMAT 1
EXT ( 1 PAGES, 1 PAGES )
MAXEXTENTS 16
BUFFERSIZE 4096
• To use the SHOW command to display the result of the SET commands used to
assign specific file-creation attributes:
> FUP
-SET TYPE K
-SET KEYLEN 2
-SET ALTKEY ("aa",FILE 0,KEYLEN 2,KEYOFF 0,INSERTIONORDER)
-SET ALTFILE (0, $DATAA.DCDTEST.ALT0)
-SHOW
TYPE K
EXT ( 1 PAGES, 1 PAGES )
FORMAT 1
REC 80
BLOCK 4096
KEYLEN 2
KEYOFF 0
ALTKEY ( "aa", FILE 0, KEYOFF 0, KEYLEN 2, INSERTIONORDER)
ALTFILE ( 0, $DATAA.DCDTEST.ALT0)
ALTCREATE
MAXEXTENTS 16
-EXIT
• Suppose these CONFIG[URE] commands are enabled:
CONFIGURE NETBLOCKSIZE 28
CONFIGURE DUP SOURCEDATE
A SHOW CONFIG[URE] command provides this result:
ALLOW ABENDS OFF
NO ECHO CONFIGURE
ECHO OBEY
NO PROMPT PURGE
NETBLOCKSIZE 28
DUP SOURCEDATE
STATUS
Reports the status of a reload operation.
OUT listfile
names an existing disk file or a device to receive the listing output from the
STATUS command. You can use either a standard file name or a spool DEFINE
name as the OUT listfile for a STATUS command. If listfile is an existing
file, FUP appends the output to the file.
Note. For more information about listfile, see Specifying Files on page 1-8.
filename
is the name of a key-sequenced file or SQL object reorganized by a reload
operation. You cannot use wild-card characters in filename or specify
qualified-fileset.
DETAIL
gives complete information on the status of the file you want listed. By default,
STATUS displays information only for fields it determines are relevant to the
current situation. For example, if a RELOAD process finishes, the STATUS
command displays only the completion time unless the DETAIL option is specified.
STATUS Guidelines
• If the reload operation is complete, FUP displays:
$vol-name.subvol-name.filename
RELOAD COMPLETED: date-time of completion
• If the reload operation is suspended (by the SUSPEND command), FUP displays:
$vol-name.subvol-name.filename
RELOAD INITIATED: date-time of initiation
SUSPENDED: date-time of suspension
DSLACK: percentage%
ISLACK: percentage%
RATE: percentage%
PCT COMPLETED: percentage%
DEALLOCATE: [YES|NO]
SHARE; [YES|NO]
RECLAIM: [YES|NO]
• If the reload operation is in progress, FUP displays:
$vol-name.subvol-name.filename
RELOAD IN PROGRESS
RELOAD INITIATED: date-time of initiation
RESUMED: date-time resumed
DSLACK: percentage%
ISLACK: percentage%
RATE: percentage%
PCT COMPLETED: percentage%
DEALLOCATE: [YES|NO]
The FUP process displays the RESUMED date and time only if the reload
operation was stopped and then restarted.
STATUS Examples
• To display the status of a RELOAD process while it is in progress:
-STATUS PAYFILE
$BASE.FUPTESTL.PAYFILE
RELOAD IN PROGRESS
RELOAD INITIATED: 10 Aug 2000, 15:35
DSLACK: 20%
ISLACK: 20%
RATE: 10%
PCT COMPLETED: 45%
DEALLOCATE: YES
• To display the status of a RELOAD process after its completion:
-STATUS PAYFILE, DETAIL
$BASE.FUPTESTL.PAYFILE
RELOAD INITIATED: 10 Aug 2000, 15:37
RELOAD COMPLETED: 10 Aug 2000, 15:39
DSLACK: 0%
ISLACK: 0%
RATE: 100%
PCT COMPLETED: 100%
DEALLOCATE: YES
SUBVOLS
Displays the names of all the subvolumes on a particular volume.
subvolset is:
OUT listfile
names an existing file or device to receive the output of the SUBVOLS command.
You can use either a standard file name or a spool DEFINE name as the OUT
Note. For more information about listfile, see Specifying Files on page 1-8.
subvolset
is a pattern describing the subvolume names you want to see. The subvolset
parameter defaults to the current volume. You can use wild-card characters in
subvolset (including the volume name), but you cannot specify
qualified-fileset.
SUBVOLS Examples
• To display the names of all the subvolumes on the $HANSEL volume:
-SUBVOLS $HANSEL
$HANSEL
BKUPLIST CARSELLA CBLIBERT D04RGEN GSPURGE
JOAND20 LATBOOT LATCA LATT001 LATT002
LATT003 LATT004 LATT005 LATT006 LATT007
LATT0 LATT013 LATT014 LATT015 LATT016
LATT017 LATT018 LATT019 LATT0 LUUTALQA
OSMAC OSMAC08 PETER RAUBATS RAUD00
• To use the wild-card option to display the names of a specific set of subvolumes:
-SUBVOLS $LARS*.GARY*
$LARS
GARYOLD GARYTACL GARYWORK GARY1
$LARS2
GARYNEW GARYZ
SUSPEND
Stops a reload operation. A subsequent RELOAD command can resume the reload
operation from the point where it was suspended. For more information, see RELOAD
on page 2-147.
OUT listfile
names an existing disk file or a device to receive the listing output from the
SUSPEND command. You can use either a standard file name or a spool DEFINE
Note. For more information about listfile, see Specifying Files on page 1-8.
filename
is the name of a key-sequenced file or SQL object (table or index) being
reorganized. You cannot use wild-card characters in filename or specify
qualified-fileset
SUSPEND Guidelines
• When you issue a SUSPEND command, FUP creates an online reload server
(ORSERV) process to stop the reload operation (which is controlled by another
ORSERV process). If FUP cannot create the second ORSERV process, an error
message appears.
• If a reload operation is not in progress for filename (or if the ORSERV process
cannot be stopped), FUP displays an error message.
• If the reload operation has already been suspended, FUP ignores the SUSPEND
command and does not display a message.
SUSPEND Example
-SUSPEND MYFILE
SYSTEM
Sets the default node. You can also set the default volume and subvolume names
used by FUP in expanding file names.
node
is a node name.
volume
names the new current default volume.
subvolume
names the new current default subvolume.
SYSTEM Guidelines
• The initial SYSTEM setting is the node enabled when you start FUP.
• You can use the SYSTEM command only to reference nodes that have a name; for
example, a node that is part of a network.
• Entering the SYSTEM command without node restores the default node name to
the local node.
• If you omit node parameter, FUP uses the local form for file-name expansion.
Using the local form allows access to all volume and device names on the node.
• When you supply node (even if the name of the local node is specified), FUP uses
the network form for file-name expansion. Using the network form causes only
volume or device names with six characters or fewer to be accessible.
• The volume and subvolume parameters are optional. If you omit these
parameters, FUP uses their current default values.
• If node is an invalid node name, FUP displays an error message and does not
change the node. However, FUP does not issue an error if volume (or its default
value) does not exist on the new system.
• FUP and the command interpreter keep separate defaults for node, volume, and
subvolume. When you exit FUP, the command interpreter defaults are still the
values that were enabled when you started FUP.
SYSTEM Examples
• To set the \LONDON node as the current default for FUP and keep your current
default volume and subvolume:
-SYSTEM \LONDON
• To set the current FUP defaults to the PAID subvolume on the $BILLS volume of
the NEWYORK node:
-SYSTEM \NEWYORK.$BILLS.PAID
TRUST
Changes the state of Trust flag to TRUST ME, TRUST SHARED or TRUST OFF. Only
a super ID (255, 255) user can set the FLTTrustFlags flag to TRUST ME or TRUST
SHARED. This flag controls whether direct I/O access to user buffers is permitted
when this process is running.
fileset-list
is a list of disk files whose state of Trust flag is to be changed.
trust-flag
is one of:
OFF
ME
SHARED
OFF
resets Trust flag.
ME
sets Trust flag to TRUST ME state.
SHARED
sets Trust flags to TRUST SHARED state.
TRUST Guidelines
• The code of the files must be 100 or 800. If a files has another file code, error 2
occurs.
• If a user who is not the super ID (255,255) attempts to change the Trust flags to
TRUST ME or TRUST SHARED, FUP gives a warning of INSUFFICIENT
CAPABILITY.
TRUST Examples
• To set TRUST flag of file MYPROG to TRUST ME:
-TRUST MYPROG, ME
• To set TRUST flag of file MYPROG to TRUSTSHARED:
-TRUST MYPROG, SHARED
• To rest TRUST flag of file MYPROG:
-TRUST MYPROG, OFF
VOLS
Displays information about volumes on a system. The VOLS command lists the
names, capacities, and current usage for all specified volumes in a format that is
similar to the Disk Space Analysis Program (DSAP) utility.
volset is:
[ \node.] $volume
OUT listfile
names an existing disk file or a device to receive the listing output from the VOLS
command. You can use either a standard file name or a spool DEFINE name as
the OUT listfile for a VOLS command. If listfile is an existing file, FUP
appends the output to that file.
Note. For more information about listfile, see Specifying Files on page 1-8.
volset
specifies volumes to display information about. The volset parameter can be the
name of a volume or a wild card, and it can include a system name. If you specify
a volume name, it must begin with a dollar sign ($). The default is all volumes on
the current system ($*).
VOLS Example
To show information on all volumes of the current system that begin with the letter M:
-VOLS $M*
-- Capacity (Mb) -- % -- Free Extents --
Volume (M) Total Free Free Count Biggest
$MCAT Y 895 346.62 38 12 332.24
$MOLD Y 415 50.04 12 25 26.67
$MONO 895 289.97 32 125 287.74
$MONO1 895 314.41 35 1 314.41
$MOSS Y 415 150.09 36 491 66.67
VOLUME
Changes the current default volume or subvolume names used during the execution of
FUP. The initial defaults are the subvolume and volume that were the current defaults
when you started FUP.
node
sets the current default node for FUP. You can set the default node only in a
named node; for example, a node that is part of a network.
volume
sets the current default volume name for FUP operations.
subvol
sets the current default subvolume name for FUP operations.
VOLUME Guidelines
• Typing the VOLUME command without a parameter restores the defaults that were
enabled when you started FUP, including the original node.
• When you supply node (even if you specify the name of the local node), FUP uses
the network form for file-name expansion. Using the network form causes only
volume or device names with six characters or fewer to be accessible.
• You cannot specify a new current default volume name that has more than six
characters after the dollar sign ($) on a remote node.
• node and subvolume are optional. If you omit these parameters, FUP uses their
current default values.
• If node is an invalid node name, FUP displays an error message and does not
change the node. However, FUP does not issue an error if volume (or its default
value) does not exist on the new system.
• FUP and the command interpreter keep separate defaults for node, volume, and
subvolume. When you exit FUP, the command interpreter defaults are still the
values that were enabled when you started FUP.
VOLUME Examples
• To set the current default subvolume for FUP to SUBVOL1 but not change the
current default volume or node:
-VOLUME SUBVOL1
• To set the current default volume for FUP to $BOOKS3 but not change the current
default subvolume or node:
-VOLUME $BOOKS3
• To set the current default node, volume, and subvolume for FUP operations to
\ITALY, $MILANO, and ARTWORK:
-VOLUME \ITALY.$MILANO.ARTWORK
Note. You should never receive a severe-problem message (CALAMITY). If you do, contact
your service provider.
Cause. During an attempt to create a new file (by using the CREATE or DUP
command) or alter an old file, a secondary file (secondary partition or alternate-key file)
was placed on a nonexistent volume.
Effect. A warning is issued, and the command execution proceeds.
Recovery. Place all alternate-key files and secondary partitions on the primary volume
or on existing volumes.
Cause. A FUP CREATE or SET command was used to specify a partition for a non-
key-sequenced file that had a different file format than an existing partition.
Effect. The command fails.
Recovery. Reenter the command with an appropriate file format.
Cause. An attempt to create or alter a file failed because an alternate-key file that had
no corresponding alternate key was specified.
Effect. The CREATE or ALTER command fails.
Recovery. Retry the command after correcting the ALTKEY or ALTFILE definition.
Cause. An attempt to create or alter a file failed because the length of the alternate
key (which was added to the offset of the alternate key) exceeded the record length
specified for the file.
Effect. The CREATE or ALTER command fails.
Recovery. Retry the command after correcting the ALTKEY or ALTKEY definition.
Cause. An attempt to create or alter a file failed. An alternate key was specified for a
file, but an alternate-key file was not specified for the alternate key.
Effect. The CREATE or ALTER command fails.
Recovery. Retry the command after correcting the ALTKEY or ALTFILE definition.
Cause. When entering the CREATE or ALTER command, you either omitted the
KEYLEN parameter from an ALTKEY specification or specified zero for KEYLEN.
Effect. The CREATE or ALTER command fails.
Recovery. Retry the command after correcting the KEYLEN parameter.
Cause. The ASSUMEID attribute of a file failed to transfer during a duplicate operation
while using the SAVEALL option to a remote node.
Cause. An attempt to load or copy a file failed because the length of a block was
either 0 or 1 byte. This error can occur if the file to be loaded or copied was filled
without specifying the VAROUT option of the COPY command. Alternatively, data in
the file might have been corrupted.
Effect. The COPY or LOAD command fails.
Recovery. Variable-length blocks must contain at least two bytes at the beginning of
the block.
Cause. The PART specification for CREATE or ALTER was wrong. For example, all
partition numbers must be sequential, starting from 1. When this message appears
after FUP LOAD... PARTOF..., it indicates that FUP found the file that should have
been the primary file, but it was not a partitioned file, or it was not the primary partition
for the destination.
Effect. The command fails.
Recovery. In the first case, retry the command after specifying all partitions up to the
maximum partition number. In the second case, correct either the destination file name
or the PARTOF volume name.
Cause. An attempt to load or copy a file failed because the record length specified for
a variable-length record was not valid. For example, the record length was a negative
number. This error can occur if the file to be loaded or copied was filled without
specifying the VAROUT option of the COPY command. Alternatively, data in the file
might have been corrupted.
Effect. The COPY or LOAD command fails.
Recovery. Retry the command after specifying the VAROUT option.
Cause. Unique and nonunique alternate keys cannot reside in the same alternate-key
file.
Effect. The CREATE or ALTER command fails.
Recovery. Retry the command after changing the alternate-key file.
CAN'T OPEN TAPE FILE FOR READ ACCESS IF DEFINE VALUE FOR USE
IS EXTEND
Cause. A TAPE DEFINE was specified with USE set to EXTEND, but the DEFINE
name was specified as an input file in either a FUP COPY or LOAD command.
Effect. The command fails.
Recovery. Modify the DEFINE or command parameters and retry the command.
CAN'T OPEN TAPE FILE FOR READ ACCESS IF DEFINE VALUE FOR USE
IS OUT
Cause. A TAPE DEFINE was specified with USE set to OUT, but the DEFINE name
was specified as an input file in either a FUP COPY or LOAD command.
Effect. The command fails.
Recovery. Modify the DEFINE or command parameters and retry the command.
CAN'T OPEN TAPE FILE FOR WRITE ACCESS IF DEFINE VALUE FOR USE
IS IN
Cause. A TAPE DEFINE was specified with USE set to IN, but the DEFINE name was
specified as an output file in either a FUP COPY or BUILDKEYRECORDS command.
Effect. The command fails.
Recovery. Modify the DEFINE or command parameters and retry the command.
Cause. An optical disk file was specified in a command that cannot operate on optical
disk files.
Effect. The command fails.
Recovery. If you intended to specify another type of file, retry the command after
correcting the file specification.
Cause. An attempt was made to copy or load a nonrelative file using the COMPACT
option.
Effect. Execution of the command proceeds.
CONSISTENCY CHECK
Cause. A logical style DEFINE name was specified for a DEFINE class, such as class
SPOOL or class TAPE, that does not exist.
Effect. The command fails.
Recovery. Specify a correct name and retry the command.
Cause. When an attempt was made to use FUP ALTER filename, NO AUDIT, FUP
encountered a “file is bad” error (CRASHOPEN flag is on). filename was partitioned.
Effect. The command is not executed.
Recovery. Reissue the FUP ALTER command for each partition.
Cause. A file-system error occurred while you tried a CREATE command with the
indicated file.
Effect. The command fails.
Recovery. For corrective action for the file-system error number indicated by nnn, see
the Guardian Procedure Errors and Messages Manual.
Cause. A file name was entered with the volume name specified and the subvolume
name unspecified. In later RVUs, you are required to specify the subvolume name
when you specify the volume name in a command.
Effect. Processing continues.
Cause. An attempt to open a labeled-tape file failed with file-system error 196 (tape
label record missing or incorrect).
Effect. The command fails.
Recovery. Verify that the specified DEFINE attributes match the label for the tape file.
Correct, if necessary, and retry the command. If the command fails again, for file-
system error 196 corrective action, see the Guardian Procedure Errors and Messages
Manual.
Cause. An internal error occurred while the system was processing DEFINEs.
Effect. The command fails.
Recovery. Note the error indicated by nnn and contact your service provider.
Cause. A RESTART operation was attempted, but the destination file was not corrupt,
which implies that the previous attempt to use DUP was successful.
Effect. The RESTART operation terminates.
Recovery. If you need to run DUP, perform an ordinary DUP operation.
Cause. The DP2 lock tables changed during the current LISTLOCK operation.
Effect. All lock information already displayed for filename is invalid.
FUP automatically starts over, retrieving and displaying current lock information for the
file identified by filename in the error message.
Recovery. Informative message only; no corrective action is needed.
Cause. The DP2 lock tables changed more than three times while FUP was
attempting to retrieve and display lock information for the filename identified in the
error message.
Effect. An error is generated, and the LISTLOCKS operation fails.
Cause. The TAPEMODE parameter was used for a tape drive that does not support
the setting.
Effect. The COPY or BUILDKEYRECORDS command continues without trying to set
the tape mode.
Recovery. Informational message only; no corrective action is needed.
Cause. A DENSITYOUT parameter was specified for a tape drive that does not
support density selection, or the DENSITYOUT parameter was specified when the tape
was not at the beginning.
Effect. The COPY or BUILDKEYRECORDS command continues without trying to set
the density.
Recovery. Informational message only; no corrective action is needed if the tape drive
was at the correct density.
Cause. The same volume was specified for more than one partition of a partitioned
file.
Effect. The CREATE, ALTER, or DUP command fails.
Recovery. Each partition of a partitioned file must reside on a separate disk volume.
EDITREAD ERROR
Cause. An error occurred while the system was reading an EDIT file (READEDIT); it
was probably a file-format error.
Effect. The command fails.
Recovery. For corrective action, see the Guardian Procedure Errors and Messages
Manual.
Cause. The source file of a LOAD command had zero records, and the EMPTYOK
option was not specified. Because all existing data in the destination file of a LOAD
command is purged at the beginning, this check for an empty source file ensures that a
mistake in typing the LOAD command (such as interchanging the source and the
destination files) does not result in the loss of data.
Effect. In interactive mode, the LOAD command terminates by returning to the FUP
prompt. In noninteractive mode, the LOAD command terminates by abnormally ending.
Recovery. Check that you entered the command as intended. If the source file is
empty, to achieve the effect of loading the destination with an empty file, use the
EMPTYOK option.
Cause. FUP could not find the primary partition on the volume given as the PARTOF
parameter.
Effect. The operation fails.
Recovery. The filename parameter indicates the name of the primary partition that
FUP required. Correct the file name and reissue the command.
Recovery. Retry the command after correcting the errors. If the file was being updated
concurrently, the error might be only a transient structure problem FUP encountered
while reading large blocks.
Cause. A file-system error occurred. filename indicates the name of the file that
FUP was attempting to process.
Effect. The operation fails.
Recovery. For corrective action for the file-system error number indicated by nnn,
see the Guardian Procedure Errors and Messages Manual.
Cause. A LISTOPENS command that specified a device other than a disk was
attempted when FUP expected a disk device name.
Effect. The command fails.
Recovery. Correct the name of the LISTOPENS object and retry the command.
Cause. A LISTOPENS command that specified a disk device name was attempted
when FUP expected a nondisk device name.
Effect. The LISTOPENS command terminates.
Recovery. Correct the name of the LISTOPENS object and retry the command.
FILE INCOMPATIBLE
Cause. The source and destination files of a DUP...,OLD command are incompatible.
Effect. The attempted duplication fails.
Recovery. You cannot use the OLD option for incompatible files. However, you can
purge the old file and retry the DUP command.
FILES CHECKSUMMED: nn
Cause. nnn indicates the number of files from fileset that were duplicated.
Effect. None.
Recovery. Informational message only, indicating successful completion for all or
some of the specified files.
Recovery. Write down the complete description and contact your service provider.
ILLEGAL CHARACTER
Cause. An illegal character was used in the attempted command or ANSI Name is
entered specifying an invalid SQL identifier character.
Effect. The command fails.
Recovery. Retry the command after correcting the illegal character.
ILLEGAL FILENAME
Cause. The file name specified in the command was not valid.
Effect. The command fails.
Recovery. Retry the command after correcting the file name.
Cause. The attempted command specified an invalid key value. For example, the
FIRST option of the COPY command specified an invalid key value for position
purposes.
Effect. The command fails.
Recovery. Retry the command after correcting the key value.
ILLEGAL OPTION
Cause. An option not allowed for the attempted command was used.
Effect. The command fails.
Recovery. Retry the command after correcting the option.
Cause. A syntax error occurred in the qualified file-set expression. The additional text
should describe the specific error.
Effect. The command fails.
Recovery. Retry the command after correcting the syntax.
ILLEGAL SYNTAX
ILLEGAL VALUE
Cause. The attempted command specified an illegal value: for example, SET EXT -1.
Effect. The command fails.
Recovery. Retry the command after correcting the value.
Cause. An attempt was made to create a DP2 key-sequenced file with the index block
size and data block size set to different values.
Effect. The CREATE command finishes, but the IBLOCK value for the file is set to its
BLOCK value.
Recovery. If the modified IBLOCK value is not the value required, reset the BLOCK
and IBLOCK values, and retry the command. The index block size must equal the data
block size for DP2 key-sequenced files.
INSUFFICIENT CAPABILITY
Cause. An attempt was made to execute a command reserved for the super ID
(255,255).
Effect. The command fails.
Recovery. Only the super ID can use the attempted command.
Recovery. Note the internal load error indicated by nnn, and contact your service
provider.
Cause. The reload server process returned invalid status information to a STATUS
command.
Effect. The command fails.
Recovery. This is an internal error. Contact your service provider.
filename : KEPT
Cause. During execution of a DUP command with the KEEP option specified, the file
indicated by filename was not duplicated.
Effect. None.
Recovery. Informational message only, indicating successful completion for all or
some of the specified files.
Cause. An attempt to create a file failed because the key length specified for the
primary key was not specified or was specified as zero.
Effect. The CREATE or ALTER command fails.
Recovery. Retry the command with a legal KEYLEN value.
Cause. An attempt to open a tape file failed with file-system error 195 (operation
requires use of $ZSVR, but it is not running). The labeled-tape server, $ZSVR, is not
running.
Effect. The command fails.
Recovery. Ask the system manager to start $ZSVR if possible and then retry the
command.
Cause. The transfer of a file’s LICENSE attribute failed during a duplicate operation
with the SAVEALL option to a remote node.
Effect. The LICENSE attribute is not preserved.
Recovery. Use the LICENSE command to license the file. Remote logon might be
required.
Cause. The file-system LOCKINFO procedure, which provides FUP with all lock
information, returned the error errnum. This message appears when errors occur
other than the detection of a DP2 lock table that is changing or has changed during a
LISTLOCKS operation.
Effect. An error is generated, and the LISTLOCKS operation fails.
Recovery. Use the ERROR program for an explanation of errnum.
Cause. FUP does not return lock information for partitioned files.
Effect. None.
Recovery. Informational message only; no corrective action is needed.
MISSING KEYWORD
Cause. A syntax error occurred. FUP required a keyword but found nothing or
something other than a keyword.
Effect. The command fails.
Recovery. Retype the command, correcting the syntax error.
MISSING NUMBER
Cause. FUP required a number but found nothing or an alphabetic value. For
example, after the SET option PART, the partition number must be provided.
Effect. The command fails.
Recovery. Retry the command after including the required number.
MISSING STRING
Cause. FUP required a string specification but did not find one. For example, for key-
sequenced files, the PART specification must include the partial-key value, which is a
string.
Effect. The command fails.
Recovery. Retry the command after including the required string specification.
Cause. This is not an error. FUP was reading from a tape file (for example, COPY,
LOAD from tape), and no more records remained in the current reel, but there are
more reels to read.
Effect. FUP waits for the next reel to be mounted.
Recovery. Mount the next reel.
Cause. An attempt was made to duplicate a partition PARTONLY. FUP detected that
the partition data would be reshuffled among the partitions.
Effect. The command fails.
Recovery. Reissue the command without the PARTONLY option, and convert the file
as a whole.
Cause. A Safeguard protected file was duplicated, but the new file is not Safeguard
protected.
Effect. The file is duplicated; this is only a warning.
Recovery. Use the Safeguard command interpreter to establish Safeguard protection
for the file.
Cause. No status information was found for a RELOAD operation, or the information
found was invalid or inconsistent.
Effect. The STATUS command fails.
Recovery. Check that you spelled filename correctly or that filename exists
before retrying STATUS.
NO SUCH LINE
Cause. The specified text for the HISTORY, !, or ? command was not found in the
buffer.
Effect. None.
Recovery. Enter the correct text and try the command again.
NO SUCH SYSTEM
Cause. The SAVEALL, SAVEID, or SOURCEDATE option of DUP was specified, and
an error occurred while saving one or more of the relevant attributes.
Effect. The DUP command finishes successfully, but not all the attributes are
transferred.
Recovery. Check which attributes did not transfer and transfer manually (if
necessary).
Cause. The attempted command was not allowed for partitioned files. For example, an
attempt was made to duplicate a partitioned file, specifying the OLD option without the
PARTONLY option.
Effect. The command fails.
Recovery. Check which commands are allowed with partitioned files.
Cause. A character was entered on the same line after the continuation character (&).
Effect. The command fails.
Recovery. Retry the command after deleting the character that follows the
continuation character.
Cause. An operation was attempted in batch mode that can be performed only from a
terminal.
Effect. None.
Recovery. Perform the operation in interactive mode from a terminal.
Cause. When an ALTER command was entered, only the PARTONLY option was
specified. No changes to the file were specified.
Effect. None.
Recovery. Retry the command, specifying the required changes.
Cause. An attempt was made to duplicate a file using the OLD option with either the
PART, ALTFILE, or EXT option. These options are mutually exclusive.
Effect. Only the OLD option is used, and the others are ignored. This is only a
warning.
Recovery. To change something using PART, ALTFILE, or EXT, do not use the OLD
option.
Cause. A RESTARTABLE DUP operation was attempted, but more than one source
file was specified.
Effect. The operation terminates.
Recovery. Specify only one source file when you perform the operation again.
Cause. A file-system error was encountered while attempting to open the indicated
file.
Effect. The command fails.
Recovery. For corrective action for the file-system error number indicated by nnn, see
the Guardian Procedure Errors and Messages Manual.
Cause. An SQL object was specified in a command that cannot operate on these
objects.
Effect. The command fails.
Recovery. If you intended to specify another type of file, retry the command after
correcting the file specification. To perform operations on SQL objects, use the SQL
Conversational Interface (SQLCI) for SQL/MP files and MXCI for SQL/MX files.
Cause. You specified a file on optical disk as the OUT file for the RUN FUP command
(for example, RUN FUP/OUT $OPT.X.Y/) or as the OUT file for the FUP commands
HELP, INFO, FILES, SUBVOLS, LISTOPENS, or SHOW (for example, HELP / OUT
$OPT.X.Y/ ALL).
Effect. The command fails.
Recovery. Retry the command with an OUT file residing on magnetic disk.
OVERFLOW OF TABLE
Cause. FUP ran out of internal memory space while trying to execute the command.
Effect. FUP ends abnormally.
Recovery. Restart FUP, and retry the command. If the same error occurs, call your
service provider.
Cause. The transfer of a file’s OWNER attribute failed during a duplicate operation
with the SAVEALL or SAVEID option to a remote node.
Effect. The OWNER attribute is not preserved.
Recovery. Use the GIVE command to change the owner. Remote logon might be
required.
Cause. When creating or altering a partitioned file, the numbering of the partitions was
not in the same order as the partial key values associated with the partitions.
Effect. The command fails.
Recovery. Retry the command after correcting the numbering of the partial key values
or the partitions.
Cause. The file attributes of the individual partitions of a partitioned file were
inconsistent with one another.
Effect. The command fails.
Recovery. Check the partitions, and ensure consistency by executing the necessary
FUP ALTER commands. For all partitions, check that:
• The file type, record length, data-block length, key length, key offset, index-block
length, and index and data-compression (where applicable) attributes are the
same.
• For relative and entry-sequenced files, the partition extent sizes are the same as
those specified when the primary partition was created.
Cause. A DUP operation that did not specify PARTONLY was attempted on a
partitioned file, and one of the partitions is an SQL object.
Effect. None of the file is duplicated. The DUP operation continues executing on the
next file in the file set.
Recovery. If you intended to specify another type of file, retry the command after
correcting the file specification.
Cause. When you created or altered a file with a nonunique alternate key, the
combined length of the primary key, the alternate key, and the 2-byte key specifier
exceeded the 255-byte total allowed. Or when you created or altered a file with a
unique alternate key, the combined length of the alternate key and the 2-byte key
specifier exceeded the allowed 255-byte total.
Effect. The command fails.
Recovery. Retry the command, specifying values within the specified byte limits.
Cause. A file-system error occurred while the system tried to purge the indicated file.
Effect. The command fails.
Recovery. For corrective action for the file-system error number indicated by nnn, see
the Guardian Procedure Errors and Messages Manual.
Cause. FUP cannot access the file specified in a qualified file set.
Effect. A file-system error (error number could be 0) is generated.
Recovery. If the error is not apparent, try using INFO on a file with a similar file set
without using a qualified file set.
Cause. Qualified file set is either not permitted in this command, or it was used in an
incorrect position.
Effect. The command fails.
Recovery. Use the qualified file set in the correct position, or retry the command
without a qualified file set (if they are not permitted in this command).
Cause. In the primary file, nnn records had partial alternate-key fields. The actual
record did not contain the full length of the alternate key. The corresponding alternate-
key records were not generated in the destination file by LOADALTFILE or
BUILDKEYRECORDS.
Effect. The LOADALTFILE or BUILDKEYRECORDS command finishes, but no
alternate-key records are generated for the records with partial alternate keys. This is
only a warning.
Recovery. Retry the command after padding the records in the primary file so that
alternate keys are always completely contained in the records.
Cause. A FUP command was entered that attempted to assign a record size
inappropriate for the file format of the file.
Effect. The command fails.
Recovery. Reenter the command with an appropriate record size. For a list of
appropriate record sizes, see Table 1-1, File Format Codes, on page 1-23.
RECORDS LOADED: nn
Cause. An attempt was made to use the REELS parameter when in-file for the
operation was a TAPE DEFINE name with LABELS set to label processing.
Effect. The command fails.
Recovery. Use the REELS and VOLUME TAPE DEFINE attributes with multiple
labeled tapes.
Cause. A RELOAD command was issued against filename while another RELOAD
command was already in progress for that file.
Effect. The command fails.
Recovery. Informational message only; no corrective action is needed.
Cause. A file-system error was encountered while attempting to rename the indicated
file.
Effect. The command fails.
Recovery. For corrective action for the file-system error number indicated by nnn, see
the Guardian Procedure Errors and Messages Manual.
Cause. FUP was expecting a delimiter but did not find it or an ANSI Name is entered
without specifying a required comma, open parenthesis, or closing parenthesis.
Effect. The command fails.
Recovery. Retry the command after adding the delimiter.
Cause. A RESTART operation was attempted, but the restart file did not have the
correct file code. The restart file was not created by FUP as part of a legitimate
RESTARTABLE DUP operation.
Cause. For a RESTARTABLE DUP operation or a RESTART operation, the restart file
was named explicitly or implicitly (that is, as a default) on an optical disk volume.
Effect. The operation terminates.
Recovery. Do not specify a restart file on an optical disk when you perform the
operation again.
Cause. A RESTART operation was attempted, but the restart file does not contain
valid information.
Effect. The RESTART operation terminates.
Recovery. Perform an ordinary DUP operation.
Cause. The transfer of a file’s SECURITY attribute failed during a duplicate operation
with the SAVEALL or SAVEID option to a remote node.
Effect. The SECURITY attribute is not preserved.
Recovery. Use the SECURE command to change the security. Remote logon might
be required.
Cause. An attempt was made to use the SKIPIN parameter when the in-file for the
operation was a TAPE DEFINE name with LABELS set to label processing.
Effect. The command fails.
Recovery. Use the FILESEQ TAPE DEFINE attribute to skip files on a labeled tape.
Cause. An attempt was made to use the SKIPOUT parameter when the out-file for
the operation was a TAPE DEFINE name with LABELS set to label processing.
Effect. The command fails.
Recovery. Use the FILESEQ TAPE DEFINE attribute to skip files on a labeled tape.
Cause. The indicated error or errors were reported by a SORT interface routine during
execution of a LOAD, LOADALTFILE, or LISTOPENS command.
Effect. The LOAD, LOADALTFILE, or LISTOPENS command fails.
Recovery. A SORT error occurred in the specified phase. Four different types of
errors are possible. FILE ERR is a standard file-system error. NEWPROCESS ERR
includes the process creation error number (n) and additional process creation error
information (%nn). INTERNAL ERR is a SORT internal error. Correct the error and retry
the command.
Cause. A RESTART operation was attempted, but the source file was modified since
the previous attempt to use DUP.
Effect. The RESTART operation terminates.
Recovery. Perform an ordinary DUP operation.
Recovery. This is an internal error. Note the error and contact your service provider.
TABLE OVERFLOW
Cause. The value of in-file was a TAPE DEFINE name, and the BLOCKLEN value
in the DEFINE was greater than 32,767 (the largest block size that FUP can
accommodate).
Effect. The command fails.
Recovery. If possible, change the BLOCKLEN value to a size that FUP can
accommodate.
Cause. The value of in-file was a TAPE DEFINE name, and the RECLEN value
was greater than 4096.
Effect. The command fails.
Recovery. If possible, change the RECLEN value to a size that FUP can
accommodate.
Cause. The value of in-file was a TAPE DEFINE name, and the BLOCKLEN value
in the DEFINE did not equal the value of the BLOCKIN FUP parameter.
Effect. The command fails.
Recovery. Change the BLOCKLEN or BLOCKIN value and retry the command.
Cause. The value of in-file was a TAPE DEFINE name, and the BLOCKLEN value
in the DEFINE did not equal the value of the BLOCKOUT FUP parameter.
Effect. The command fails.
Recovery. Change the BLOCKLEN or BLOCKOUT value and retry the command.
Cause. The value of in-file was a TAPE DEFINE name, and the DENSITY value in
the DEFINE did not equal the value of the DENSITYOUT FUP parameter.
Effect. The command fails.
Recovery. Change the DENSITY or DENSITYOUT value and retry the command.
Cause. The value of in-file was a TAPE DEFINE name, the EBCDIC value in the
DEFINE was OFF, and the EBCDICIN FUP parameter was specified.
Effect. The command fails.
Recovery. Change the EBCDIC value in the DEFINE or omit EBCDICIN, and retry the
command.
Cause. The value of in-file was a TAPE DEFINE name, the EBCDIC value in the
DEFINE was OFF, and the EBCDICOUT FUP parameter was specified.
Effect. The command fails.
Recovery. Change the EBCDIC value in the DEFINE or omit EBCDICOUT, and retry
the command.
Cause. The value of in-file was a TAPE DEFINE name, and the RECLEN value in
the DEFINE did not equal the value of the RECIN FUP parameter.
Effect. The command fails.
Recovery. Change the RECIN or RECLEN value and retry the command.
Cause. The value of in-file was a TAPE DEFINE name, and the RECLEN value in
the DEFINE did not equal the value of the RECOUT FUP parameter.
Effect. The command fails.
Recovery. Change the RECLEN or RECOUT value and retry the command.
Cause. The TAPE DEFINE attribute for TAPEMODE and the FUP parameter for
TAPEMODE were both specified, and they conflict.
Effect. The COPY or BUILDKEYRECORDS command terminates.
Recovery. Modify one of the TAPEMODEs and retry the operation.
Cause. This is not an error. FUP was writing to a tape file and reached the end of the
current reel.
Effect. FUP waits for the next reel to be mounted.
Recovery. Mount the next reel.
Cause. A write request was made, but the write ring was missing.
Effect. The write request to the tape fails.
Recovery. Unmount the tape, insert a write ring, and remount the tape. Press
RETURN to restart the operation.
Cause. An attempt to open a tape file failed with file-system error 194 (device use or
mount request rejected by operator). The operator rejected the request to mount the
specified tape.
Effect. The command fails.
Recovery. Determine why the tape mount request was rejected, resolve the problem,
and retry the command. For the corrective action for file-system error 194, see the
Guardian Procedure Errors and Messages Manual.
TOO LONG
TOO MANY
Cause. Too many objects were specified for the attempted command. For example,
more than 10 file sets were specified in a file-set list for FUP, or too many key tags
were specified for the BUILDKEYRECORDS command.
Effect. The command fails.
Recovery. Retry the command with an acceptable number of specified objects.
Cause. Too many (more than 1000) volumes were specified for the system.
Effect. Only the first 1000 volumes are sorted by name. The remaining volumes are
printed in hardware device number order.
Recovery. Retry the command with 1000 or fewer volumes specified.
TRUNCATION OCCURRING
Cause. During execution of the COPY, LOAD, or LOADALTFILE command, the input
records were truncated.
Effect. The command proceeds; this is only a warning.
Recovery. Check the BLOCKIN, RECIN, or input record length values. If truncation is
intended, ignore the warning.
Cause. This is a warning message. The VARIN option was used with the COPY or
LOAD command, and the last variable-length record in the block was truncated
because the record, as indicated by the record length at the beginning of the record,
extended beyond the end of the block.
Effect. Execution of the command proceeds; this is only a warning.
Recovery. Check that the BLOCKIN value (if specified) is correct and that the file was
generated by FUP COPY with a VAROUT option.
Cause. The LISTOPENS command does not return information for devices that have
subdevices.
Effect. You receive an error message instead of the information you requested.
Recovery. You need to execute LISTOPENS for specific terminals. An example is
LISTOPENS $TC1.#C13.
Cause. A SUSPEND command was issued, but the RELOAD command could not be
stopped.
Effect. The command fails.
Recovery. Check that you have proper security access to stop the reload server
process and then retry the command.
Cause. An attempt to create or alter a file failed because unique keys were not the
same length within the same alternate-key file.
Effect. The command fails.
Recovery. Place unique keys with different lengths in different key files.
UNKNOWN COMMAND
Recovery. Check that what you are trying can be done using FUP. If it can, retry the
command after correcting any errors in typing.
UNIMPLEMENTED FUNCTION
UNTERMINATED STRING
Cause. An attempt was made either to duplicate a partition, PARTONLY, by using the
EXT or PART option to specify the extent sizes, or to alter the extent sizes of a
secondary partition in the file label of the primary partition of a DP2 key-sequenced file.
Effect. The ALTER finishes; this is only a warning.
Recovery. Check that the extent sizes in the file label of the primary partition reflect
the actual extent sizes of the secondary partitions.
Cause. The device specified in a LISTOPENS command does not return open
information.
Effect. The command fails.
Recovery. For information on correcting the indicated file-system error number, see
the Guardian Procedure Errors and Messages Manual.
WRITEEDIT ERROR
Cause. An error occurred while writing an EDIT file (WRITEEDIT); it was probably a
file-format error.
Effect. The command fails.
Recovery. For corrective action, see the Guardian Procedure Errors and Messages
Manual.
Cause. FUP returns this error message when the ANSI Names mapping service
(ANMS) is unable to launch the ANSI Name Mapping Server Process (ANSP). The
error code returned is the specific PROCESS_CREATE error.
Effect. The command fails.
Recovery. For corrective actions contact the SQL/MX ANSI Names developers.
Cause. Retuned by ANSI Names Mapping Service (ANMS) to FUP when ANSI Name
is entered and some miscellaneous SQL error occurs.
Effect. The command fails.
Recovery. For corrective actions, contact the SQL/MX ANSI Names developers.
Cause. The SQL/MX ANSI Names mapping service (ANMS) is unable to perform the
mapping due to a file-system error.
Effect. The command fails.
Recovery. For corrective actions contact the SQL/MX ANSI Names developers.
Cause. ANSI Name entered is not fully qualified, i.e. A standard SQL three-part name
consisting of three SQL identifiers representing, respectively, a catalog name, a simple
schema name, and a simple object name is not entered.
Effect. The command fails.
Recovery. Retry the command by entering the fully qualified ANSI Name.
Cause. ANSI Name is entered with SQL identifier that is too long.
Effect. The command fails.
Recovery. Retry the command after correcting the SQL identifier.
Cause. The delimited ANSI Name is entered without a matching double quote.
Effect. The command fails.
Recovery. Retry the command after adding the missing double quote.
Cause. ANSI Name is entered with the SQL/MX SCHEMA USER keyword.
Effect. The command fails.
Recovery. None.
Cause. ANSI Name is entered with CATALOG or SCHEMA in the INFO or RELOAD
command.
Effect. The command fails.
Recovery. None.
Cause. ANSI Name is entered with TABLE or INDEX in the RELOAD command.
Effect. The command fails.
Recovery. None.
Cause. Returned when FUP tries to launch one more ANSI Name Server Process
(ANSP) when it already has one active ANSP.
Effect. The command fails.
Recovery. This message should never appear. If it does, stop FUP and start it again
and then issue the command.
Cause. ANSI Name entered does not exist on the system or is inaccessible.
Effect. The command fails.
Recovery. Check whether the input ANSI Name really exists in the system or not and
then retry the command with the correct ANSI Name..
Cause. ANSI Name is encountered after the Guardian name in any command
supporting ANSI Names or in any command that does not support ANSI Names.
Effect. The command fails.
Recovery. Retry the command by entering either the ANSI Name or the Guardian
name.
Cause. Returned by FUP when an internal error occurs in the ANSI Name Server
Process (ANSP) due to a memory allocation failure.
Effect. The command fails.
Recovery. For corrective actions, contact the SQL/MX ANSI Names developers.
Cause. ANSI Name is entered without specifying a required starting single quote or if
guardian file name is entered after the first ANSI Name.
Effect. The command fails.
Recovery. Retry the command after adding the required missing single quote. If you
have entered the Guardian name along with the ANSI Name in the same command,
then retry the command by entering either the ANSI Name or the Guardian name.
Cause. ANSI Name is entered without specifying a required ending single quote or the
ANSI Name is not entered in the proper format.
Effect. The command fails.
Recovery. Retry the command after adding the required missing single quote or by
correcting the ANSI Name.
Effect. ANSI Name is entered with space character missing between the ANSI Name
keyword and the SQL identifier.
Effect. The command fails.
Recovery. Retry the command after adding the required space character.
Cause. ANSI Name is entered but SQL/MX is not installed on the system.
Effect. The command fails.
Recovery. Install SQL/MX on the system.
Cause. Returned by FUP when SQL/MX ANSI Names APIs are not installed on the
system.
Effect. The command fails.
Cause. There is a version mismatch between FUP and the SQL/MX MAPPING
SOFTWARE (SQL/MX ANSI Names mapping service - ANMS).
Effect. The command fails.
Recovery. Install the version of FUP and the SQL/MX MAPPING SOFTWARE
(SQL/MX ANSI Names mapping service ANMS) that are compatible with each other.
Cause. There is a version mismatch between FUP and the SQL/MX parser.
Effect. The command fails.
Recovery. Install the version of FUP and SQL/MX parser that are compatible with
each other.
Note. Do not use TAPE DEFINE attributes that conflict with your FUP command parameters.
For more information about DEFINEs, see the Guardian User’s Guide .
Table A-1. How FUP Input Options Work With TAPE DEFINEs
FUP Parameter FUP Parameter DEFINE
FUP Parameter Only and DEFINE Only DEFINE
BLOCKIN BLOCKIN is the FUP quits with BLOCKIN is set BLOCKLEN
size of the an error if to BLOCKLEN
buffer used by values conflict
FUP
EBCDICIN FUP performs FUP performs FUP lets label EBCDIC
or translation translation and processing do
XLATE sets DEFINE the translation
OFF, or an error
occurs if values
conflict
RECIN RECIN used for FUP quits with RECIN is set to RECLEN
FUP deblocking an error if RECLEN
values conflict
REELS Error Error FUP ignores REELS
DEFINE
REWINDIN DEFINEs do not affect this N.A.
parameter
SKIPIN Error Error FUP ignores FILESEQ
DEFINE
UNLOADIN DEFINEs do not affect this N.A.
parameter
Table A-2. How FUP Output Options Work With TAPE DEFINEs
FUP Parameter FUP Parameter DEFINE
FUP Parameter Only and DEFINE Only DEFINE
BLOCKOUT BLOCKLEN is FUP quits with BLOCKOUT is BLOCKLEN
set to an error if set to
BLOCKOUT values conflict BLOCKLEN
DENSITYOUT FUP sets FUP sets FUP ignores the DENSITY
DEFINE to density, or an DEFINE
DENSITYOUT error occurs if
and sets density values conflict
EBCDICOUT FUP performs FUP performs FUP lets label EBCDIC
or translation translation and processing do
XLATE sets DEFINE the translation
OFF, or an error
occurs if values
conflict
RECOUT RECLEN is set FUP quits with RECOUT is set RECLEN
to RECOUT and an error if to RECLEN,
is used by FUP values conflict and it is then
for blocking used by FUP for
blocking
REWINDOUT DEFINEs do not affect this N.A.
parameter
SKIPOUT Error Error FUP ignores FILESEQ
DEFINE
UNLOADOUT DEFINEs do not affect this N.A.
parameter
Note. For information on SAFECOM commands and their syntax, see the Safeguard
Reference Manual.
Table B-4 lists commands that you can use to manage files with either Guardian or
Safeguard security codes.
fileset is:
[[[ \node.]$volume.]subvolume.| *.]{file-id | *}
fileset-list is:
{ fileset | ( fileset [ , fileset ] ... ) }
command-option | environment-option
command-option is:
COPY copy-option [, copy-option ]...
DUP[LICATE] dup-option [, dup-option ]...
LOAD load-option [, load-option ]...
LOADALTFILE loadaltfile-option [, loadaltfile-option
]...
RELOAD reload-option [, reload-option ]...
environment-option is:
ALLOW allow-option [, allow-option ]...
DISPLAYBITS bitcount
[ NO ] ECHO [ CONFIG[URE] ] [ OBEY ]
[ NO ] DISPLAYALLNUM
IOTIMEOUT time
NETBLOCKSIZE size
[ NO ] PROMPT [ PURGE ]
REPORTWIDTH width
RESTARTUPDATE time
XLATE xlate-table-name [ TEXT|CHARMAP ]
[ IN filename ]
rename-option is:
[ ALTFILE ( key-file-number , filename ) ]
[ PART ( sec-partition-number , [ \node.]$volume ]
[ [ , pri-extent-size [ , sec-extent-size ] ] ) ]
EXIT
[ , DETAIL ]
[ , EXTENTS ]
[ , STAT[ISTICS] [ , PARTONLY ] [ , PARTIAL num ] ]
[ , USER { groupnum , usernum } ]
[ { groupname.username } ]
LICENSE fileset-list
OBEY filename
REPORTWIDTH width
RESTART [ restart-filename ]
QUALEXPR qualifier
qualifier is:
EXCLUDE fileset-list
FROM CATALOG[S] catalog-list
START file-id
WHERE expression
catalog-list is:
[ \node.] $volume [.subvol ] | subvol
expression is:
( expression )
expression AND expression
expression OR expression
NOT expression
file-attribute
OWNER userid
timestamp-field time-conditional time-value
FILECODE conditional-number
EOF conditional-number
file-attribute is:
ALTKEY
AUDITED
BROKEN
COLLATION
CORRUPT
CRASHOPEN
ENSCRIBE
ENTRYSEQUENCED
FORMAT1|FORMAT2
INDEX
[ SHORTHAND | PROTECTION ] VIEW
KEYSEQUENCED
LICENSED
OPEN
PROGID
RELATIVE
[ PRIMARY | SECONDARY ] PARTITION
ROLLFORWARDNEEDED
SAFEGUARD
SQL
SQLPROGRAM
TABLE
TRUSTED
TRUSTME
TRUSTSHARED
UNSTRUCTURED
user-id is:
group-name.user-name
group-name.*
group-number, user-number
group-number,*
timestamp-field is:
CREATIONTIME
EXPIRATIONTIME
LASTOPENTIME
MODTIME
time-conditional is:
AFTER | >
BEFORE | <
time-value is:
[ date ] time
[ time ] date
time is:
hh:mm [:ss]
date is:
dd mmm yyyy
mmm dd yyyy
access mode. A specification that determines the different types of operations (read-write,
read-only, or write-only) that a process can perform against a file. The mode is
established when the file is opened.
alternate key. A data field that provides an alternative access path to the records in a file.
This path differs from the inherent access path (the primary key) defined for the file.
Unlike primary key values, alternate key values need not be unique. An example is an
employee data file. An alternate key might be used for the department number of the
employee. Using this key, a program could access the data records for all employees
who work in a particular department.
American Standard Code for Information Interchange (ASCII). A method of coding data
that consists of seven bits for each character (plus a parity bit). Designed for
synchronous or asynchronous use, the code has 128 standard characters. It can also
be an 8-bit code representing characters and control codes. One of two major codes
used to represent and exchange data. (The other is EBCDIC.)
backup. The hardware and software resources available to recover after a degradation or
failure of one or more system components.
binary. A numbering system based on two’s rather than 10’s. The binary system uses only
the digits 0 and 1.
binary digit (bit). In the binary notation, either of the characters 0 or 1. Smallest unit of
information used in data processing. Eight bits equal one byte. A single unit of
information in a circuit.
block. A data collection that can be read from or written to storage media, or read and
written in a single operation. This grouping increases the efficiency with which the
media are used.
buffer. One of three things: (1) a storage area for a block of data, (2) a storage device used
to compensate for a difference in the rate of data flow when transmitting data from one
device to another, or (3) a temporary or dynamic storage facility. It carries both logical
(use within a program) and physical connotations.
byte. An 8-bit storage unit; the smallest addressable unit of memory representing one
alphanumeric character.
cache. A portion of memory that temporarily stores an entire directory, a set of names, or a
set of files.
catalog. A collection of files that maintains context for virtual disk processes, pools, or
$ZSMS.
COBOL. The compiler and run-time support for the American National Standard
Programming Language Common Business-Oriented Language (COBOL). The code
for most Pathway server processes is written in this language.
direct file. A file that is not on an SMF virtual disk; its logical (external) name and physical
(internal) file are identical. All files on direct volumes are direct, and files on physical
volumes in pools can also be direct.
direct volume. A volume that is not in any pool. All files contained in a direct volume are
direct files.
Disk Compression Program (DCOM). A software product for NonStop systems that
compresses or joins together the pieces of fragmented disk files to increase usable
disk space.
Disk Space Analysis Program (DSAP). A software product for NonStop systems that is
used to analyze use of space on a disk.
Disk Process 2 (DP2). The portion of the operating system software that performs read,
write, and lock operations on disk volumes. The disk process also implements
Enscribe and SQL/MP file types; creates TMF audit-trail records; performs logical
REDO operations for Remote Duplicate Database Facility (RDF); and manages disk
space, disk controllers, and paths to the disks. This disk process provides enhanced
performance, throughput, recoverability, and reliability improvements in high-volume,
online transaction processing situations.
EBCDIC. An 8-bit coded character set. One of two major codes used to represent and
exchange data. (The other is ASCII.) EBCDIC uses the eighth bit for an information bit,
which extends the range of distinct characters to 256.
EDIT file. An unstructured file containing text that can be processed by either of the editors
for NonStop systems—EDIT or PS Text Edit (TEDIT). An edit file typically contains
either source program code or documentation.
end-of-file (EOF) marker. A marker placed at the end of a file to indicate that the file
contains no additional information.
Enscribe. The database file-management software provided as part of the Guardian file
system. The Enscribe software provides access to—and manipulation of—records in a
database on a NonStop system. Files on a Guardian system can be either Enscribe
files or SQL/MP tables.
entry-sequenced file. A file in which each new record is stored at the end of the file in
chronological sequence and whose primary key is the system-generated record
address. Records (rows) can be updated but not deleted.
exclusion mode. A specification that determines the degree of access that competing
processes can have to a file simultaneously (shared access, exclusive access, or
protected access). Exclusion mode is specified when the file is opened.
extent. Either a contiguous area on disk for allocating one file, or a block of physical storage
for allocating files.
extent sizes. The size in bytes of a contiguous area on disk for allocating one file.
external name. The file name known and used by applications. For logically named SMF
files, it is the logical name. For direct files, it is the direct name.
file partitioning. The action of subdividing a large file into smaller units that can be handled
more effectively.
file set. A file or a set of files that can be specified (optionally) with the wild-card option to
define name patterns. The file or file set can usually be qualified with an additional set
of exclusions.
Format 1 file. A Format 1 file is a file created on RVUs preceding D46.00 or G06.00.
Format 2 file. Beginning with the D46.00 and G06.00 RVUs, a Format 2 file can contain
larger blocks and partitions than a Format 1 file.
internal file name. The name by which a file is known to DP2. For a direct file, it is the
same as the file name. For a file that is name-managed by SMF, the physical file name
is on a subvolume beginning with ZYS (for logically named files) or ZYT (for logical
templates). See also physical file name.
keyword. A reserved word with a fixed meaning in a program. It cannot be changed by the
user.
location-independent naming (LIN). The naming of files so that their names do not
determine their physical location.
logical file name. A file name whose volume component is an SMF virtual disk. See also
external name.
logical temporary file. A temporary file (the file name beginning with “#”) created on a
logical volume. The volume component of its file name is that logical volume.
node. A uniquely identified system location on a network. Guardian files are stored at a
particular node, in a specific disk volume and subvolume. The node name is preceded
by a backslash (\); for example, \WEST.
nonaudited file. A disk file that is not flagged for auditing by the HP NonStop Transaction
Management Facility (TMF).
null value. A value indicating that a program has made no explicit assignment to a variable
or field. In the context of the SPI and DSM programmatic interfaces, a field of a
structure has a null value if the application has made no explicit assignment to that
field after calling the SSNULL procedure to initialize the structure.
pad character. A character inserted as fill when insufficient data characters are present to
satisfy a length requirement.
parameter. A name specified in a prepared command for which the user substitutes a value
when executing the command.
partition. The portion of a partitioned table, index, or physical file that resides on a particular
disk volume.
physical file name. The internal name by which a file is known to DP2. For a direct file, this
name is the same as the file name. For a file name-managed by SMF, the physical file
name is on a subvolume beginning with ZYS (for logically named files) or ZYT (for
logical templates). See also internal file name.
PHYSVOL option. A physical volume name optionally specified when a logical file is
created (either programmatically or through a command), instructing SMF to create the
file on that volume. Examples include CREATE interfaces and FUP RELOCATE.
pool. A collection of physical volumes managed by SMF. Also known as storage pool.
pool process. The SMF process managing a storage pool, with the same name as that
storage pool.
qualified file-set list. A list of file sets qualified by a clause (optionally) that selects objects
based on characteristics of the file. To determine which qualifiers are permitted, refer to
the syntax.
record. A basic unit of storage in a file or database pertaining to a particular item. A record
is the smallest logical unit of data that can be read from or written to a file. A record in
a relational table is represented as a row. A record can also be any audit record that
describes the outcome of a transaction. More specifically, a status record is either a
commit record or an abort record. A record can also be a Data Definition Language
(DDL) dictionary object that describes the structure of an Enscribe disk file. A record
usually includes file-creation information, which allows the File Utility Program (FUP) to
create a file from the record structure. If the file is to be key sequenced, a record also
contains the key attributes.
relative files. A file in which each new record is stored at the relative record location
specified by its primary key, and whose primary key is either a user-defined or system-
defined relative record number that indicates the record location. Each record location
can be regarded as a numbered slot that holds one fixed-length record, with the
primary key identifying the slot number. In general use, records can be updated or
deleted, but not lengthened or shortened. In SQL/MP, records (rows) can be updated
or deleted, and VARCHAR columns can be lengthened or shortened.
relocate. To move a logically named file from one physical location on disk to another
without changing its logical name. Offline relocation does not preserve opens.
Safeguard. A system-level security tool that provides users of NonStop systems and
distributed networks with a set of services for protecting the components of the system
or network from unauthorized use. Safeguard services include authentication,
authorization, and auditing.
storage manager. A process responsible for performing storage management for some set
of devices. For example, a pool process is the storage manager for the volumes in its
storage pool.
structured query language (SQL) object program. The object file containing the set of
executable machine language instructions produced from a host language source
program with embedded SQL statements.
subvolume. A part of a fully qualified file name. A subvolume is a named logical area on a
disk where users can store files that are usually related to one another. Any user can
create a subvolume by creating a file and naming the new subvolume as part of the file
specifier. A subvolume ceases to exist when the last file in the subvolume is purged. A
subvolume name consists of one to eight alphanumeric characters, the first of which
must be alphabetic.
super ID. A privileged user who can read, write, execute, and purge all files on the system.
The super ID is usually a member of a system-supervisor group. The super ID has the
user ID 255,255.
unstructured file. An Enscribe file type on a NonStop system that is essentially a byte
array on disk that starts at byte address 0 and continues sequentially upward through
whatever byte address is identified by the end of file (EOF).
virtual disk process (VDP). The SMF process that implements location-independent
naming. This is a process pair that is treated like a physical disk process from an
application viewpoint. Also known as virtual volume process.
virtual volume. A process name used instead of a physical volume name to provide
location-independent naming.
virtual volume process. The SMF process managing a logical volume. This is a process
pair. Also known as virtual disk process.
volume. A physical storage device (disk) for files on a NonStop system. Volume names
always begin with a dollar sign ($). It is the part of the designation that identifies where
users store a document. The volume (like a file cabinet) holds subvolumes (like file
drawers) that contain files (like individual folders).
E File (continued)
FUP LOAD 2-130
EBCDICIN option FUP RELOAD 2-147
FUP COPY 2-41
loading empty 2-131
FUP LOAD 2-133
partitioned
EBCDICOUT option
See Partitioned files
FUP BUILDKEYRECORDS 2-22
relative
FUP COPY 2-46
See Relative file
Empty records, copied with FUP
COPY 2-41, 2-52 Safeguard protected
EMPTYOK option, FUP LOAD 2-131 See Safeguard-protected files
EXIT command 2-71 structured
EXT option See Structured files
FUP DUP 2-65 unstructured
FUP SET 2-168 See Unstructured files
Extent sizes File codes 2-86
specifying File extents
FUP ALTER 2-16 allocating, FUP ALLOCATE 2-6
FUP SET 2-168 specifying with FUP ALTER 2-16
upward rounding 2-178, 2-179 upward rounding at file creation 2-7
EXTENTS option, FUP INFO 2-83 File ID, defined 1-10
FILE option, FUP SET 2-14, 2-176
F File security
changing 2-162
FC command 2-71
changing ownership, FUP GIVE 2-77
File
FUP INFO display 2-87
alternate-key
File set 1-9
See Alternate-key files
File type
audited
FUP INFO display 2-88
See Audited files
setting 2-170
creating with FUP CREATE 2-58/2-61
FILENAMES command 2-73
deleting with FUP PURGE 2-140 FILES command 2-75
DP2 FIRST option, FUP COPY 2-38, 2-131
See DP2 files FOLD option, FUP COPY 2-47
FUP LISTLOCKS display 2-120 Format 1 (files)
key-sequenced description 1-22
See Key-sequenced files INFO DETAIL option 2-100
licensed qualified file sets 1-15
See Licensed files SET option 2-169
loading data
FUP BUILDKEYRECORDS 2-21
V
Variable-length blocked records with FUP
COPY 2-45, 2-49
VARIN option
FUP COPY 2-45
FUP LOAD 2-133
VAROUT option, FUP COPY 2-49
VERIFIEDWRITES option
FUP ALTER 2-13
FUP SET 2-170
VOLS command 2-193
Thank you for taking the time to provide us with your comments.
You can submit this form online, e-mail it as an attachment to pubs.comments@hp.com, fax it to
408-285-5520, or mail it to:
Hewlett-Packard Company
NonStop Enterprise Division
19333 Vallco Parkway, MS 4421
Cupertino, CA 95014-2599
Attn.: Product Manager, Software Publications