i/p 590-595 Network Integrity Check

v.15 11/16/2007
iSeries and pSeries 590/595 Network Integrity Check

• This document was produced by John T. Stephenson, Poughkeepsie, NY, USA.

• This is intended for support purposes, and as such, is an ongoing and improving
document -- please send comments to fatjohn@us.ibm.com. If possible, include a
copy of the actual text as it appears in the document and include the modifications
you think are required.

The procedures in this document have been created to resolve IP addressing problems on
the 590/595 private service network. IP addressing problems are known to cause
unexpected incidents. However, this document DOES NOT apply to 590/595 in cluster
environment.

These checks will be performed by clients and servicers prior to certain 590/595
upgrades. These checks should also be run anytime an IP addressing problem is
suspected.

The Checks to be performed are dependent upon the HMC/Server configuration. Only
those Checks applicable to your configuration should be performed.

Usage
There are 4 Network Checks contained in this document. They have headings Network
Check 1, 2, 3 and 4. These should be run before performing server upgrades and they
should be run if you suspect system errors due to network issues. There are also recovery
procedures with headings MESTIP1, 2, 3 and 4. The MESTIP recovery procedures
should only be run when directed to do so by one of the Network Check procedures.

Note: Please ensure that you obtain from hardware support both pesh password (for
HMC) and celogin password (for ASMI) for the date when the Network Integrity Check
is to be performed. This information will be required in Network Check 4, steps 3 and 4.
In addition, obtain the root password (for HMC) from customer if any recovery
procedures are to be performed (MESTIPs).

-1-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

Table of Contents
Prerequisite firmware check
Network Check 1
Network Check 2
Network Check 3
Network Check 4
MESTIP1
MESTIP2
MESTIP3
MESTIP4

Prerequisite firmware check

Client task
The following set of checks will be done by clients prior to all significant hardware upgrades
to 9119-590, 9119-595, and 9406-595 servers. This check is to prevent an unexpected outage
of any server on the network or a customer impact event on the server being upgraded.

Since this document is primarily intended for use with MES upgrades, it is assumed that the
servers meet minimum firmware requirements, (firmware versions SF230 and BP230 or
higher plus the prerequisite HMC level).

Code level recommendations can be viewed at

https://www14.software.ibm.com/webapp/set2/sas/f/power5cm/supportedcode.htmlFirmware
prerequisites can be viewed by selecting the ‘hardware’ tab at
https://www-912.ibm.com/e_dir/eServerPrereq.nsf

Are the firmware prerequisites installed?

-2-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

Network Check 1
Client and Installer task
Verify DHCP addressing
Static addressing is not supported for the 9119-590, 9119-595, and 9406-595. In addition,
stale addresses can cause unexpected results.

If your HMC is configured as a DHCP server on a private network and is not

communicating correctly with the managed system, or you have recently modified your
network configuration (moved a managed system, replaced an HMC, or added a second
HMC), use the Network Check 1 procedure to determine if your DHCP-managed private
network is configured correctly.

Note: If your HMC is set up as a DHCP server on a private network, the mksysconn
command should not be used to add a static connection for models 575/590/595. Follow
Network Check 1 and use the rmsysconn command to remove IP addresses which have
not been assigned to the servers by DHCP.

Warning: If rmsysconn is used to remove a currently assigned and working static IP

address, that address will be lost and the Server & Frame Restart Procedure (in
MESTIP1) will need to be performed to recover the connection.
Has the DHCP server addressing been verified?

Network Check 1

Abstract: Verify all IP addresses

Description: This procedure guides you through verification that all known devices
(BPCs and FSPs) demonstrate connectivity and there are no extraneous IP addresses.
This information is necessary during the IP address verification process.

When to perform? All configurations

Procedure:

1. Open a rshterm window on the HMC

2. Create a list of the HMC IP connection addresses in Appendix A. In the rshterm
window, type the following command:
lssysconn -r all

Note: The command output displays information in the following format for service
processors and Bulk Power Controllers (BPCs): ). A minimum of two (2) Service
Processors and two (2) Bulk Power Controllers must be displayed. When a Powered
Expansion Frame is a component of the server, two (2) additional Bulk Power
Controllers will be displayed

-3-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

Example output for a CEC Frame only:

resource_type=sys,type_model_serial_num=9119-595*0200WXYZ0,sp=primary,ipaddr=<unique>,state=Connected
resource_type=sys,type_model_serial_num=9119-595*0200WXYZ0,sp=secondary,ipaddr=<unique>,state=Connected
resource_type=frame,type_model_serial_num=9458-100*ppppppp,side=a,ipaddr=<unique>,state=Connected
resource_type=frame,type_model_serial_num=9458-100*ppppppp,side=b,ipaddr=<unique>,state=Connected

3. Create a list of the IP addresses in Appendix B created by the HMC DHCP server. In
the rshterm window type the following command:

lshmc -n -F clients

4. Compare the output from the two commands entered above. Any IP address found in
the output list of the lssysconn command that is not also found in the output list of the
lshmc command must be removed.

Notes:

a. When comparing the output from the lssysconn and lshmc commands, if an
IP address is displayed in the output for both the lssysconn and lshmc
commands, the IP address was assigned by the HMC DHCP server and the
connection is being managed by the HMC DHCP server.

b. There may be an issue with this step if other non-575/590/595 servers have
been configured with static addresses on this network interface. The other
servers should use DHCP addresses on this network or be moved to another
network.

5. If any IP addresses needs to be removed, in the rshterm window type the following
command for each IP address to be removed.

rmsysconn -o remove --ip [ip address]

Example usage:

rmsysconn -o remove --ip 192.168.0.117

6. Repeat these steps on the redundant/backup HMC if on exists.

-4-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

7. In the rshterm window, use the ping command to check each IP address to ensure
good connectivity.

This step also attempts to verify there is no VLAN crosstalk when done in the
redundant HMC environment.

ping [IP address]

Example:

ping 192.168.0.115

For each IP address you ping, you should get a response similar to that shown in the
following example.

If an IP address you ping fails, the command will respond with a Request timed out
as shown in the following example.

-5-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

If you have followed this procedure and not all of your connections appear to be
active (for example, two connections for each 590/595 system, one for each 575
system, and two for each frame), call an authorized Service Representative for
additional support. A Service Representative should follow the Server & Frame
Restart Procedure section of MESTIP1 found later in this document.

-6-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

Network Check 2
Client and Installer task
Verify DHCP address ranges are different on redundant HMC configurations.
When redundant HMCs are configured, unexpected network crosstalk can occur if the
DHCP servers are configured to the same address range. Use the Network Check 2
Procedure to verify that the HMCs are set to different ranges.
Have you confirmed that the DHCP address ranges on the redundant HMCs are different?

Network Check 2 Procedure

Abstract: Verify DHCP address ranges are different

Description: This procedure guides you through verification that all known devices
demonstrate connectivity and there are no extraneous IP addresses.

When to perform? For Redundant/Dual HMC configurations only

Procedure:

Perform the following steps on each HMC.

1. In the Navigation Area, expand the HMC that you want to work with. HMC’s are
listed by hostname or IP address.

2. Expand HMC Management.

3. Click HMC Configuration.

4. In the Contents pane, click Customize Network Settings.

5. Click the LAN Adapters tab.

6. Select Ethernet eth0 in the LAN adapters list.

-7-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
7. Click the Details… button.

8. Click the LAN Adapter tab

9. Record the Address Range of the DHCP Server.

10. Repeat steps 1 through 9 on the 2nd HMC (DHCP server).

11. Compare the address ranges collected from the two HMC’s (DHCP servers) and
verify they are different.

• If the address ranges are different the check is successful.

• If the address ranges are the same, corrective action is required on both the
HMC’s and the attached servers. Service Representatives should follow
MESTIP1 found later in this document.

-8-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Network Check 3
Client and Installer task
Verify that the 192.168.x.x address range has been corrected.
In a configuration with multiple servers and a single HMC, there is an issue that can
result in commands being executed by BPCs or FSPs that are at “default” IP addresses.
Use the Network Check 3 Procedure to verify that the HMC has been corrected to
eliminate exposure to this issue.
Has the 192.168.x.x range been corrected on a system with multiple servers and a single
HMC?

Network Check 3 Procedure

Abstract: Verify that a valid 192.168.x.x IP address range is being used

Description: This procedure confirms that if a 192.168.x.x IP address range is being used
by the DHCP server, it is a valid range.

When to perform? For Single HMC configurations managing multiple servers only

Procedure:

1. In the Navigation Area, expand the HMC that you want to work with. HMC’s are
listed by hostname or IP address.

2. Expand HMC Management.

3. Click HMC Configuration.

4. In the Contents pane, click Customize Network Settings.

5. Click the LAN Adapters tab.

6. Select Ethernet eth0 in the LAN adapters list.

-9-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
7. Click the Details… button.

8. Click the LAN Adapter tab

9. Record the Address Range of the DHCP Server.

10. Check the IP address range

• If the IP address range recorded IS NOT 192.168.0.2 - 192.168.255.254, no

further action related to this Check is required.

• If the IP address range recorded IS 192.168.0.2 - 192.168.255.254, service

action is required on the HMC and the attached servers. Service Representatives
should follow MESTIP2 found later in this document.

-10-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

Network Check 4
Important Notes!

• This section can only be performed by an IBM SSR/CE or other qualified Service
Representative.

• Network Check 1 must be performed successfully before performing this section.

Abstract: This procedure will verify that the SPCN IP addressing (on FSP) is correctly
associated to the BPC IP addresses in the system.

Description: This procedure guides you through the methodology to acquire, convert and
verify SPCN IP addressing.

When to perform? All configurations.

Procedure:

1. Please verify the physical network cabling is correct (HMC –> BPC -> FSP) by
referring to the below diagram. If cables are not attached correctly, then the private
network environment will be corrupted.

-11-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
2. If there is any miscabling found in Step 1, DO NOT proceed the next step. Open a
PMH and contact support for the recovery procedure. You should provide the
following data on the PMH:

- HMC and System Firmware code level (e.g.: HMC v6r.1.2 MH00915 Firmware
SF240_320)
- System configuration (Single/Dual HMC? Single/Dual frame?)
- Cabling description (Which Ports? Which component(s)?)
- Perform a Network Reset on the HMC GUI by the following procedure and
collect the pedbg –c.

Network Reset Procedure:

a) Under Frame Management, right click on the target CEC frame and select
Reset Or Remove Connection:

b) Click YES in the pop-up window:

c) At this point, the Frame state will be shown as “No Connection” and will take
approximately 4 -5 minutes to return to “Standby/Standby”. Ensure the State
is at “Standby/Standby” before you continue with the next step.
d) Collect PEDBG –C (Retain Tip H186878)

-12-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
If the cabling is correct on all Ethernet ports, continue with step 3.

3. Open a rshterm window on the HMC

4. Obtain the HMC Serial Number by executing the command string…

lshmc –v

The output line beginning with *SE will contain the HMC Serial Number.

Example: *SE 10ABCDE

5. Obtain the 9119 server serial number(s) by executing the command string…

lssysconn –r all

The serial number(s) will be found following the asterisk (*) in the output as shown
in the following example.

Example: 9119-595*02ABCDE

6. With the information obtained in steps 2 and 3, contact Support to obtain a pesh
password for the HMC and celogin passwords for all the 9119 servers.

7. If this is NOT a single HMC configuration, perform steps 1 through 4 on the second
HMC.

8. Execute the following steps for each 9119 to verify that the SPCN IP address list (on
FSP) matches the IP addresses assigned by the HMC(s) to the BPCs.

a) In the Navigation Area (left pane),

1) Expand the HMC that you want to work with. HMC’s are listed by
hostname or IP address.

2) Expand Service Applications

3) Click Service Focal Point

-13-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

b) In the Contents Area (right pane), click Service Utilities.

c) A window will appear titled Service Utilities. In this window click the 590
or 595 server MTMS so it is highlighted.

d) From the menubar, click Selected -> Launch ASM Menu…

e) When the window titled Launch ASM Interface appears, click the OK
button.

f) If a Security Alert appears, click the Yes button.

g) In the ASM window, enter the User ID of celogin and the Password
provided to you by Support then click the Log in button.

-14-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

h) Access the Service Processor Command Line field by expanding System

Service Aids then click Service Processor Command Line in the left pane.

i) In the Service Processor Command Line field enter the command string…

echo "0000D000EC220000" | spif -

Note! Make certain you include the space and minus sign (-) at the end of
the command string.

This command returns the SPCN Connection Table.

Example of output for a single frame server with a Single HMC:

0000d000 01a2001d ec3c0000 02007a00

00000000 acac11fe fe01fa05 08000005
acac11fe fd

-15-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Example of output for a single frame server with dual HMCs:

0000d000 01a20035 ec3c0000 04007a00

00000005 acc0a8fd ff017a00 01000005
00ac11fe fe027b01 01000000 00ac11fc
fe037b01 00000007 acc0a8fe fd

Example of output for a dual frame server with dual HMCs:

0000d000 01a20069 ec3c0000 04007a00

00fa0501 00090618 fb017b01 00fa0501
00090618 f3027a00 01000001 ac090619
f7037b01 01000000 ac090619 fb3c0100
04007a00 00fa0501 00090618 ef017b01
00fa0501 00090618 f0027a00 01000001
ac090619 f8037b01 01000000 ac090619
fc

j) Record the IP addresses in Appendix C that were listed in the SPCN

Connection Table.

For a single frame server with a single HMC, record bytes 22 through 25
and 34 through 37.

For single frame server with dual HMCs, record bytes 22 through 25, 34
through 37, 46 through 49 and 58 through 61.

For dual frame server with dual HMCs, record bytes 22 through 25, 34
through 37, 46 through 49, 58 through 61, 74 through 77, 86 through 89, 98
through 101 and 110 through 113.

k) Decode the IP addresses by converting the hex value of each byte to its
decimal value. In the table below, use the first hex digit of the address byte
to select a row (gray digits) then use the second hex digit of the address byte
to select a column (orange digit). Record the translation into table under
Appendix C.

example AC = 172

Each set of 4 sequential bytes from the SPCN Connection Table is an IP

address. The Dotted Decimal value will be realized after your conversion.

Example IP addresses using the Single HMC sample data from above:

ac11fefe = 172.17.254.254
ac11fefd = 172.17.254.253

-16-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
X 0 1 2 3 4 5 6 7 8 9 A B C D E F
0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 16 17 18 19 20 21 22 23 24 35 26 27 28 28 30 31
2 32 33 34 35 36 37 38 38 40 41 42 43 44 45 46 47
3 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
4 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
5 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
6 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
7 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
8 128 129 130 131 132 1333 134 135 136 137 138 139 140 141 142 143
9 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
A 160 162 162 163 164 165 166 167 168 169 170 171 172 173 174 175
B 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191
C 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207
D 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
E 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
F 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255

l) Check that the IP addresses from the SPCN Connection Table match what
the HMC(s) are using. If they do not match, go to tip MESTIP3 to correct
stale SPCN addresses.

m) Check that the number of IP addresses is correct. If a server with one HMC
has more than 2 addresses for a single frame server or more than 4 addresses
for a dual frame server, go to disruptive tip MESTIP4 to remove lingering
addresses for each affected server.

-17-
i/p 590-595 Network Integrity Check
v.15 11/16/2007

*** TIPS *** This section does not contain network checks.
This section contains workarounds that are to be run when a
network check detects a problem.

MESTIP1
Recover from redundant HMCs setup on the same DHCP address range.

1. Re-configure the IP address range on the redundant HMC to a different range.

2. Restart the redundant HMC.
3. Execute Server Restart Procedure

Server & Frame Restart Procedure

4. Identify the servers attached to the eth0 network of the redundant HMC
5. Identify the Machine Types and code levels of the servers attached to the eth0
network of the redundant HMC.
6. Perform the following tasks on the attached servers & frames depending on machine
type and code level.

Servers

a. 9119-590, 9119-595 and 9406-595 servers at SF240_xxx, no action

required.
b. 9119-590, 9119-595 and 9406-595 servers at SF235_xxx:
i. reset FSP network configuration on ASMI for each FSP
c. 9119-590, 9119-595 and 9406-595 servers at SF230_xxx:
i. quiesce the partitions
ii. power off the server
iii. EPO off, EPO on
iv. Power on the server
v. Return to customer to start partitions
d. 9118 servers prior to SF240:
i. reset FSP network configuration on ASMI
e. All other servers at SF240_xxx, no action required.
f. All other servers prior to SF240_xxx:
i. Disconnect Ethernet cable at the server for 1 minute. Re-insert
cable.

Frames

g. 9119 frames, 9118 frames, and 24” I/O frames at BP240_xxx, no action
h. 9119 frames, 9118 frames, and 24” I/O frames at BP230_xxx or
BP235_xxx (Not required if frame was EPO off/on in previous steps):

-18-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Note: root login is required to proceed, with the hscpe userid use pesh with
the password generated from to access root.

Check the health of the power subsystem on each BPC and then restart
each BPC, **one at a time** if the health is good.

i. Use the BPC IP addresses obtained on the HMC from the output of
the lssysconn command. Telnet to the BPC-A menu ---> telnet
xxx.xxx.xxx.xxx 8890, where xxx.xxx.xxx.xxx is the IP address of
the BPC.

ii. Enter c 970000 to obtain the BPA Power Status. If the second
byte of returned data is set to 91(example rc =
00:xx91xxxxxxxxxxxxxx where xx=any number) then the health is
good.
iii. Type 'x' to exit the PTLIC menu.
iv. Repeat steps 2 - 4 for BPC-B.

***WARNING*** If BPC-A and BPC-B report 91 then continue. If

there is something different, DO NOT proceed with this instruction. Power
repairs **MUST** be performed before continuing.

Action to reboot BPCs if health is good.

i. Telnet directly to BPC-A ---> telnet xxx.xxx.xxx.xxx, where
xxx.xxx.xxx.xxx is the IP address of the BPC. Use "root" for the
user ID and "FipSroot" for the password.

ii. At the command prompt, enter ‘reboot’ then press ENTER. The
amber LED on that BPC will turn off. Soon that LED will begin to
blink. The LED will stop blinking after a few minutes and will stay
on solid to indicate that the reboot process is complete.

*** WARNING *** WAIT *** BPC-A reboot must be complete to

continue or system CRASH will occur ****

iii. Repeat steps 1 - 2 for BPC-B.

Recovery procedure complete.

-19-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
MESTIP2
Recover from HMC using the default address range.
Option 1
1. Upgrade the HMC to V6R1.0 plus MH00839 (or higher)
2. Activate the DHCP address range
a. In the navigation area, expand the HMC that you want to work with.
HMCs are listed by hostname or IP address.
b. Expand HMC Management.
c. Click HMC Configuration.
d. In the contents pane, click Customize network settings.
e. Click the LAN Adapters tab.
f. Select the eth0 LAN adapter.
g. Click ‘details’.
h. Click LAN Adapter tab
i. Select address range 192.168.128.2-192.168.255.254
j. Click ok
k. Reboot the HMC
3. Restart the servers using the ‘Server & Frame Restart Procedure’ in MESTIP1

Option 2

1. If the client does not want to upgrade the HMC, follow the same steps as option 1 but
select a different address range. Do not use 192.168.0.2-192.168.255.254
2. Restart the servers using the ‘Server & Frame Restart Procedure’ in MESTIP1

Recovery Procedure complete.

MESTIP3

-20-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Recover from SPCN using stale addresses on the active links. Follow this tip when the IP
addresses in the SPCN table are different from the HMC assigned addresses.

Update the system to SF240_299 and BP240_211 (or higher) to resolve the problem. If
the client does not want to update the code to resolve the problem, then follow the steps
below to resolve the problem. However, this procedure will only fix the issue at the
moment and could occur again in the future. The permanent fix for this problem is to
update the firmware to the levels noted above.

Note: To perform this procedure, it will require command line access to the HMC with
root authority which means you must log into the HMC, acquire a PE passcode for the
day and use the “pesh” command to gain full command line access. Ensure that a hscpe
id is set up on the HMC.

Check the health of the power subsystem on each BPC and then restart each BPC, **one
at a time** if the health is good.

1. Use the BPC IP addresses obtained on the HMC from the output of the
lssysconn command. Telnet to the BPC-A menu ---> telnet
xxx.xxx.xxx.xxx 8890, where xxx.xxx.xxx.xxx is the IP address of the
BPC.

2. Enter c 970000 to obtain the BPA Power Status. If the second byte of
returned data is set to 91(example rc = 00:xx91xxxxxxxxxxxxxx where
xx=any number) then the health is good.

3. Type 'x' to exit the PTLIC menu.

4. Repeat steps 2 - 4 for BPC-B.

***WARNING*** If BPC-A and BPC-B report 91 then continue. If there is something

different, DO NOT proceed with this instruction. Power repairs **MUST** be
performed before continuing.

Action to reboot BPCs if health is good.

iv. Telnet directly to BPC-A ---> telnet xxx.xxx.xxx.xxx, where
xxx.xxx.xxx.xxx is the IP address of the BPC. Use "root" for the
user ID and "FipSroot" for the password.

v. At the command prompt, enter ‘reboot’ then press ENTER. The

amber LED on that BPC will turn off. Soon that LED will begin to
blink. The LED will stop blinking after a few minutes and will stay
on solid to indicate that the reboot process is complete.
*** WARNING *** WAIT *** BPC-A reboot must be complete to continue or system
CRASH will occur ****

-21-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
vi. Repeat steps 1 - 2 for BPC-B.

Recovery procedure complete.

-22-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
MESTIP4
Recover from lingering addresses.

Clear the condition on each 9119 server frame non-concurrently by:

1. power off the server
2. Telnet directly to BPC-A ---> telnet xxx.xxx.xxx.xxx, where xxx.xxx.xxx.xxx is the
IP address of the BPC. Use "root" for the user ID and "FipSroot" for the password.
3. Enter ‘userconfig –reset’ (Note: there are 2 dashes before the word “reset”)
4. On servers with SF235_xxx and SF240 xxx, reset the network configuration on
ASMI for each FSP. On servers with SF230_xxx, perform the reset on the primary
FSP.
5. EPO the server
6. On servers with SF240_219, SF240_258, and SF240_261, perform the following to
prevent FSP GARD after EPO-ON {or reference TDR H187952}:

WORKAROUND: There are 2 workarounds which prevent garding after AC power

is restored (ie. EPO on). Select PROCEDURE1 or PROCEDURE2.

PROCEDURE1: Perform this procedure if the HMC is at level 6.1.0 and MH00833
is installed. Otherwise perform PROCEDURE2.
a. Login to the HMC as user hscpe.
b. Right click on the HMC desktop and select "rshterm".
c. Enter 'pesh' and enter the password provided by support
d. Enter 'lssyscfg -r sys' and determine the managed system name
e. Enter 'updlic -o b -m <managed system name>'
This will restart the backup FSP. PROCEDURE1 complete.

PROCEDURE2: Perform this procedure if the HMC is not at level 6.1.0

a. Start ASM Menu for Primary FSP

b. Log on using celogin
c. Issue "rmgrcmd --disp-fsp-info=role" in the "Service Processor Command Line"
menu. This should respond with the FSP role as "Primary FSP".
NOTE: The command line has two "-" symbols before the "disp-fsp-info" keyword.
d. Issue "smgr mfgState" in the "Service Processor Command Line" menu. This
should respond back with "Standby" to indicate the system is in Standby state
(Managed System in power off state)
e. Issue "rmgrcmd --power=off" to power off the backup FSP.
NOTE: The command line has two "-" symbols before the "power" keyword.
f. Issue "rmgrcmd --power=on" to power on the backup FSP.
NOTE: The command line has two "-" symbols before the "power" keyword.
This will restart the backup FSP. PROCEDURE2 complete.

-23-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
7. On servers with SF240 you might observe your frame connections in a "Pending
Authentication" state due to using the “userconfig --reset” command on the BPC.
a. Attempt to clear the condition by selecting "Frame Management" from the
HMC GUI and use "Update Managed Frame Password" to clear the
condition. If the condition persists, follow the workaround below to
resolve the issue {or reference TDR H187613} :
Note: To perform this procedure, it will require command line access to the
HMC with root authority which requires that you acquire a PE passcode for
the day.

b. Log into the HMC command line using rshterm, and use the "pesh"
command to gain full command line access.
Note: Before performing the steps below,, it is recommended that you issue
the following command from the HMC command line: "mksysconn -o auto".

c. Type "telnet <IP address>" of BPC in the Pending Authentication state.

d. Login = "dev".
e. Enter the Password = "FipSdev" (case sensitive)
f. Type "registry -l nets/PasswordStatusFile" (case sensitive so key
carefully)
Output Example:
9458-10002A000A_B# "registry -l nets/PasswordStatusFile
nets/PasswordStatusFile:"
"e0" <<<< Check this value it is the result from command issued A value
of anything other than "00" proves you have the error condition, continue
with next step To set the PasswordStatusFile to a good value, enter the
following:
g. Type "registry -Hw nets/PasswordStatusFile 0". This command sets the
value to "00".
h. Re-enter the command "registry -l nets/PasswordStatusFile". Review the
output per the example above to verify the output is now “00”.
i. Type "exit" to log out of the BPC.
j. From the HMC GUI, select Frame Management. Right click on the
appropriate frame and choose "Reset or Remove Connection".
k. Verify the correct frame component and choose "Yes" to reset the
connection.
l. From the HMC command line, issue "lssysconn -r all" and verify that you
now have all good connections.
8. Power on the system.

Recovery procedure complete.

-24-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Appendix A
List of IP addresses provided from “lssysconn –r all” command on HMC 1

System or BPC s/n Pri/Sec or IP address HEX IP address

A/B Conversion
Example: Example: Example: Example:
9119.595.02AD52E (System) Primary 9.6.25.247 09.06.19.f7
9458.100.99204HM (BPC) B 9.6.25.253 09.06.19.fd

* Hex conversion of IP address needed for Network Check 4

-25-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Appendix A (continued)
List of IP addresses provided from “lssysconn –r all” command on HMC 2

System or BPC s/n Pri/Sec or IP address HEX IP address

A/B Conversion*
Example: Example: Example: Example:
9119.595.02AD52E (System) Primary 9.6.25.247 09.06.19.f7
9458.100.99204HM (BPC) B 9.6.25.253 09.06.19.fd

* Hex conversion of IP address needed for Network Check 4

-26-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Appendix B
List of IP addresses provided from “lshmc –n –F clients” command from HMC 1

-27-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Appendix B (continued)
List of IP addresses provided from “lshmc –n –F clients” command from HMC 2

-28-
i/p 590-595 Network Integrity Check
v.15 11/16/2007
Appendix C
Single Frame Server on Single HMC

Byte Hex IP address Decimal IP address HMC 1

22-25
34-37

Single Frame Server on Dual HMC or Dual Frame Server on Single HMC

Byte Hex IP address Decimal IP address HMC 1 or 2

22-25
34-37
46-49
58-61

Dual Frame Server on Dual HMC

Byte Hex IP address Decimal IP address HMC 1 or 2

22-25
34-37
46-49
58-61
74-77
86-89
98-101
110-113

-29-