How To
Summary
This document explains how to change your NIM Master or NIM Clients CPUID after you have moved it to new HW to match the system's new CPUID.
Objective
QUESTION:
When does a CPUID change occurs, how does his affect a NIM client and NIM master and how do we correct it?
ANSWER:
When Live Partition Mobility (LPM) is used to move a machine from one physical server to another and the machine is defined as a Network Installation Management client (NIM Client) or is a Network Install Manager itself ( NIM master ), the NIM administrator must update the CPUID attribute of the system to reflect the new hardware value after the LPM migration completes.
CUPID changes when moving your LAPR to new HW. This can be done via LPM as mentioned above, mksysb restore or reassigning luns to new HW.
Steps
- How do I change my NIM Client CPUID after I have moved it to new HW so that I can reestablish communication with the NIM Master.
After you have moved your NIM Client to new HW the CPUID has now been changed. This will impact communication with the NIM Master as the NIM Master is not aware of the client’s CPUID change.
In this example I will use a NIM Client – Instlab163 and NIM Master – Sentinel.
Before I have moved the NIM client to new HW it had the following CPUID:
root@instlab163(/)# uname -a
AIX instlab163 2 7 00FF48B55C00
After I moved it to new HW the CPUID changed:
root@instlab163(/)# uname -a
AIX instlab163 2 7 00F84BE54C00
Communication is now not working between Instla163 and Sentinel.
root@sentinel(/)# nim -o lslpp instlab163 | grep bos.mp
0042-001 nim: processing error encountered on "master":
0042-006 m_lslpp: (From_Master) connect Error 0
0042-008 nimsh: Request denied - sentinel.aus.stglabs.ibm.com
If we take a look in the /var/adm/ras/nimsh.log file on the NIM Client, we can see the following issue:
Mon Mar 15 12:02:05 2021 success: we got 2nd write local id is 00FF48B55C00
Mon Mar 15 12:02:05 2021 success: we got 3rd write remote id is 00FB43B94C00
Mon Mar 15 12:02:05 2021 success: we got 4th write command is /usr/lpp/bos.sysmgt/nim/methods/c_nimpush "/usr/lpp/bos.sysmgt/nim/methods/c_ckspot" "-l" "-ast_applied=3" "-ast_committed=5" "-aplatform=yes" "-aname=instlab163" "-alocation=/usr"
Mon Mar 15 12:02:05 2021 passing OpenSSL setting of 0
Mon Mar 15 12:02:05 2021 error: local value passed, '00FF48B55C00', does not match environment value '00F84BE54C00'
This indicates that the NIM Master thinks that the NIM Client’s CPUID is still '00FF48B55C00', however once compared with the environment value in order to establish the connection it fails due to mismatch as seen above.
To check what CPUID the Master has for the NIM Client we can run the following command on the NIM Master:
root@sentinel(/)# lsnim -l instlab163 | grep -i cpuid
cpuid = 00FF48B55C00 <---- wrong/old CPUID
To check the current CPUID of the NIM Client:
root@instlab163(/)# uname -a
AIX instlab163 2 7 00F84BE54C00
To change the CPUID on the NIM Master’s client definition to match with the actual CPUID of the nim client:
root@sentinel(/)# nim -o change -a cpuid=<new CPUID> <name of client>
root@sentinel(/)#nim -o change -a cpuid=00F84BE54C00 instlab163
root@sentinel(/)# lsnim -l instlab163 | grep -i cpuid
cpuid = 00F84BE54C00 <---- correct CPUID
We can verify connectivity:
root@sentinel(/)# nim -o lslpp instlab163 | grep bos.mp
bos.mp64 7.2.4.0 COMMITTED Base Operating System 64-bit
bos.mp64 7.2.4.0 COMMITTED Base Operating System 64-bit
- How do I change my NIM Master CPUD after I have moved it to new HW to reestablish connectivity with all NIM Clients?
After you have moved your NIM Master to new HW the CPUID has now been changed. This will impact communication with all NIM Clients.
Example:
root@sentinel(/)# nim -o lslpp instlab163 | grep bos.mp
0042-001 nim: processing error encountered on "master":
0042-006 m_lslpp: (From_Master) connect Error 0
On the client:
# tail /var/adm/ras/nimsh.log
Tue Dec 3 14:59:27 2019 count equals 3
Tue Dec 3 14:59:27 2019 count equals 4
Tue Dec 3 14:59:27 2019 got stderr port 1021
Tue Dec 3 14:59:27 2019 success: we got 1st write query is 0
Tue Dec 3 14:59:27 2019 success: we got 2nd write local id is 00FADBC84C00
Tue Dec 3 14:59:27 2019 success: we got 3rd write remote id is 00C655204B00
Tue Dec 3 14:59:27 2019 success: we got 4th write command is /usr/lpp/bos.sysmgt/nim/methods/c_nimpush "/usr/lpp/bos.sysmgt/nim/methods/c_ckspot" "-l" "-ast_applied=3" "-ast_committed=5" "-aplatform=yes" "-aname=lawdbfo1" "-alocation=/usr"
Tue Dec 3 14:59:27 2019 passing OpenSSL setting of 0
Tue Dec 3 14:59:27 2019 error: remote value passed, '00FB43B94C00', does not match environment value '00FB43C85B00'
To correct this, we would have to make changes to the NIM Master itself and make changes to all nim clients as well.
On the NIM Master:
To check the current system CPUID:
root@sentinel(/)# uname -a
AIX sentinel 2 7 00FB43B94C00
To check the NIM Master definition:
root@sentinel(/)# lsnim -l master | grep -i cpuid
cpuid = 00FB43C85B00 ß old/wrong CPUID
To change the CPUID:
# /usr/lpp/bos.sysmgt/nim/methods/m_chattr -a cpuid="new CPUID" master
Example:
root@sentinel(/)# /usr/lpp/bos.sysmgt/nim/methods/m_chattr -a cpuid="00FB43B94C00" master
To verify:
root@sentinel(/)# lsnim -l master | grep -i cpuid
cpuid = 00FB43B94C00
On the NIM Clients:
Since we have moved our NIM to NEW HW and its CPUID has changed, we would need to manually update each NIM Client’s niminfo file. As the niminfo file of each nim client contains the CPUID of the NIM Master it muse be updated to reflect the new CPUID. This can be done as follows:
# mv /etc/niminfo /etc/niminfo.old
# niminit -a name=<client name> -a master=<master hostname> -a connect=nimsh
Example:
root@instlab163(/)# mv /etc/niminfo /etc/niminfo.old
root@instlab163(/)# niminit -a name=instlab163 -a master=sentinel -a connect=nimsh
nimsh:2:wait:/usr/bin/startsrc -g nimclient >/dev/console 2>&1
0513-044 The nimsh Subsystem was requested to stop.
0513-059 The nimsh Subsystem has been started. Subsystem PID is 4391334.
We can verify connectivity:
root@sentinel(/)# nim -o lslpp instlab163 | grep bos.mp
bos.mp64 7.2.4.0 COMMITTED Base Operating System 64-bit
bos.mp64 7.2.4.0 COMMITTED Base Operating System 64-bit
Thank you for the time to read through this guide. I hope you found the information both useful and helpful. If you feel there are any mistakes or inconsistencies, please email me at boris.borisov@bg.ibm.com. If there are any technical questions regarding this document, please follow support procedures and open a PMR by calling 1-800-426-7378, and select the option for software support.
Related Information
Document Location
Worldwide
Was this topic helpful?
Document Information
Modified date:
23 March 2021
UID
ibm16430695