BL870c i4 System auto restart often

[DATNT]

Hello!

I have two Bl870c i4 running HP-UX B.11.31 . I temporarily call server 1 and server 2 with the same hardware configuration and work in parallel.

Server 1 runs main and works normal but server 2 often automatically restarts about once a week and can't boot, must perform hot shutdown and turn on again to boot into OS.

I checked the event log from Ilo and found some error:
767 informational 2 SFW 2,0,7,0 OS_INIT_CHECKSUM_PASSED This indicates OS INIT CHECKSUM passed. Implementation dependent data field,5684122F1DE105F1 00005E0100001290

768 informational 2 SFW 2,0,7,0 BR_TO_OS_INIT  Branch to OS INIT Implementation dependent data field,
568400721DE105F3 00000010420570A0

769 major 5 HPUX 1,1,5,0 HP-UX_HEX_FAULT_CODE OS legacy hex fault code (Bxxx) Legacy 20-bit PA HEX chassis code,
BF80033855E105F5 000000000002BA00

770 minor 3 HPUX 1,1,5,0 HP-UX_CRASHDUMP_STARTED OS crashdump started (D700) Legacy 20-bit PA HEX chassis code,
7F80033755E105F7 00000000000AD700

775 major 5 ILO 1 BLADE_FAILED_BOOT 08 Aug 2023 17:44:08 This blade has failed to show signs of boot.

I checked the hardware information in the OS and found that the CPU information of Server 2 was wrong, and the server displayed the correct parameters.

Server 2:  Intel(R) Itanium(R) Processor 9540 (194.14 GHz, 24 MB)
8 cores, 16 logical processors per socket
582.43 GT/s QPI, CPU version D0
Active processor count:
3 sockets
24 cores (8 per socket)
24 logical processors (8 per socket)
LCPU attribute is disabled

Server 1:  Intel(R) Itanium(R) Processor 9540 (2.13 GHz, 24 MB)
8 cores, 16 logical processors per socket
6.38 GT/s QPI, CPU version D0
Active processor count:
3 sockets
24 cores (8 per socket)
24 logical processors (8 per socket)
LCPU attribute is disabled

So, is it because Server 2 received wrong CPU information, so it led to Server 2 being overloaded with CPU and frequently restarting automatically?

Thanks for your support!!

[shiva_jr]

Hi, 

It is hard to believe that the server 2 showing as 582.43 GT/s QPI, CPU version DO.  Maximum supported bandwidth of 6.4GT/S.  Please refer this document page No. 1. 
Please reboot the server, don't run any processes, fetch the cpu information again.
Check the crash dump logs in the HP-UX OS again. You may get a specific root cause for the System restart.
If no solution, and the  crash dump error persists, you can contact HPE support team so that you will get the solution.

Regards,
Shiva_JR

[Sunitha_Mod]

Hello @DATNT,

Let us know if you were able to resolve the issue.

If you have no further query and you are satisfied with the answer then kindly mark the topic as Solved so that it is helpful for all community members.