Unexpected storage processor reboot

Hi

Yes, you are correct, in this case, the bugcheck occured is nothing but a fancy name of Windows BSD.

But, the alert due to which SPB rebooted is not a false alert. any bugcheck or Windows BSD may occur due to hardware interrupt caused or any driver/application related issues..in this case it is the SPB CPU module which may have caused the hardware interrupt which resulted in the bugcheck error and rebooted the SPB..this clearly states in the event message. It is recommended for SPB replacement.

If you need more investigation, get Dell involved to analyse the dump files generated due to bugcheck. This can provide you exact root cause.

Thanks

Anirudh

Greetings,

Emc161432 talks about the 0x7127ca2a alert.

That's an old primus however I reckon that we have some issue with the SP's CPU Module.

You mighrt need to replace the part.

Machine Check Exception (MCE) is a type of computer hardware error that occurs when a computer's central processing unit (CPU) detects a hardware problem. The error usually occurs due to failure or overstressing of hardware components where the error cannot be more specifically identified with a different error message.

I would recomend you to get an EMC Ticket raised, you would be asked to provide SP collects and DUMP file located at C:\dumps\crash.dmp from the rebooted SP.

Bugchecks sometimes could be hazardous causing memory leaks.

Regards,

Sheron Godfred

Hi null Poore,

I believe we are not running this CX4-120 on R30.524 and most probably it is on R28-29!

Before, you go for any hardware replacements, I would request you to check the below points:

- Get the Health Check Done by Dell Support. (Agreed, there was an SP Reboot due to BugCheck)

- Check if the fault LED on the storage system is Amber.(If you have physical access to the Storage)

- Verify from the Navisphere/Unisphere Fault Status Report if the CPU Module is faulted or if anything is faulted.

- Remember, If you have got Error: 7127ca2a due to BugCheck there is a high chance that CPU Module could go back online. (Do NOT replace it until and unless fault is verified with SP Collects Analysis)

- Lastly, upgrade the FLARE to R30.524 once you are sure the storage system is normal.

PS: We like the SP to get itself back online, but if the hardware is marginal, the problem could return and so L2 or Engineering should review for 'proactive' replacement recommnedations and this FRU can not be done with out verifying the exact fault by analyzing the SP Collects.

Thank you for the feedback.  I will proceed with having SPB replaced.  Is this something that can be done while the array remains up and running on SPA?

Hi

Yes, you can replace SPB without scheduling any downtime as long as SPA is alive.. Always recommended to schedule downtime and then go for replacement of a SP..this is always a best practice. Like Sheron said, plan for FLARE upgrade activity as well to latest version.

Thanks

Anirudh

Yes it can be done online if you have a proper failover.

And replacement may not be the actual resolution.

We also would need to upgrade the FLARE, perhaps the issue would have been created due to a bug in the FLARE.

Regards.

Sheron Godfred