English Community

Datacenter SystemsBladeCenter / Flex Systems
All Forum Topics
Options

35 Posts

07-01-2015

SA

117 Signins

923 Page Views

  • Posts: 35
  • Registered: ‎07-01-2015
  • Location: SA
  • Views: 923
  • Message 1 of 16

x240 -8737 will not complete Post

2018-03-29, 21:09 PM

Hi Team,

 

Having issue with the x240 8737 flaex node adter firmware update, server shutting off during post.

Kindly advice. Logs attached. Thanks


System Health: Critical
   Node Node 06 message: Memory device 22, (DIMM 22) memory scrub failed.
   Node Node 06 message: Group 4, (CPUs) bus uncorrectable error.
   Node Node 06 message: Memory device 24, (DIMM 24) uncorrectable ECC memory error.


3     77777703   E   C   Node_06 2018-03-29T16:30:26Z   3741773555         13623 00Y2774       MM1P
Text: Node 06- Node Node 06 message: Memory device 24, (DIMM 24) uncorrectable ECC memory error.

4     77777703   E   C   Node_06 2018-03-29T16:30:22Z   3741773554         13622 00Y2774       MM1P
Text: Node 06- Node Node 06 message: Group 4, (CPUs) bus uncorrectable error.

22    0600B006   I   N   Node_06 2018-03-29T15:58:23Z   3741773536         13605 00Y2774       MM1P
Text: Node 06- The node Node 06 has entered maintenance mode for up to 20 minutes.
------------------------------------------------------------------------------------------------------
23    04210001   I   N   Node_06 2018-03-29T15:55:44Z   3741773535         13604 00Y2774       MM1P
Text: Node 06- Node Node 06 system-management processor exited update mode.

Solved! See the solution
Reply
Options

53 Posts

07-20-2015

BR

60 Signins

815 Page Views

  • Posts: 53
  • Registered: ‎07-20-2015
  • Location: BR
  • Views: 815
  • Message 2 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 16:24 PM

Put this node in a minimal configuration - 1 DIMM per CPU, no mezzanine cards - then run a bootable update ISO (BOMC).

Add more memory when the system is fully up-to-date. You may have a bad DIMM if the error persists after all updates, if that's the case, contact IBM to get it replaced.

Reply
Options

35 Posts

07-01-2015

SA

117 Signins

923 Page Views

  • Posts: 35
  • Registered: ‎07-01-2015
  • Location: SA
  • Views: 923
  • Message 3 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:09 PM

Thanks for the reply, but have already update the fw for the server.Have  attached the pic.Only LSI sas cont fw failed.The server is not covered under warranty.  Do you suspect its memory issue or systemboard related?

Reply
Options

53 Posts

07-20-2015

BR

60 Signins

815 Page Views

  • Posts: 53
  • Registered: ‎07-20-2015
  • Location: BR
  • Views: 815
  • Message 4 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:21 PM

It's very unlikely that the onboard SAS firmware would be the reason.

You can better isolate moving the memory around. If the problem follows a specific DIMM, then it's most likely a DIMM problem. If the problem stays in a specific slot, swap the processors and if it still persists, then it's most likely a board problem.

Reply
Options

35 Posts

07-01-2015

SA

117 Signins

923 Page Views

  • Posts: 35
  • Registered: ‎07-01-2015
  • Location: SA
  • Views: 923
  • Message 5 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:26 PM
Ok, Is it possible to boot from backup Uefi of server by turning on/off the jumper switch on systemboard?
Reply
Options

53 Posts

07-20-2015

BR

60 Signins

815 Page Views

  • Posts: 53
  • Registered: ‎07-20-2015
  • Location: BR
  • Views: 815
  • Message 6 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:33 PM

It is possible to boot from a secondary copy, but there's no correlation to your problem.

We would only use that when there's a corrupt image or a problem upgrading / downgrading.

Reply
Options

35 Posts

07-01-2015

SA

117 Signins

923 Page Views

  • Posts: 35
  • Registered: ‎07-01-2015
  • Location: SA
  • Views: 923
  • Message 7 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:40 PM
I agree, but can you suggest , what causes the error and what likely is a failed part. Somehow i want to save the systemboard for the customer, any other part can be replaced.
4 77777703 E C Node_06 2018-03-29T16:30:22Z 3741773554 13622 00Y2774 MM1P
Text: Node 06- Node Node 06 message: Group 4, (CPUs) bus uncorrectable error.
Reply
Options

53 Posts

07-20-2015

BR

60 Signins

815 Page Views

  • Posts: 53
  • Registered: ‎07-20-2015
  • Location: BR
  • Views: 815
  • Message 8 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:53 PM

From the logs I'd say it's more likely a DIMM error. The bus uncorrectable message itself is, most of the times, not what we're really looking for, but whatever other error that happened before it. In your case, one memory had an error that the ECC could not repair, so that's the component we would consider first as root cause.

Reply
Options

35 Posts

07-01-2015

SA

117 Signins

923 Page Views

  • Posts: 35
  • Registered: ‎07-01-2015
  • Location: SA
  • Views: 923
  • Message 9 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 20:59 PM
Appreciate your comments, The last thing i want to ask , would the memory failure or fault result in server abrupty going into restart process continously.As per my understanding it should disable the failed mem module or slot and complete the post process. Pls correct me.thanks
Reply
Options

53 Posts

07-20-2015

BR

60 Signins

815 Page Views

  • Posts: 53
  • Registered: ‎07-20-2015
  • Location: BR
  • Views: 815
  • Message 10 of 16

Re: x240 -8737 will not complete Post

2018-03-30, 21:28 PM

For an error like this, the machine would have to restart. But just once. Then the memory would go into a disabled state.

If you are seen recursive reboots after the error, it would need further analysis to understand the reasons.

Reply
Forum Home

Community Guidelines

Please review our Guidelines before posting.

Learn More

Check out current deals!

Go Shop
X

Save

X

Delete

X

No, I don’t want to share ideas Yes, I agree to these terms