Topic
  • 7 replies
  • Latest Post - ‏2013-02-10T12:14:30Z by Novikov_Alexander
herico
herico
1 Post

Pinned topic system x 3650 7979 b1g SMI handler has reported a PCI SERR.

‏2011-12-16T12:04:20Z |
Hi,
having some problem with this system, RSAII reports this error:
326 E SERVPROC 11/14/11, 17:27:37 CPU 2 IERR, the CPU has been disabled
327 W SERVPROC 11/14/11, 17:27:37 CPU 2 IERR detected, the system has been restarted
328 E SERVPROC 11/14/11, 17:27:17 Bus Uncorrectable Error.
329 E SERVPROC 11/14/11, 17:27:17 CPU 1 IERR, the CPU has been disabled
330 W SERVPROC 11/14/11, 17:27:17 CPU 1 IERR detected, the system has been restarted
331 E SERVPROC 11/14/11, 17:26:27 Bus Uncorrectable Error.
332 E SERVPROC 11/14/11, 17:25:17 Bus Uncorrectable Error.
333 E SERVPROC 11/14/11, 17:25:09 Device signaled SERR on PCI primary. Chassis#=NA Slot#=0 Bus#=0 Dev.ID=0x25e2 Vend.ID=0x8086 Status=0x4010 DevFun#=0xba
334 E SERVPROC 11/14/11, 17:25:09 System Error PCI Bus
335 E SERVPROC 11/14/11, 17:25:09 SMI handler has reported a PCI SERR.

When i try to power on the system, it doesn't show anything on the display.
Already tried to insert other CPUs, but the error is the same

47 E SERVPROC 12/15/11, 13:14:45 CPU 2 IERR, the CPU has been disabled
48 W SERVPROC 12/15/11, 13:14:45 CPU 2 IERR detected, the system has been restarted
49 E SERVPROC 12/15/11, 13:14:25 CPU 1 IERR, the CPU has been disabled
50 W SERVPROC 12/15/11, 13:14:25 CPU 1 IERR detected, the system has been restarted
51 I SERVPROC 12/15/11, 13:13:34 BMC Self Test Results: Passed

So it is obvious that it's CPU to blame for.
What actions should i take?
Thanks.
Updated on 2013-02-10T12:14:30Z at 2013-02-10T12:14:30Z by Novikov_Alexander
  • Novikov_Alexander
    Novikov_Alexander
    6932 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2011-12-17T10:08:02Z  
    Dear Tomas,

    try easy configuration (one CPU, 2 RDIMMs. no cards, no disks ...).
    Do you have previous DSA log? Post it.

    Regards,
    Alexander Novikov
    Russia, Moscow
  • MohammadSafir
    MohammadSafir
    4 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2013-02-09T08:14:29Z  
    Hi,

    i have same problem with same machine,
    please guide me how you solved this problem,

    Already tried swapping following with other system,
    CPU
    Memory
    Power Cage
    VRM
    Power Supplies

    If we reseat both power cables , system will turn on without error but after random time it will restart may be after 20 min of 3Hrs.
    And again we get the same error.
    already upgraded firmware levels to the latest.

    RSAII output
    Index Sev Source Date Time Text

    ----
    --------
    -------------------------------
    1 INFO SERVPROC 02/06/13 13:16:45 Remote Login Successful. Login ID:''USERID' from WEB browser at IP@=192.168.18.25'
    2 INFO SERVPROC 02/06/13 13:14:03 System Complex Powered Down
    3 ERR SERVPROC 02/06/13 13:14:00 CPU 2 IERR, the CPU has been disabled
    4 WARN SERVPROC 02/06/13 13:14:00 CPU 2 IERR detected, the system has been restarted
    5 ERR SERVPROC 02/06/13 11:49:28 Bus Uncorrectable Error.
    6 ERR SERVPROC 02/06/13 11:33:08 CPU 1 IERR, the CPU has been disabled
    7 WARN SERVPROC 02/06/13 11:33:08 CPU 1 IERR detected, the system has been restarted
    and there is CPU light on at LPD Panel.

    Please help
  • Novikov_Alexander
    Novikov_Alexander
    6932 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2013-02-10T09:20:27Z  
    Hi,

    i have same problem with same machine,
    please guide me how you solved this problem,

    Already tried swapping following with other system,
    CPU
    Memory
    Power Cage
    VRM
    Power Supplies

    If we reseat both power cables , system will turn on without error but after random time it will restart may be after 20 min of 3Hrs.
    And again we get the same error.
    already upgraded firmware levels to the latest.

    RSAII output
    Index Sev Source Date Time Text

    ----
    --------
    -------------------------------
    1 INFO SERVPROC 02/06/13 13:16:45 Remote Login Successful. Login ID:''USERID' from WEB browser at IP@=192.168.18.25'
    2 INFO SERVPROC 02/06/13 13:14:03 System Complex Powered Down
    3 ERR SERVPROC 02/06/13 13:14:00 CPU 2 IERR, the CPU has been disabled
    4 WARN SERVPROC 02/06/13 13:14:00 CPU 2 IERR detected, the system has been restarted
    5 ERR SERVPROC 02/06/13 11:49:28 Bus Uncorrectable Error.
    6 ERR SERVPROC 02/06/13 11:33:08 CPU 1 IERR, the CPU has been disabled
    7 WARN SERVPROC 02/06/13 11:33:08 CPU 1 IERR detected, the system has been restarted
    and there is CPU light on at LPD Panel.

    Please help
    Dear Mohammad,

    could you provide DSA log of your server. Use DSA Preboot Edition and post resulted *.xml.gz file.

    Regards,
    Alexander Novikov
    Russia, Moscow
  • MohammadSafir
    MohammadSafir
    4 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2013-02-10T09:24:18Z  
    Dear Mohammad,

    could you provide DSA log of your server. Use DSA Preboot Edition and post resulted *.xml.gz file.

    Regards,
    Alexander Novikov
    Russia, Moscow
    attached is DSA and RSA2 Logs...sorry but logs are in HTML.

    Attachments

  • Novikov_Alexander
    Novikov_Alexander
    6932 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2013-02-10T09:58:59Z  
    attached is DSA and RSA2 Logs...sorry but logs are in HTML.
    Mohammad,

    read the next doc:

    CPU Ierror with Emulex adapter - IBM System x
    http://www-947.ibm.com/support/entry/portal/docdisplay?lndocid=migr-5073430

    Regards,
    Alexander Novikov
    Russia, Moscow
  • MohammadSafir
    MohammadSafir
    4 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2013-02-10T11:01:56Z  
    Mohammad,

    read the next doc:

    CPU Ierror with Emulex adapter - IBM System x
    http://www-947.ibm.com/support/entry/portal/docdisplay?lndocid=migr-5073430

    Regards,
    Alexander Novikov
    Russia, Moscow
    according to this retain tip, emulex HBA cards should be moved to PCI slot 3 & 4. but in my case HBA cards are already on slot 3 & 4.
    Do you think we should upgrade firmware level of HBA cards also.
  • Novikov_Alexander
    Novikov_Alexander
    6932 Posts

    Re: system x 3650 7979 b1g SMI handler has reported a PCI SERR.

    ‏2013-02-10T12:14:30Z  
    according to this retain tip, emulex HBA cards should be moved to PCI slot 3 & 4. but in my case HBA cards are already on slot 3 & 4.
    Do you think we should upgrade firmware level of HBA cards also.
    Mohammad,

    you must do it for fix the issue.

    Regards,
    Alexander Novikov
    Russia, Moscow