Announcement Banner
UNSOLVED

David Len.

updated

6 years ago

DL

David Len.

1 Rookie

•

29 Posts

0

1607

February 12th, 2021 12:00

memory 'dances'

R620 with 128 GB of 16*8GB DIMM

This is a refurbished server that originally had several 'Correctable memory error rate exceeded for..' on 3 DIMM

It has 'settled down', and finally there was a single DIMM, A5, showing consistent, recurring errors

I have replaced the A5 with a 4 GB DIMM, and the errors started showing on A7

I have replaced the original DIMM in the A5 socket and the errors came back to A5

I have swapped A5 and A1 with each other, and got errors reported on A3 and A7

I'm planning to swap A5 and A1 back to original sockets and am pretty sure that errors will be reported on A5 only. 

Any idea as to what might be happening here?

Thanks, David L.

  • DELL-Young E

    Moderator

    •

    5421 Posts

    •

    37 Points

    909

    0

    Posted February 14th, 2021 18:00

    Hi, I found this from an old post, hope this helps.

     

    Social Media and Communities Professional

    Dell Technologies | Enterprise Support Services

    #IWork4Dell

    Did I answer your query? Please click on ‘Mark as Accepted Answer’. ‘Thumbs up’ the posts you like!

  • David Len.

    1 Rookie

    •

    29 Posts

    899

    0

    Posted February 15th, 2021 05:00

    thanks for the quote, but this is not the case here.

    David L.

  • David Len.

    1 Rookie

    •

    29 Posts

    897

    0

    Posted February 15th, 2021 06:00

    Thanks Stefan

    It'll be a while as the winter storm messed up my internet connection, I'll respond once I have some results.

    David L.

  • DELL-Stefan R

    Moderator

    •

    790 Posts

    896

    0

    Posted February 15th, 2021 06:00

    No worries m8.

     

    Take your time to get it fixed and respond back as soon as possible

     

    BR
    Stefan

  • DELL-Stefan R

    Moderator

    •

    790 Posts

    898

    0

    Posted February 15th, 2021 06:00

    Hi David.

     

    First of all, I recommend updating the system to the latest firmware versions.

    If you want to do this as simple as possible try the Platform Specific Bootable ISO, PER620, 20.12.00:

    https://www.dell.com/support/home/en-us/drivers/driversdetails?driverid=y1h70&oscode=ws8r2&productcode=poweredge-r620

     

    Here is also a how-to:
    https://www.dell.com/support/kbdoc/en-ie/000178586/update-poweredge-servers-with-platform-specific-bootable-iso

     

    As the failure occurs on the CPU1 you might also use the Intel Diagnostic Tool to check if the CPU is ok.

    https://www.dell.com/support/kbdoc/en-us/000148306/how-to-run-a-diagnostic-on-the-processor-when-the-system-performance-is-slow

     

    Let us know the results.

     

    BR
    Stefan

  • David Len.

    1 Rookie

    •

    29 Posts

    872

    0

    Posted February 21st, 2021 06:00

    well... this was a no fun experience

    I found out that attempting to install the firmware remotely mush be done through a hop to a local LAN workstation, as the load times otherwise are nerve racking. When done over LAN using a virtual CD the process is slow, but bearable.

    The kicker was that the iDRAC was flashed as well, killing the virtual terminal, without any indication as to what happened.

    When I finally reconnected to the iDRAC and logged in the screen display was there, so I could see that the flashing was completed, but I had quite a while to meditate on the need to get to the client location and pull the plug on the system as it seemed hung, that was not fun.

    Anyway, the firmware was updated, the 2 offending DIMM swapped back to original sockets previously, I'll know in a few days if this fixed the issue.

    Takeaway for me, flashing should be done on location using an attached monitor and a USB.

    David L.

  • David Len.

    1 Rookie

    •

    29 Posts

    838

    0

    Posted February 24th, 2021 12:00

    I have completed the firmware flush successfully, and both CPU passes all tests, the problem persists.

    David L

  • DELL-Chris H

    7 Practitioner

    •

    9682 Posts

    •

    48046 Points

    834

    0

    Posted February 24th, 2021 13:00

    David, 

     

    Would you confirm if the BIOS and iDrac are up to date as well?

     

     

    DELL-Chris H

    Social Media and Communities Professional

    Dell Technologies | Enterprise Support Services

    #IWork4Dell

    Did I answer your query? Please click on ‘Mark as Accepted Answer’. ‘Thumbs up’ the posts you like!

  • David Len.

    1 Rookie

    •

    29 Posts

    817

    0

    Posted February 25th, 2021 06:00

    BIOS 2.9.0
    Firmware 2.65.65.65

    David L.

  • DELL-Stefan R

    Moderator

    •

    790 Posts

    193

    0

    Posted February 25th, 2021 07:00

    Hi David.

     

    thanks for letting us know, so we excluded the CPUs, changing the DIMMS leads to a swapped error on either A3, A5, or A7 - for me, this sounds like a broken mainboard.

    The initially mentioned correctable memory error on 3 DIMM seemed to be the beginning of the now seen issue.

     

    Here is a list of part numbers you might watch out for: 

    KCKR5, XWDCF, VV3F2, PXXHP

     

    ASSY,PWA,PLN,SV,DELL,R620
    ASSY,PWA,PLN,SV,DELL,R620

     

    There are a couple of boards avail, you might check the one that is installed in yours by entering the ServiceTag on our support page:
    https://www.dell.com/support/home/en-us/product-support/product/poweredge-r620/overview

     

    BR
    Stefan