Page MenuHomePhabricator

hw troubleshooting: "Comm Error: backplane 0" for wikikube-worker1057.eqiad.wmnet
Closed, ResolvedPublic

Description

  • - Provide FQDN of system.
  • - If other than a hard drive issue, please depool the machine (and confirm that it’s been depooled) for us to work on it. If not, please provide time frame for us to take the machine down.
  • - Put system into a failed state in Netbox.
  • - Provide urgency of request, along with justification (redundancy, dependencies, etc): Medium to high as we already have 5 servers of this k8s cluster in this exact state
  • - Describe issue and/or attach hardware failure log. (Refer to https://wikitech.wikimedia.org/wiki/Dc-operations/Hardware_Troubleshooting_Runbook if you need help)
  • - Assign correct project tag and appropriate owner (based on above). Also, please ensure the service owners of the host(s) are added as subscribers to provide any additional input.

I am seeing the following in logs after a failed reimage on wikikube-worker1057 (previously known as kubernetes1038):

racadm>>getsel                         
Record:      1
Date/Time:   08/09/2023 02:06:35
Source:      system
Severity:    Ok
Description: Log cleared.
-------------------------------------------------------------------------------
Record:      2
Date/Time:   08/09/2023 02:15:20
Source:      system
Severity:    Ok
Description: OEM software event.
-------------------------------------------------------------------------------
Record:      3
Date/Time:   08/09/2023 02:15:21
Source:      system
Severity:    Ok
Description: OEM software event.
-------------------------------------------------------------------------------
Record:      4
Date/Time:   08/09/2023 02:15:28
Source:      system
Severity:    Ok
Description: C: boot completed.
-------------------------------------------------------------------------------
Record:      5
Date/Time:   08/09/2023 02:15:28
Source:      system
Severity:    Ok
Description: OEM software event.
-------------------------------------------------------------------------------
Record:      6
Date/Time:   12/06/2024 12:27:25
Source:      system
Severity:    Critical
Description: The System Configuration Check operation resulted in the following issue: Comm Error: Backplane 0.
-------------------------------------------------------------------------------

I have power cycled via racadm but it hasn't fixed itself.

Event Timeline

The following commands have to be executed when the host is back (just noting it down so I don't forget it):

cookbook sre.hosts.reimage --puppet 7 --new -t T377876 --os bookworm wikikube-worker1057
./add_k8s_node.py --netbox-token $NETBOX_TOKEN --netbox-commit --task-id T377876 wikikube-worker1057.eqiad.wmnet
homer 'cr*eqiad*' commit 'T377876'
cookbook sre.k8s.pool-depool-node --k8s-cluster wikikube-eqiad -t T377876 pool wikikube-worker1057.eqiad.wmnet

and add the host to the relabel task for DC-Ops.

JMeybohm renamed this task from Comm Error: backplane 0 when reimaging wikikube-worker1057 to hw troubleshooting: "Comm Error: backplane 0" for wikikube-worker1057.Jan 6 2025, 1:40 PM
JMeybohm renamed this task from hw troubleshooting: "Comm Error: backplane 0" for wikikube-worker1057 to hw troubleshooting: "Comm Error: backplane 0" for wikikube-worker1057.eqiad.wmnet.
JMeybohm assigned this task to Jclark-ctr.
JMeybohm updated the task description. (Show Details)
JMeybohm updated the task description. (Show Details)

Reimaged server without issues. it was posted onto T381789 ticket by mistake

Thanks @Jclark-ctr for the quick help and running the reimage one more time. The host looks good to me now.

I executed our add_k8s_node.py script, run homer and pooled the node, so this host is also done regarding T377876.