https://rancher.com/ logo
Join Slack
Powered by
hello there! one of my nodes has become "unschedul...
# longhorn-storage
a
hello there! one of my nodes has become "unschedulable", i think it has to do with one of the volumes being in "degraded" state because the replica that should be assigned to it is, for some reason, in state "N/A". I have no idea what to do next 😄 where could I look to address this situation?
b
As long as there are other healthy replicas it should be safe to just delete it and let it recreate a new one.
a
i'm getting "Scheduling FailureReplica Scheduling Failure Error Message: precheck new replica failed"
node is still unschedulable, apparently it wasn't that volume
how do i go about finding out why a node is unschedulable?
b
Look at the events
There's might be clues in the status section of the yaml object too
a
whose events exactly? there's a lot of pods
b
Look in the longhorn-system namespace
Look*
a
yep that's where i saw a lot of pods 😅
b
Yep just gonna have to wade through all the messages
Job security if nothing else
a
so none of the pods running on that node has events 🫠
the only one with events is one in CrashLoopBackOff, but that's because it doesn't have the iscsi stuff as it's a control node - but it's not the one that i have a problem with
also weird, it is "Unschedulable / Ready"
god i'm stupid, for some reason the node was cordoned... 🙃
please forget i ever said anything, and thank you for the info 🙂
b
It happens friend
🙌 1
32 Views
Open in Slack
PreviousNext
Powered by