Don't reset cluster and collect capz controller log if cluster creation timeout #4534
Labels
kind/bug
Categorizes issue or PR as related to a bug.
priority/backlog
Higher priority than priority/awaiting-more-evidence.
/kind bug
[Before submitting an issue, have you checked the Troubleshooting Guide?]
What steps did you take and what happened:
[A clear and concise description of what the bug is.]
Notice: The issue is about collecting log not the cluster failure below.
There's a cluster creation failure recently and its capz controller log is lost.
https://prow.k8s.io/view/gs/kubernetes-jenkins/logs/cloud-provider-azure-master-ipv6-capz-1-26/1754248546623688704
The reason why log is lost is that the script reset the mgmt cluster
capz
after workload cluster creation failure. I think the logic should be improved.What did you expect to happen:
CAPZ controller log is collected after cluster creation timeout.
Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]
Environment:
kubectl version
):/etc/os-release
):The text was updated successfully, but these errors were encountered: