Repository navigation
Re-initalize failed Kubernetes clusters< 8000 span class="sr-only"> - #7234
Conversation
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: tstromberg The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/ok-to-test |
|
Error: running mkcmp: exit status 1 |
|
/ok-to-test |
|
|
||
| if rr, err := k.c.RunCmd(exec.Command("/bin/bash", "-c", cmd)); err != nil { | ||
| return errors.Wrapf(err, "kubeadm reset: cmd: %q", rr.Command()) | ||
| rr, derr := k.c.RunCmd(exec.Command("/bin/bash", "-c", cmd)) |
There was a problem hiding this comment.
This code seems similar to Kic and None Driver Stop()... I wonder if we could get rid of those (not sure) since we have it here.
given if minikube stop calls delete cluster....we could get rid of those?
|
for this PR, we still see the bind problems: |
Correct, this does not prevent the error from happening, it only provides recovery from that error. There isn't enough detail available to show why the dashboard failed to deploy. |
If we fail to bring-up a Kubernetes cluster, this PR will reset it and bring it up again.
This helps primarily for two cases:
nonedriver initialization where untracked Kubernetes pods are running from a previous install.Fixes #6425 and #6312 and numerous other errors which resemble: