bootstrap on gke fails and unable to cancel bootstrap process
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Canonical Juju |
Triaged
|
High
|
Unassigned | ||
Bug Description
juju bootstrap gke-k8s gke-controller
Creating Juju controller "gke-controller" on gke-k8s/us-central1
Creating k8s resources for controller "controller-
^CInterrupt signalled: waiting for bootstrap to exit
j^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
^CInterrupt signalled: waiting for bootstrap to exit
I had to kill it and now it's left some cruft behind to manually clean up.
ERROR controller "gke-controller" already exists
10:58:14 DEBUG cmd supercommand.go:537 error stack:
/build/
When I went into the pods there was an error about a "unschedulable" pod and in that pod "controller-0" there was an error that it was unschedule-able because "insufficient memory".
I used the default cluster setup so perhaps there's missing documentation and error handling that helps detect this situation?
I created another node-pool with 8gb of ram to try to test it out with.
| Richard Harding (rharding) wrote | #1 |
| Richard Harding (rharding) wrote | #2 |
| Ian Booth (wallyworld) wrote | #3 |
| Changed in juju: | |
| milestone: | none → 2.8-next |
| status: | New → Triaged |
| importance: | Undecided → High |
| Heather Lanigan (hmlanigan) wrote | #4 |
| Ian Booth (wallyworld) wrote | #5 |
I've unregistered the controller locally, removed the workloads, service and ingress, and storage from the gke webui. It still won't let me reuse the controller name so trying to find the list of what it created.