Open juan-sg opened 1 year ago
@juan-sg Apologies for late reply. kubectl delete
getting stuck after the last line means that there is a Datashim managed object still around in the cluster.
What could've happened is that the PVC got deleted but, for some reason, the associated dataset
and datasetinternal
objects were not. The following commands will confirm this:
$ kubectl get datasets --all-namespaces
$ kubectl get datasetsinternal --all-namespaces
In case, there are Datasets/DatasetInternal objects still around, you can patch them the same way you did for the PVC, and they would go away (along with the namespace and the rest of the install)
Also, noted the suggestion to improve documentation. if you could list here what you expected and what was missing, we'll update accordingly.
Summary:
kubectl delete
command gets stuck aftermutatingwebhookconfiguration.admissionregistration.k8s.io "dlf-mutating-webhook-cfg" deleted
(which seems to be the last declaration in the YAML).Series of events: This was working as expected (although documentation can be a bit better given that the s3 bucket[s] do[es]n't work OOTB) until I tried to switch from a
stringData
secret to justdata
. This resulted in the PVC getting stuck on 'terminating' status, so I decided to delete the PV and PVC. PVC did not get deleted, so I issuedkubectl patch pvc example-dataset -p '{"metadata":{"finalizers":null}}'
and the PVC was removed but creating the dataset deployment again would not create the PVC this time. I decided to uninstall the whole thing and reinstall, but thekubectl delete -f
command got stuck aftermutatingwebhookconfiguration.admissionregistration.k8s.io "dlf-mutating-webhook-cfg" deleted
.The entire stdout is found below.
Please let me know if there are any additional details needed to troubleshoot this. Thanks!