Delete replication group to wait for delete completion #34
Conversation
kumargauravsharma
requested review from
echen-98,
jaypipes and
nmvk
and removed request for
jaypipes and
nmvk
There was a problem hiding this comment.
Looking great @kumargauravsharma but I think you may have forgotten to git add a file...? :)
| } | ||
| status := *r.ko.Status.Status | ||
| return status == "deleting" | ||
| } |
| @@ -0,0 +1,4 @@ | |||
| if isDeleting(r) { | |||
| return requeueWaitWhileDeleting | |||
There was a problem hiding this comment.
Did you need to git add a conditions.go file? :) I don't see where requeueWaitWhileDeleting is defined?
There was a problem hiding this comment.
requeueWaitWhileDeleting is defined in hooks.go
There was a problem hiding this comment.
doh! sorry, not sure how I missed that :)
| path: Events | ||
| AuthToken: | ||
| is_secret: true | ||
| hooks: |
There was a problem hiding this comment.
Not sure but maybe you need a sdk_find_post_set_output hook as well? (similar to dk_delete_pre_build_request.go.tpl)
There was a problem hiding this comment.
It was not needed.
I have a upcoming separate PR to patch resource which is being deleted, so that kubectl describe provides latest state of the resource as it gets deleted.
But that is independent of this PR.
generator.yaml
Outdated
| sdk_delete_pre_build_request: | ||
| template_path: hooks/sdk_delete_pre_build_request.go.tpl | ||
| sdk_delete_post_request: | ||
| template_path: hooks/sdk_delete_post_request.go.tpl |
There was a problem hiding this comment.
| template_path: hooks/sdk_delete_post_request.go.tpl | |
| template_path: hooks/sdk_delete_post_request.go.tpl |
There was a problem hiding this comment.
Removed the extra space.
kumargauravsharma
force-pushed
the
rg-delete
branch
from
be34d9e to
984b927
Compare
kumargauravsharma
force-pushed
the
rg-delete
branch
from
984b927 to
1ab7e69
Compare
kumargauravsharma
force-pushed
the
rg-delete
branch
from
1ab7e69 to
c004a76
Compare
|
/approve |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: echen-98, jaypipes, kumargauravsharma, nmvk The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
| @@ -0,0 +1,8 @@ | |||
| if respErr == nil { | |||
| if foundResource, err := rm.sdkFind(ctx, r); err != ackerr.NotFound { | |||
There was a problem hiding this comment.
If a DELETE operation on ReplicationGroup has returned no error, doesn't this mean that the resource will always be in a deleting status? Is it necessary to do another sdkFind() operation here?
Further, if a DELETE operation on ReplicationGroup returns no error, is there any chance that the ReplicationGroup can end up not being deleted in the backend Elasticache service? If there is no chance of this, is it really necessary to do anything here other than return nil and let the elasticache-controller reconcile the resource and remove its finalizer in cleanup() and get rid of the CR on the k8s side?
6 participants
Issue #, if available:
Replication group delete takes some time to delete while
kubectl delete -f replication_group.yamlremoves the custom resource record from cluster.Description of changes:
This PR adds logic inside
sdk::sdkDelete()to requeue reconcile while replication group is deleting.By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.