-
Notifications
You must be signed in to change notification settings - Fork 20
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Prevent infinite CAPIProvider reconciliations #908
Prevent infinite CAPIProvider reconciliations #908
Conversation
818568c
to
cab64f2
Compare
…atches Signed-off-by: Andrea Mazzotti <[email protected]>
Signed-off-by: Andrea Mazzotti <[email protected]>
Signed-off-by: Andrea Mazzotti <[email protected]>
Signed-off-by: Andrea Mazzotti <[email protected]>
Signed-off-by: Andrea Mazzotti <[email protected]>
cab64f2
to
c9df7d5
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There seems to be no regression in tests which verify RV stability on the secret resource, and SSA patch is notoriously unpredictable. I can also see one of the flakes in
turtles/internal/sync/provider_sync_test.go
Line 350 in a45c388
g.Expect(conditions.Get(capiProvider, turtlesv1.CheckLatestVersionTime).Message).To(Equal(fmt.Sprintf("Updated to latest %s version", CAPIVersion))) |
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
thanks!
What this PR does / why we need it:
The CAPIProvider syncing secret is getting a new
resourceVersion
even on empty patch. This creates an infinite loop where the CAPIProviderReconciler is continuously invoked due to the secret version change.This PR tries to address this, however further issues escalate after applying it.
For example there is a conflict between the
rancher-turtles-controller-manager
and thecluster-api-operator
when it comes to managing some resources.For example giving the following bootstrap, there is a conflict in how finalizers are handled:
Which issue(s) this PR fixes (optional, in
fixes #<issue number>(, fixes #<issue_number>, ...)
format, will close the issue(s) when PR gets merged):Fixes #892
Special notes for your reviewer:
Checklist: