Skip to content

Conversation

@manusa
Copy link
Member

@manusa manusa commented Dec 2, 2025

Fixes #522
Replaces #531

The kubernetes.Provider watch-related functionality is restarted for each watch notification that affects its state.

@manusa
Copy link
Member Author

manusa commented Dec 2, 2025

Once this one is merged or the related bug is fixed we should start with the 0.0.55 release which should be stable enough.

Copy link
Collaborator

@nader-ziada nader-ziada left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

}

allClusterManagers[name] = nil
p.managers[name] = nil
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

would these managers have any resources that should be cleaned up

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's a good question.

We don't provide a kubernetes.Manager.Clean method.

The manager itself has a pointer reference to the global config.StaticConfig and an instance of kubernetes.AccessControlClientset.

The clientset has instances of the discovery client, dynamic client, and kubernetes interface. But AFAICT, none of those clients provide clean up methods.

I might have missed something, but I'd say that there's no clean up needed.

The kubernetes.Provider watch-related functionality is
restarted for each watch notification that affects its
state.

Signed-off-by: Marc Nuri <[email protected]>
@manusa manusa force-pushed the fix/provider-watch branch from 565c57e to d3430b3 Compare December 2, 2025 16:21
@nader-ziada nader-ziada merged commit f4e6a94 into containers:main Dec 2, 2025
10 of 11 checks passed
@manusa manusa deleted the fix/provider-watch branch December 3, 2025 05:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Provider watchers to update underlying manager

2 participants