flow: change clustering interface to be a call to NotifyClusterChange #4790
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Previously, components would opt in to clustering notifications through a call to ClusterUpdatesRegistration. Then, if that method returned true, they would be re-evaluated. In preparation for the move to services, the clustering service would not be able to function the same way, as services are not permitted to trigger a re-evaluation of a component. The new interface is a NotifyClusterChange method, which allows components to opt-in to whether they want to ignore that signal or not. This is not reflected in the CHANGELOG as it is a no-op for users.
mattdurham
reviewed
Aug 14, 2023
captncraig
reviewed
Aug 14, 2023
|
|
||
| // Schedule a reload so targets get redistributed. | ||
| select { | ||
| case c.onUpdate <- struct{}{}: |
There was a problem hiding this comment.
This is a case where this change can probably simplify things a lot. Trying to differentiate a clustering update vs an argument update was a real pain. We can probably not send on this channel, and just call c.manager.ClusteringUpdated() directly, and simplify the logic in Run by a bit.
There was a problem hiding this comment.
Can we defer that to a separate PR? I want to minimize the impact of changes as much as possible here.
There was a problem hiding this comment.
yes. I can make that change after.
captncraig
approved these changes
Aug 14, 2023
clayton-cornell
pushed a commit
that referenced
this pull request
Aug 14, 2023
…#4790) * flow: change clustering interface to be a call to NotifyClusterChange Previously, components would opt in to clustering notifications through a call to ClusterUpdatesRegistration. Then, if that method returned true, they would be re-evaluated. In preparation for the move to services, the clustering service would not be able to function the same way, as services are not permitted to trigger a re-evaluation of a component. The new interface is a NotifyClusterChange method, which allows components to opt-in to whether they want to ignore that signal or not. This is not reflected in the CHANGELOG as it is a no-op for users. * flow: remove unused Reevaluate function in ComponentNode * component: add more documentation around NotifyClusterChange * prometheus.operator.*: change Mutex to a RWMutex
github-actions
bot
added
the
frozen-due-to-age
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Previously, components would opt in to clustering notifications through a call to ClusterUpdatesRegistration. Then, if that method returned true, they would be re-evaluated.
In preparation for the move to services, the clustering service would not be able to function the same way, as services are not permitted to trigger a re-evaluation of a component.
The new interface is a NotifyClusterChange method, which allows components to opt-in to whether they want to ignore that signal or not.
This is not reflected in the CHANGELOG as it is a no-op for users.