-
Notifications
You must be signed in to change notification settings - Fork 3.9k
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
4.1, by @Ayanda-D: stop QQ replicas when a QQ is forced to shrink to a single replica (backport #12468) #12481
Conversation
Cherry-pick of d9de6d9 has failed:
Cherry-pick of 10dbde1 has failed:
Cherry-pick of b03637f has failed:
Cherry-pick of c9d97e6 has failed:
Cherry-pick of 16170d0 has failed:
To fix up this pull request, you can check it out locally. See documentation: https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/reviewing-changes-in-pull-requests/checking-out-pull-requests-locally |
Already backported manually. |
This is #12427 by @Ayanda-D.
Shrinking operations did not stop QQ replicas. This was easy to miss because QQs are usually shrunk before a node is removed from the cluster.
However, there is a scenario where this is not the case. If some nodes (replicas) need to be replaced, in particular when a majority of nodes cannot be recovered for any reasons, the recovery process will involve shrinking a QQ to just one member so that it has an online quorum (of 1 node out of 1) before new replicas can be added.
For this to succeed, the older replicas must be stopped and deleted from the (QQ) cluster.
This is an automatic backport of pull request #12468 done by Mergify.