scheduler: speed up exit #5341
scheduler: speed up exit #5341
Conversation
|
[REVIEW NOTIFICATION] This pull request has not been approved. To complete the pull request process, please ask the reviewers in the list to review by filling The full list of commands accepted by this bot can be found here. Reviewer can indicate their review by submitting an approval review. |
ti-chi-bot
added
the
release-note
han-ian
force-pushed
the
coordinator_exit_speed_up
branch
3 times, most recently
from
c2a4fd6 to
fa8db5b
Compare
Signed-off-by: yin.han <[email protected]>
han-ian
force-pushed
the
coordinator_exit_speed_up
branch
from
fa8db5b to
5f63dfe
Compare
| @@ -364,6 +364,11 @@ func (l *balanceLeaderScheduler) Schedule(cluster schedule.Cluster, dryRun bool) | |||
|
|
|||
| result := make([]*operator.Operator, 0, batch) | |||
| for sourceCandidate.hasStore() || targetCandidate.hasStore() { | |||
| // if coordinator is stopping, speed up exit. | |||
| if l.BaseScheduler.OpController.Ctx().Err() != nil { | |||
There was a problem hiding this comment.
Maybe we can consider adding the context field for all schedulers?
There was a problem hiding this comment.
Every scheduler have a BaseScheduler, and BaseScheduler have opController with a context filed, I think it's no difference.
ti-chi-bot
added
the
needs-rebase
|
@in-han: PR needs rebase. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
What problem does this PR solve?
Issue Number: Close #5274
What is changed and how does it work?
Check List
Tests
For our cluster with more than 2000 tikv-servers, pd transfer leader can be done in 1 second.
Code changes
Side effects
Related changes
Release note