Conversation
Issue Number: tikv#12680 Signed-off-by: pingyu <yuping@pingcap.com>
Signed-off-by: pingyu <yuping@pingcap.com>
Signed-off-by: pingyu <yuping@pingcap.com>
|
[REVIEW NOTIFICATION] This pull request has been approved by:
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. DetailsReviewer can indicate their review by submitting an approval review. |
|
@pingyu: GitHub didn't allow me to request PR reviews from the following users: haojinming. Note that only tikv members and repo collaborators can review this PR, and authors cannot review their own PRs. DetailsIn response to this: 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. |
|
PTAL, thanks~ |
|
@haojinming: Thanks for your review. The bot only counts approvals from reviewers and higher roles in list, but you're still welcome to leave your comments. DetailsIn response to this:
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 ti-community-infra/tichi repository. |
|
@haojinming: Thanks for your review. The bot only counts approvals from reviewers and higher roles in list, but you're still welcome to leave your comments. DetailsIn response to this:
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 ti-community-infra/tichi repository. |
|
/merge |
|
@iosmanthus: It seems you want to merge this PR, I will help you trigger all the tests: /run-all-tests You only need to trigger If you have any questions about the PR merge process, please refer to pr process. DetailsInstructions 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 ti-community-infra/tichi repository. |
|
This pull request has been accepted and is ready to merge. DetailsCommit hash: 394f255 |
|
cherry pick to release-6.1 failed |
| fn flush_timestamp(&self, region: &Region, reason: &'static str) { | ||
| fail::fail_point!("causal_observer_flush_timestamp", |_| ()); | ||
|
|
||
| if let Err(err) = self.causal_ts_provider.flush() { |
There was a problem hiding this comment.
So it will be blocked every time a new leader is elected or a region is merged?
There was a problem hiding this comment.
Yes. It seems to be a performance issue when API v2 is enabled.
The P99 duration of timestamp flush is 1.0 ~ 1.5ms in a Kingsoft cloud env.
I will think about how to optimize this. As it's not easy to reproduce the causality issues (#12498, #12680), it may also mean that it is unnecessary to flush timestamp for many circumstances.
How do you think ?
There was a problem hiding this comment.
If the ts can be invalidated and flushed asynchronously, then the latency be best optimized.
There was a problem hiding this comment.
If the ts can be invalidated and flushed asynchronously, then the latency be best optimized.
OK, get it.
Signed-off-by: pingyu yuping@pingcap.com
What is changed and how it works?
Issue Number: Close #12680
What's Changed:
In the scenario of region merge, the target region would merge some entries from source region with larger timestamps, when leader of source region is in another store with larger TSO batch than the store of target region's leader.
So we need a flush after commit merge.
Also see analysis in #12680.
Related changes
pingcap/docs/pingcap/docs-cn:No.
release-6.1
Check List
Tests
Side effects
Release note