flush cache before segment merge (#4955) #4971
Conversation
Signed-off-by: ti-chi-bot <[email protected]>
|
[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. Reviewer can indicate their review by submitting an approval review. |
|
@ti-chi-bot: This cherry pick PR is for a release branch and has not yet been approved by release team. To merge this cherry pick, it must first be approved by the collaborators. AFTER it has been approved by collaborators, please ping the release team in a comment to request a cherry pick review. 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. |
ti-chi-bot
added
do-not-merge/cherry-pick-not-approved
release-note
ti-chi-bot
added
size/M
| // keep flush until success if not abandoned | ||
| if (left->hasAbandoned()) | ||
| { | ||
| LOG_FMT_DEBUG(left->log, "Give up merge segments left [{}], right [{}]", left->segmentId(), right->segmentId()); |
There was a problem hiding this comment.
should use LOG_DEBUG in 4.0
| // keep flush until success if not abandoned | ||
| if (right->hasAbandoned()) | ||
| { | ||
| LOG_FMT_DEBUG(right->log, "Give up merge segments left [{}], right [{}]", left->segmentId(), right->segmentId()); |
|
I will fix the compile when release-4.0 plan to have a new release version. |
|
This pull request is closed because it's related version has closed automatic cherry-picking. You can find more details at: |
This is an automated cherry-pick of #4955
What problem does this PR solve?
Issue Number: close #4956
Problem Summary:
When do segment split, we try to copy the tail column files in the delta layer of the original segment to the new result
segments. So the new segments may contain data that doesn't belong to its segment range.
And this is ok for most cases, because the redundant data will be filtered out by the segment range when serve the read requests to the segment. So the redundant is invisible in almost all cases.
But when do segment merge later, if the previous redundant data is still not flushed to disk, it will be directly copied to the new merged segment again.
So the redundant data in each segment become visible again after segment merge which may cause potential data incorrectness.
What is changed and how it works?
Flush cache before every merge operation. So the potential unsaved data will be filtered out by the segment range when do merge.
Check List
Tests
Side effects
Documentation
Release note