WIP: Add limit for total size of concurrent requests in ingesters's memory, and reject additional requests. #5958
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
Signed-off-by: Peter Štibraný <pstibrany@gmail.com>
Closed
3 tasks
This was referenced Sep 8, 2023
3 tasks
|
Replaced with #6492. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
What this PR does
This PR adds
-ingester.instance-limits.max-inflight-push-requests-bytesinstance limit that rejects additional requests, if ingester is already handling requests that altogether have more specified size. This is similar to distributor's-distributor.instance-limits.max-inflight-push-requests-bytes.This limit can work better than existing
-ingester.instance-limits.max-inflight-push-requests, because "max inflight push requests" set to 1000 has different effect on memory if average request size is 1 KiB or 100 KiB.One possible negative of this change is that ingester now spends extra cpu cycles computing request size. I'm going to test this PR with some workloads and see if this shows as a problem in the profiles. If so, we can easily move the computation to distributor, and pass the size in the WriteRequest.
I'm also going to test this in combination with #5921.
Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX]