Load balancing fixes #89
Merged
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
Allows for transfers to be disabled which can be useful in some situations and evaluating performance.
jessejlt
approved these changes
Aug 22, 2023
This is a temp workaround for slow peers to reduce the sending node getting backed up. The transfer/upload stack needs some larger rework to better handle overload and large backups.
Collector connections (and other clients) use HTTP keep-alive which prevents connections from being closed on the server. This can cause problems when scaling out ingestor pods because the existing connections persist and are not balanced to new pods. This decreases the idle timeout to forces connections to close so a new connection can be established which can land on a different host.
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Outdated suggestions cannot be applied.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.
This has a few changes to improve load balancing where new nodes are not used until the clients close connections. This decrease the idle timeout on ingestor so that they are closed more regualarly.
For segment transfers, this adds a little bit of concurrency to handle slow peers, but this fix really isn't sufficient and a larger follow up PR will handle this more reliably.
The last bit is to add an option to disable peer discovery/transfer which is useful for performance testing or validating the overhead of peer transfers/COGS. It can also be used to prevent the slow peers for transfers as s stop-gap fix.