rpc: retry server busy #3758
Merged
rpc: retry server busy #3758
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
Closed
Add RPC retry with exponential back-off in case of quic `server busy` error. Reduce synchronous part of accept loop. With the bump to quic-go v0.15.5 (scionproto#3732) the quic library now enforces a max accept queue length. With our approach of every RPC is a new quic connection, and some work being done synchronously in the accept loop this bites on CI heavily. With the exponential back-off and immediately spawning a go routine, we can improve failure rate significantly. But this should not be the final solution. Applications need to be able to deal with requests failing. Currently, we only do one shot testing, which might be too strict. For investigation results, see the buildkite builds associated with the draft PR scionproto#3750.
lukedirtwalker
approved these changes
May 11, 2020
There was a problem hiding this comment.
Reviewed 5 of 5 files at r1.
Reviewable status:complete! all files reviewed, all discussions resolved (waiting on @scrye)
scrye
approved these changes
May 11, 2020
There was a problem hiding this comment.
Reviewable status:
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Outdated suggestions cannot be applied.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.
Add RPC retry with exponential back-off in case of quic
server busyerror.
Reduce synchronous part of accept loop.
With the bump to quic-go v0.15.5 (#3732) the quic library now enforces a
max accept queue length. With our approach of every RPC is a new quic
connection, and some work being done synchronously in the accept loop
this bites on CI heavily.
With the exponential back-off and immediately spawning a go routine, we
can improve failure rate significantly.
But this should not be the final solution. Applications need to be
able to deal with requests failing. Currently, we only do one shot
testing, which might be too strict.
For investigation results, see the buildkite builds associated with
the draft PR #3750.
This change is