Don't block the event loop on context end and related fixes #9
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
mstoykov
requested review from
oleiade and
olegbespalov
and removed request for
a team
olegbespalov
approved these changes
Jan 12, 2024
mstoykov
force-pushed
the
unblockEventLoop
branch
from
92cb8b7
to
6fdef09
Compare
mstoykov
changed the title
Previously setTimeout (and co) might block the event loop in cases where timeout hasn't triggered but the context has been done. While fixing that some other potential races were found. Most of them due to trying to write less code - but it is actually important that if we are using the event loop we do all the stuff on it. Additionally fixing a panic in init context if a timer gets interrupted - all due to using the logger from the lib.State. The test added did tease out a bunch of this issues, and does reproduce the original issue *on* occasion. Unfortunately a better reproducible test seems to be very hard to write
mstoykov
force-pushed
the
unblockEventLoop
branch
from
6fdef09
to
6981438
Compare
mstoykov
commented
Jan 12, 2024
Comment on lines
+216
to
+222
| select { | ||
| case ch <- struct{}{}: | ||
| // wait for this to happen so we don't need to hit the event loop again | ||
| // instead this just closes the queue | ||
| <-ch | ||
| case <-e.vu.Context().Done(): // still shortcircuit if the context is done as we might block otherwise | ||
| } |
There was a problem hiding this comment.
This is the original fix
mstoykov
commented
Jan 12, 2024
Comment on lines
+176
to
+186
| (async () => { | ||
| let poll = async (resolve, reject) => { | ||
| await (async () => 5); | ||
| setTimeout(poll, 1, resolve, reject); | ||
| interrupt(); | ||
| } | ||
| setTimeout(async () => { | ||
| await new Promise(poll) | ||
| }, 0) | ||
| })() | ||
| `) |
There was a problem hiding this comment.
Not certain which parts make more of a difference ... or less, but this is the smallest I got to actually reproduce the issue without needing the whole k6.
olegbespalov
approved these changes
Jan 12, 2024
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.
Previously setTimeout (and co) might block the event loop in cases where
timeout hasn't triggered but the context has been done.
While fixing that some other potential races were found. Most of them
due to trying to write less code - but it is actually important that if
we are using the event loop we do all the stuff on it.
Additionally fixing a panic in init context if a timer gets interrupted -
all due to using the logger from the lib.State.
The test added did tease out a bunch of this issues, and does reproduce
the original issue on occasion. Unfortunately a better reproducible
test seems to be very hard to write