patch cancel bug: immediately mark predictions as cancelled #1798
Conversation
This file contains hidden or 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
technillogue
force-pushed
the
syl/patch-cancel-bug
branch
2 times, most recently
from
e3d1c78 to
df1d8d0
Compare
|
I'm not sure how I feel about this, fwiw. By doing this we tell director that work has been stopped when it may not have been, and it seems to me there's a real risk that opens the door to errors later on that may be much harder to debug. If we know for a fact that's not the case for the specific models we're trying to patch, is there a way we can address this in the model? Or only turn it on for those models? |
|
it's definitely not possible to patch it in the model, and I think it should occur for any sufficiently high-throughput model (but we only had one or two of those). we can definitely gate this behind a feature flag, though |
…cellation to succeed while I struggle with #1786, we see queue spikes because not responding to cancellation promptly causes the pod to get restarted. this is a dirty hack to pretend like cancellation works immediately. as soon as we fix the race condition (and possibly any issues with task.cancel() behaving differently from signal handlers), we can drop this. Signed-off-by: technillogue <technillogue@gmail.com>
this makes the cancel patch cleaner, but might be a small speedup for high-throughput outputs Signed-off-by: technillogue <technillogue@gmail.com>
technillogue
force-pushed
the
syl/patch-cancel-bug
branch
from
568552a to
db8542e
Compare
mattt
approved these changes
Jul 22, 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Outdated suggestions cannot be applied.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.
unconditionally mark predictions as cancelled without waiting for cancellation to succeed
while I struggle with #1786, we see queue spikes because not responding to cancellation promptly causes the pod to get restarted. this is a dirty hack to pretend like cancellation works immediately. as soon as we fix the race condition (and possibly any issues with task.cancel() behaving differently from signal handlers), we can drop this.
this is implemented by pretending as though the right event was read from the pipe