[SDK] reset transient failure count if function made progress on latest retry #6808
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
785f632 to
d3786dd
Compare
SeungjinYang
commented
Aug 21, 2025
sky/server/rest.py
Outdated
Comment on lines
133
to
140
Collaborator
Author
There was a problem hiding this comment.
This change isn't necessary as part of this PR but I think it's a nice improvement where we try immediately on the first instance of transient error to improve responsiveness of the code.
cg505
approved these changes
Aug 21, 2025
Collaborator
Author
|
/quicktest-core |
25c0159 to
96b1e2a
Compare
Collaborator
Author
|
/quicktest-core -k job |
Collaborator
Author
|
test_managed_jobs bw compatibility test is failing, unsure why |
aylei
approved these changes
Aug 22, 2025
Collaborator
aylei
left a comment
There was a problem hiding this comment.
LGTM, thanks @SeungjinYang !
Collaborator
Author
|
test_managed_jobs bw compatibility test is also failing on master and seems unrelated to this issue. merging |
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Outdated suggestions cannot be applied.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.
On transient retries - if we determine the retry to have succeeded, then reset the retry count. This allows the function decorator to deal with any number of transient errors that the function can recover from.
To determine if a retry has succeeded, we use the
line_processedfield as a proxy for progress. The idea is that if the function, on retry, were able to process more lines, the function must have overcome whatever transient error was thrown before.This PR has no effect on functions that don't interact with
line_processedfield, preserving the current behavior where requests are retriedmax_retriestimes.Tested (run the relevant ones):
bash format.sh/smoke-test(CI) orpytest tests/test_smoke.py(local)/smoke-test -k test_name(CI) orpytest tests/test_smoke.py::test_name(local)/quicktest-core(CI) orpytest tests/smoke_tests/test_backward_compat.py(local)