Repository navigation
Fix Dags with deadlines failing to parse after a database retry - #71628
Closed
ephraimbuddy wants to merge 1 commit into
Closed
ephraimbuddy wants to merge 1 commit into
ephraimbuddy wants to merge 1 commit into
Conversation
ephraimbuddy
force-pushed
the
fix-deadline-dag-parse-retry
branch
from
August 15, 2026 15:34
70a915b to
75aa479
Compare
A transient database error while persisting a Dag with deadlines could turn a recoverable retry into a serialization failure. The Dag processor then discarded the file's parse results, making otherwise valid Dags disappear until a later successful parse.
ephraimbuddy
force-pushed
the
fix-deadline-dag-parse-retry
branch
from
August 15, 2026 16:31
75aa479 to
2d5f8a7
Compare
Contributor
Author
|
Closing this as #70148 fixed same issue and is merged now |
Contributor
|
To be fair, I think your tests are more thorough and still worth saving, but the bug itself is fixed. |
2 tasks done
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Writing a Dag with a deadline rewrites its serialized deadline entries from definitions into UUID references, and it did so on the object the caller passed in. When update_dag_parsing_results_in_db hit a transient database error it rolled back and retried the same objects, which by then held UUID strings where definitions were expected. The retry raised AttributeError while re-encoding them, and since that is not a database error it escaped the retry entirely, so the manager discarded the whole file's parse results rather than one Dag's.
Rewrite a copy instead. Only the "dag" sub-dict is copied, so the bulk of the serialized structure is still shared, and the Dag the caller holds stays in the shape it arrived in however many times the write is attempted.
Was generative AI tooling used to co-author this PR?
Generated-by: Claude Opus 5 following the guidelines