Repository navigation
Post rebase and rebuild failure feedback to Jira on the final retry - #611
majamassarini merged 3 commits into
Conversation
Apply the same error feedback pattern from PR packit#603 to the rebase and rebuild agents. When a task crashes (exception path) and exhausts all retries, post the error message to Jira if the task was user-triggered. This ensures users get feedback for unexpected crashes that never reach the workflow's comment_in_jira step. Graceful failures (workflow-level) already post comments via the workflow, so we skip comment_text there to avoid double-commenting. Related: packit#603 Assisted-by: Claude Sonnet 4.5 <noreply@anthropic.com>
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates both rebase_agent.py and rebuild_agent.py to post a final failure feedback comment to Jira when a user-triggered run exhausts all retry attempts on a crash path, while avoiding double-commenting on graceful failure paths. The reviewer suggested deduplicating the list of Jira issues in rebuild_agent.py using dict.fromkeys() to prevent posting duplicate comments on the same issue.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
Use dict.fromkeys() to deduplicate all iterations over rebuild_data.all_jira_issues and all_issues to prevent: - Posting duplicate comments on the same issue - Redundant label updates on the same issue - Triggering multiple email notifications to stakeholders This is a defensive measure to handle edge cases where the issue list might inadvertently contain duplicates. Assisted-by: Claude Sonnet 4.5 <noreply@anthropic.com>
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request adds final failure feedback reporting to Jira for both the rebase and rebuild agents during user-triggered runs, and deduplicates Jira issue keys in the rebuild agent to prevent redundant operations. A critical feedback item was raised regarding the rebase agent, where a plain string is passed to the retry function on graceful failure instead of a serialized ErrorData JSON object, which could lead to downstream parsing errors in the Redis error queue.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request updates the backport, rebase, and rebuild agents to improve retry logic, deduplicate Jira issue keys using dict.fromkeys, and post final failure feedback to Jira once the final retry attempt is exhausted. The review feedback highlights two critical issues in backport_agent.py and rebase_agent.py where instantiating ErrorData with potentially None error values inside an else block could raise an unhandled AttributeError or Pydantic ValidationError, leading to a worker process crash. Code suggestions have been provided to robustly handle these cases with fallback values.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
46cc5a0 to
e68e962
Compare
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request improves error handling, retry mechanisms, and Jira notifications across the backport, rebase, and rebuild agents. It introduces structured ErrorData serialization, posts final failure feedback to Jira on exhausted retries for user-triggered tasks, and deduplicates Jira issue keys to prevent redundant operations. Feedback points out a potential ValidationError in rebuild_agent.py if state.rebuild_error is None, suggesting a fallback string to ensure robustness.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
In the graceful failure path (workflow-level failures), wrap the error string in an ErrorData object before passing to retry(). The error queue expects serialized ErrorData JSON, not plain strings. This matches the pattern used in: - The exception/crash path in all agents - rebuild_agent.py graceful failure path (already correct) Without this fix, downstream consumers of the error queue would fail to parse plain string payloads. Assisted-by: Claude Sonnet 4.5 <noreply@anthropic.com>
e68e962 to
d89ccb9
Compare
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request updates the backport, rebase, and rebuild agents to improve error reporting and failure feedback in Jira. Key changes include wrapping retry errors in serialized ErrorData objects, posting final failure comments to Jira on exhausted retry attempts for user-triggered tasks, and deduplicating Jira issue keys using dict.fromkeys to prevent duplicate operations. I have no feedback to provide as there are no review comments.
Important
The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.
2 participants
Apply the same error feedback pattern from PR #603 to the rebase and rebuild agents. When a task crashes (exception path) and exhausts all retries, post the error message to Jira if the task was user-triggered.
This ensures users get feedback for unexpected crashes that never reach the workflow's comment_in_jira step. Graceful failures (workflow-level) already post comments via the workflow, so we skip comment_text there to avoid double-commenting.
Related: #603
Assisted-by: Claude Sonnet 4.5 noreply@anthropic.com