fix: incident enrichment via workflow when using mysql #5098
Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 1 Skipped Deployment
|
… happen if the WorkflowManager._instance already set, and stop called after)
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #5098 +/- ##
==========================================
+ Coverage 46.18% 46.85% +0.67%
==========================================
Files 173 174 +1
Lines 17970 18090 +120
==========================================
+ Hits 8299 8476 +177
+ Misses 9671 9614 -57 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
dosubot
bot
added
size:L
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
PR Summary
Comprehensive fix for incident enrichment visibility in MySQL deployments by addressing UUID format inconsistencies and workflow-related issues across different database dialects.
- Added migration script
2025-07-01-12-19_alertenrichment_uuid_dashes.pyto standardize UUID formatting across database dialects by removing dashes - Modified
keep/api/core/db.pyto introduce__convert_to_uuid_dialect_strfor proper UUID handling across different databases - Added safeguard in tests by setting
KEEP_STORE_WORKFLOW_LOGS=falseto prevent test failures due to db_session fixture table drops - Fixed scheduler initialization edge case in
keep/workflowmanager/workflowmanager.pyto ensure proper workflow manager restarts
7 files reviewed, 3 comments
Edit PR Review Bot Settings | Greptile
keep/api/models/db/migrations/versions/2025-07-01-12-19_alertenrichment_uuid_dashes.py
Show resolved
Hide resolved
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com> Signed-off-by: Kirill Chernakov <yakiryous@gmail.com>
| if not self.scheduler: | ||
| self.logger.debug("Scheduler was stopped, reinitializing it") | ||
| self.scheduler = WorkflowScheduler(self) | ||
|
|
There was a problem hiding this comment.
first, how is it possible that "self.started" is true and not self.scheduler? if its because of raise condition, than it means that right after you start it, the self.started will become false and stop won't work anymore
…eep into fix/5097-wf-incident-enrichment # Conflicts: # keep/workflowmanager/workflowmanager.py
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
…ts.id (UUID) to not to break existing alert enrichments
…eep into fix/5097-wf-incident-enrichment
Closes #5097
Problem
When enriching incident via workflow, python-side str(uuid) is called to get
alert_fingerprintforalertenrichmententry.https://github.com/keephq/keep/blob/v0.45.11/keep/providers/base/base_provider.py#L244-L245
Therefore, when we look up incident enrichments using cast(alert_fingerprint as char) no enrichments will be found, because python will convert uuid to string with dashes, while mysql will convert to string without dashes:
https://github.com/keephq/keep/blob/v0.45.11/keep/api/core/db.py#L3979-L3990
Summary
__convert_to_uuid_dialect_strinkeep/api/core/db.pyto introduce for proper UUID handling across different databases.2025-07-01-12-19_alertenrichment_uuid_dashes.pyto standardize UUID formatting across database dialects by removing dashes.