!227592 (comment 3170197970) wouldn't have happened, if BBM runner picked this up instead.
Considering this a non-blocker.
@krasio - Thanks for the fix! Going ahead since !227592 (comment 3168898482) should not be blocker.
Execute BBM affected by single record table bug
Execute BBMs affected by #590848. This is in addition to !225461 which was not enough as just marked the migrations as paused.
Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.
Resolving since bringing finalization of the migration with 1 record to forefront should be fine.
Related: !227592 (comment 3170197970)
WITH migrations AS (
SELECT *
FROM batched_background_migrations
WHERE queued_migration_version BETWEEN '20250905091200' AND '20260216140430'
AND status = 0 -- paused
)
SELECT m.id, m.job_class_name, m.queued_migration_version
FROM migrations m
LEFT JOIN batched_background_migration_jobs j ON m.id = j.batched_background_migration_id
WHERE j.id IS NULL
AND (
(m.min_value IS NOT NULL AND m.min_value = m.max_value)
OR
(m.min_cursor IS NOT NULL AND m.min_cursor = m.max_cursor)
);
---
id | job_class_name | queued_migration_version
---------+--------------------------------------------------+--------------------------
2001086 | FixIncompleteInstanceExternalAuditDestinationsV2 | 20251027170720
(1 row)
Notes:
The Migration got failed in the main database because of 0005-Patch-finalize-BBM-helper.patch.
Surprising that there were a few matching BBMs in Gitlab.com prod.
thought (non-blocking): Setting the status to active will make those BBMs to get picked up in the next run of BBM workers.
suggestion: Would be nice to not use send.
public_send(