* Fix race condition causing unhealthy repository message to be lost
This commit fixes a race condition in the provisioning repository controller
where the "Repository is unhealthy" message in the sync status could be lost
due to status updates being based on stale repository objects.
## Problem
The issue occurred in the `process` function when:
1. Repository object was fetched from cache with old status
2. `RefreshHealth` immediately patched the health status to "unhealthy"
3. `determineSyncStatusOps` used the stale object to check if unhealthy
message was already set
4. A second patch operation based on stale data would overwrite the
health status update
## Solution
Introduced `RefreshHealthWithPatchOps` method that returns patch operations
instead of immediately applying them. This allows batching all status updates
(health + sync) into a single atomic patch operation, eliminating the race
condition.
## Changes
- Added `HealthCheckerInterface` for better testability
- Added `RefreshHealthWithPatchOps` method to return patch ops without applying
- Updated `process` function to batch health and sync status updates
- Added comprehensive unit tests for the fix
Fixes the issue where unhealthy repositories don't show the "Repository is
unhealthy" message in their sync status.
* Fix staticcheck lint error: remove unnecessary nil check for slice
* provisioning: detect stale sync status and trigger resync
When sync jobs expire and are cleaned up by the expired job cleanup
controller, the Repository sync status remains stuck in Pending or
Working state. This prevents new sync jobs from being queued because
shouldResync() blocks on these states.
This change adds detection logic in shouldResync() to check if a sync
job referenced in the sync status still exists. If the job doesn't exist
(NotFound), we trigger a resync to reconcile the stale state.
Fixesgrafana/git-ui-sync-project#626
* test: remove unused mocks and fix test case
- Remove unused mockRepositoryLister and mockRepositoryNamespaceLister types
- Remove unused imports (labels, listers)
- Remove test case for sync disabled scenario as we don't care about sync enabled state when detecting stale status
* adding some logs to better understand what might be happening
* only focus this PR on improve logging in finalizer handling
* debug log before calling finalizers
* working on finalizers
* removing last todos, adding unit tests
* better use SupportedFinalizers name
* addressing comments
* wip: fix tests and add delete error in status
* chore: codegen
* chore: codegen openapi
* Merge remote-tracking branch 'origin/main' into provisioning-finalisers-fix-2
* update frontend client
* fix: errors in testing
* fix: breaking test
---------
Co-authored-by: Daniele Ferru <daniele.ferru@grafana.com>
Co-authored-by: Ryan McKinley <ryantxu@gmail.com>
Daniele Stefano Ferru
Roberto Jiménez Sánchez
Stephanie Hingtgen
