Update cleanup policy parameters
🎩 Context
Cleanup policies for container images is a pretty complex piece of background job logic that take care of removing unwanted tags on container repositories from the Container Registry.
The background execution of cleanup policies have been heavily refactored during the last 12 months. This refactoring has been done behind a feature flag. The refactoring is now humming smoothly on gitlab.com.
Time to remove the feature flag to deliver the refactoring to self-managed instances. The rollout issue is here: #238190 (closed)
Before doing that, we need to update default values and documentation around cleanup policies background execution parameters.
That's exactly what this MR does.
🔬 What does this MR do and why?
- Update the default value of
container_registry_expiration_policies_worker_capacity
in the Application Settings from0
to4
. This defines how many concurrent jobs we can have for the policies execution. - If
container_registry_expiration_policies_worker_capacity
is set to0
, we update it to the default value:4
. - Update the documentation to recommend
200
forcontainer_registry_cleanup_tags_service_max_list_size
. - Update the documentation to present all the default values.
📸 Screenshots or screen recordings
n / a
📐 How to set up and validate locally
- Checkout this MR branch
- In a rails console, check:
ApplicationSetting.first.container_registry_expiration_policies_worker_capacity
value. - Run the migration
- In a rails console, check:
ApplicationSetting.first.container_registry_expiration_policies_worker_capacity
value.
- If in (2.) you had
0
, (4.) will be4
. - If in (2.) you had something else than
0
, (4.) will be the same value.
🚦 MR acceptance checklist
This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.
-
I have evaluated the MR acceptance checklist for this MR.
Related to #238190 (closed)