Enable container_registry_phase_2_deny_list feature flag in bulk

Production Change

Change Summary

We're about to start Phase 2 of the GitLab.com container registry upgrade/migration (gitlab-org&5523 (closed)).

As for Phase 1 (https://gitlab.com/gitlab-com/gl-infra/production/-/issues/5770), we have a special feature flag that acts as deny list for VIP top-level namespaces that should be left to the end of the migration in order to reduce risk.

We need to enable this feature flag (container_registry_phase_2_deny_list) in bulk for all target top-level namespaces.

This is identical to https://gitlab.com/gitlab-com/gl-infra/production/-/issues/5770, but here we use the ID instead of the path to identify the target top-level groups. This is done to account for potential group renames that may have occurred.

The list of target VIP top-level namespaces can be seen at https://gitlab.com/gitlab-com/gl-infra/production/-/snippets/2292214. This snippet contains a single JSON file with an array of integers, one for each top-level namespace ID.

This change is to execute a script on a production Rails console that will do the following:

  1. Read JSON document from the private snippet.
  2. Iterate over the list of top-level namespace IDs and for each:
    1. Find the corresponding Group object by ID;
    2. Enable container_registry_phase_2_deny_list feature flag for the Group.

This change won't trigger any application behavior, either on GitLab Rails or the Container Registry, as all the functionality sits behind the container_registry_migration_phase2 feature flag which is disabled. Additionally, we're starting Phase 2 by migrating repositories under the gitlab-org group first and then on the free tier only. We'll only move to paid tiers afterward.

Related to gitlab-org/gitlab#350543 (closed).

Change Details

  1. Services Impacted - ServiceGitLab Rails ServiceContainer Registry
  2. Change Technician - @ahyield
  3. Change Reviewer - @ahyield
  4. Time tracking - 10 mins
  5. Downtime Component - NA

Detailed steps for the change

Pre-Change Steps - steps to be completed before execution of the change

Estimated Time to Complete (mins) - 5 minutes

  • Set label changein-progress on this issue
  • Obtain a PAT for the GitLab API with read access to private snippets in this project

Change Steps - steps to take to execute the change

Estimated Time to Complete (mins) - 5 minutes

  • Open rails console with write access in production

  • Execute the following script. Make sure to fill the PAT variable with your private access token and keep the console open for the validation and rollback steps if necessary:

    PAT = "<TOKEN>"
    SNIPPET_URL = "https://gitlab.com/api/v4/snippets/2292214/raw"
    FEATURE_FLAG = :container_registry_phase_2_deny_list
    
    response = Gitlab::HTTP.get(SNIPPET_URL, headers: { 'PRIVATE-TOKEN' => PAT })
    ids = JSON.parse(response.body)
    
    puts "Loaded #{ids.size} group IDs"
    
    before_count = Feature::FlipperGate.where(feature_key: FEATURE_FLAG.to_s).count
    
    ids.each do |id|
      group = Group.find_by_id(id)
      unless group
        puts "WARNING: #{id} not found!"
        next
      end
      Feature.enable(FEATURE_FLAG, group)
    end
    
    after_count = Feature::FlipperGate.where(feature_key: FEATURE_FLAG.to_s).count
    diff_count = after_count - before_count
    
    if diff_count == ids.size
      puts "enabled for {after_count} groups"
    else
      puts "WARNING: Expected feature flag to be enabled for #{ids.size} groups, but only did for #{diff_count}"
    end

Post-Change Steps - steps to take to verify the change

Estimated Time to Complete (mins) - 1 Minute

  • No WARNING: .* messages should be seen during the execution of the script. Otherwise, report to @jdrpereira.
  • Run /chatops run feature get container_registry_phase_2_deny_list on Slack to ensure the feature flag scope was updated.

Rollback

Rollback steps - steps to be taken in the event of a need to rollback this change

Estimated Time to Complete (mins) - 2 Minutes

  • Execute the following snippet:

    before_count = Feature::FlipperGate.where(feature_key: FEATURE_FLAG.to_s).count
    
    ids.each do |id|
        group = Group.find_by_id(id)
        Feature.disable(FEATURE_FLAG, group) if group
    end
    
    after_count = Feature::FlipperGate.where(feature_key: FEATURE_FLAG.to_s).count
    unless after_count == 0
       puts "WARNING: Expected feature flag to be disabled for all groups, but #{after_count} remain"
    end
  • Report any WARNING: .* messages to @jdrpereira.

Monitoring

Key metrics to observe

This change won't trigger any behavior change, either on GitLab Rails or the Container Registry, as all the functionality sits behind the container_registry_migration_phase2 feature flag. Additionally, we're starting Phase 2 by migrating repositories under the gitlab-org group first and then on the free tier only.

Change Reviewer checklist

C4 C3 C2 C1:

  • The scheduled day and time of execution of the change is appropriate.
  • The change plan is technically accurate.
  • The change plan includes estimated timing values based on previous testing.
  • The change plan includes a viable rollback plan.
  • The specified metrics/monitoring dashboards provide sufficient visibility for the change.

C2 C1:

  • The complexity of the plan is appropriate for the corresponding risk of the change. (i.e. the plan contains clear details).
  • The change plan includes success measures for all steps/milestones during the execution.
  • The change adequately minimizes risk within the environment/service.
  • The performance implications of executing the change are well-understood and documented.
  • The specified metrics/monitoring dashboards provide sufficient visibility for the change. - If not, is it possible (or necessary) to make changes to observability platforms for added visibility?
  • The change has a primary and secondary SRE with knowledge of the details available during the change window.

Change Technician checklist

  • This issue has a criticality label (e.g. C1, C2, C3, C4) and a change-type label (e.g. changeunscheduled, changescheduled) based on the Change Management Criticalities.
  • This issue has the change technician as the assignee.
  • Pre-Change, Change, Post-Change, and Rollback steps and have been filled out and reviewed.
  • This Change Issue is linked to the appropriate Issue and/or Epic
  • Necessary approvals have been completed based on the Change Management Workflow.
  • Change has been tested in staging and results noted in a comment on this issue.
  • A dry-run has been conducted and results noted in a comment on this issue.
  • SRE on-call has been informed prior to change being rolled out. (In #production channel, mention @sre-oncall and this issue and await their acknowledgement.)
  • Release managers have been informed (If needed! Cases include DB change) prior to change being rolled out. (In #production channel, mention @release-managers and this issue and await their acknowledgment.)
  • There are currently no active incidents.
  • If the change involves doing maintenance on a database host, an appropriate silence targeting the host(s) should be added for the duration of the change.
Edited by Amy Phillips