Re-organize queues to use for Sidekiq

Dumping too many jobs in the same queue (e.g. the "default" queue) is a
dangerous setup. Jobs that take a long time to process can effectively
block any other work from being performed given there are enough of
these jobs.

Furthermore it becomes harder to monitor the jobs as a single queue
could contain jobs for different workers. In such a setup the only
reliable way of getting counts per job is to iterate over all jobs in a
queue, which is a rather time consuming process.

By using separate queues for various workers we have better control over
throughput, we can add weight to queues, and we can monitor queues
better. Some workers still use the same queue whenever their work is
related. For example, the various CI pipeline workers use the same
"pipeline" queue.

This commit includes a Rails migration that moves Sidekiq jobs from the
old queues to the new ones. This migration also takes care of doing the
inverse if ever needed. This does require downtime as otherwise new jobs
could be scheduled in the old queues after this migration completes.

This commit also includes an RSpec test that blacklists the use of the
"default" queue and ensures cron workers use the "cronjob" queue.

Fixes gitlab-org/gitlab-ce#23370

3 files changed, 12 insertions, 15 deletions

diff --git a/app/workers/repository_check/batch_worker.rb b/app/workers/repository_check/batch_worker.rb
index a3e16fa5212..c3e7491ec4e 100644
--- a/app/workers/repository_check/batch_worker.rb
+++ b/app/workers/repository_check/batch_worker.rb
@@ -1,14 +1,13 @@
 module RepositoryCheck
   class BatchWorker
     include Sidekiq::Worker
-  
+    include CronjobQueue
+
     RUN_TIME = 3600
-  
-    sidekiq_options retry: false
-  
+
     def perform
       start = Time.now
-  
+
       # This loop will break after a little more than one hour ('a little
       # more' because `git fsck` may take a few minutes), or if it runs out of
       # projects to check. By default sidekiq-cron will start a new
@@ -17,15 +16,15 @@ module RepositoryCheck
       project_ids.each do |project_id|
         break if Time.now - start >= RUN_TIME
         break unless current_settings.repository_checks_enabled
-  
+
         next unless try_obtain_lease(project_id)
-  
+
         SingleRepositoryWorker.new.perform(project_id)
       end
     end
-  
+
     private
-  
+
     # Project.find_each does not support WHERE clauses and
     # Project.find_in_batches does not support ordering. So we just build an
     # array of ID's. This is OK because we do it only once an hour, because
@@ -39,7 +38,7 @@ module RepositoryCheck
         reorder('last_repository_check_at ASC').limit(limit).pluck(:id)
       never_checked_projects + old_check_projects
     end
-  
+
     def try_obtain_lease(id)
       # Use a 24-hour timeout because on servers/projects where 'git fsck' is
       # super slow we definitely do not want to run it twice in parallel.
@@ -48,7 +47,7 @@ module RepositoryCheck
         timeout: 24.hours
       ).try_obtain
     end
-  
+
     def current_settings
       # No caching of the settings! If we cache them and an admin disables
       # this feature, an active RepositoryCheckWorker would keep going for up
diff --git a/app/workers/repository_check/clear_worker.rb b/app/workers/repository_check/clear_worker.rb
index b7202ddff34..1f1b38540ee 100644
--- a/app/workers/repository_check/clear_worker.rb
+++ b/app/workers/repository_check/clear_worker.rb
@@ -1,8 +1,7 @@
 module RepositoryCheck
   class ClearWorker
     include Sidekiq::Worker
-
-    sidekiq_options retry: false
+    include RepositoryCheckQueue
 
     def perform
       # Do small batched updates because these updates will be slow and locking
diff --git a/app/workers/repository_check/single_repository_worker.rb b/app/workers/repository_check/single_repository_worker.rb
index 98ddf5d0688..3d8bfc6fc6c 100644
--- a/app/workers/repository_check/single_repository_worker.rb
+++ b/app/workers/repository_check/single_repository_worker.rb
@@ -1,8 +1,7 @@
 module RepositoryCheck
   class SingleRepositoryWorker
     include Sidekiq::Worker
-
-    sidekiq_options retry: false
+    include RepositoryCheckQueue
 
     def perform(project_id)
       project = Project.find(project_id)