Run background migrations with a minimum intervaldelay-background-migrations

This adds a minimum interval to BackgroundMigrationWorker, ensuring
background migrations of the same class only run once every 5 minutes.
This prevents a thundering herd problem where scheduled migrations all
run at once due to their delays having been expired (e.g. as the result
of a queue being paused for a long time).

If a job was recently executed it's rescheduled with a delay that equals
the remaining time of the job's lease. This means that if the lease
expires in two minutes we only need to wait two minutes, instead of
five.

Fixes https://gitlab.com/gitlab-org/gitlab-ce/issues/41624

1 files changed, 21 insertions, 2 deletions

diff --git a/spec/workers/background_migration_worker_spec.rb b/spec/workers/background_migration_worker_spec.rb
index 1c54cf55fa0..d67e7698635 100644
--- a/spec/workers/background_migration_worker_spec.rb
+++ b/spec/workers/background_migration_worker_spec.rb
@@ -1,13 +1,32 @@
 require 'spec_helper'
 
-describe BackgroundMigrationWorker, :sidekiq do
+describe BackgroundMigrationWorker, :sidekiq, :clean_gitlab_redis_shared_state do
+  let(:worker) { described_class.new }
+
   describe '.perform' do
     it 'performs a background migration' do
       expect(Gitlab::BackgroundMigration)
         .to receive(:perform)
         .with('Foo', [10, 20])
 
-      described_class.new.perform('Foo', [10, 20])
+      worker.perform('Foo', [10, 20])
+    end
+
+    it 'reschedules a migration if it was performed recently' do
+      expect(worker)
+        .to receive(:always_perform?)
+        .and_return(false)
+
+      worker.lease_for('Foo').try_obtain
+
+      expect(Gitlab::BackgroundMigration)
+        .not_to receive(:perform)
+
+      expect(described_class)
+        .to receive(:perform_in)
+        .with(a_kind_of(Numeric), 'Foo', [10, 20])
+
+      worker.perform('Foo', [10, 20])
     end
   end
 end