Fix schedule head pipeline updatefix-schedule-head-pipeline-update-method

Currently, schedule head pipeline update method which executed after
pipeline creation does not take into account of merge reqeust
pipelines. We should use dedicated `all_merge_requests` method
in this case.

1 files changed, 1 insertions, 7 deletions

diff --git a/app/services/ci/create_pipeline_service.rb b/app/services/ci/create_pipeline_service.rb
index 252f5778644..c17712355af 100644
--- a/app/services/ci/create_pipeline_service.rb
+++ b/app/services/ci/create_pipeline_service.rb
@@ -104,17 +104,11 @@ module Ci
     end
 
     def schedule_head_pipeline_update
-      related_merge_requests.each do |merge_request|
+      pipeline.all_merge_requests.opened.each do |merge_request|
         UpdateHeadPipelineForMergeRequestWorker.perform_async(merge_request.id)
       end
     end
 
-    # rubocop: disable CodeReuse/ActiveRecord
-    def related_merge_requests
-      pipeline.project.source_of_merge_requests.opened.where(source_branch: pipeline.ref)
-    end
-    # rubocop: enable CodeReuse/ActiveRecord
-
     def extra_options(options = {})
       # In Ruby 2.4, even when options is empty, f(**options) doesn't work when f
       # doesn't have any parameters. We reproduce the Ruby 2.5 behavior by