fix: mirror broadcast error and retry

app/jobs/broadcast_mirror_repo_msg_job.rb

@@ -18,12 +18,29 @@ class BroadcastMirrorRepoMsgJob < ApplicationJob
       id: project.id,
       type: project.numerical_for_project_type
     }
+    # 新增失败重试机制, 重试三次
+    result = broadcast(project, json_data)
+
+    if result == 0 
+      count = 3
+      while count > 0 
+        result = broadcast(project, json_data)
+        if result > 0 
+          break
+        end
+        count -= 1
+      end
+    end
+  end
+
+  def broadcast(project, json_data)
     puts "############ broadcast start.......... "
     puts "############ broadcast channel_name: channel_room_#{project.id}"
     puts "############ broadcast project data: #{json_data} "
 
     cable_result = ActionCable.server.broadcast "channel_room_#{project.id}", project: json_data
 
-    puts "############ broadcast result: #{cable_result == 1 ? 'successed' : 'failed'} "
+    puts "############ broadcast result: #{cable_result > 0 ? 'successed' : 'failed'} "
+    return cable_result
   end
 end

app/jobs/migrate_remote_repository_job.rb

@@ -16,5 +16,6 @@ class MigrateRemoteRepositoryJob < ApplicationJob
     else 
       repo&.mirror&.failed!
     end
+    BroadcastMirrorRepoMsgJob.perform_later(repo.id) unless repo&.mirror.waiting?
   end
 end

app/jobs/sync_mirrored_repository_job.rb

@@ -26,6 +26,7 @@ class SyncMirroredRepositoryJob < ApplicationJob
     result = Gitea::Repository::SyncMirroredService.call(repo.owner.login,
       repo.identifier, token: user.gitea_token)
     repo&.mirror.set_status! if result[:status] === 200
+    BroadcastMirrorRepoMsgJob.perform_later(repo.id) unless repo&.mirror.waiting?
   end
 
 end