[YARN-9384] FederationInterceptorREST should broadcast KillApplication to all sub clusters

• Type: Bug
• Status: Patch Available
• Priority: Major
• Resolution: Unresolved
• Component/s: federation
• Labels: federation
• Assignee: Zhang
• Reporter: Zhang
• Created: 13/Mar/19 06:14
• Updated: 01/Aug/19 23:05

Description

Today KillApplication request from user client only goes to home cluster. As a result the containers in secondary clusters continue processing for 10~15 minutes (UAM heartbeat timeout).

This is not an favorable user experience especially when user has a streaming job and sometime needs to restart it (like updating config or resources). In this case, containers created by new job and old job can run at the same time.