[MESOS-8887] Unreachable tasks are not GC'ed when unreachable agent is GC'ed.

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.4.3, 1.5.2, 1.6.1, 1.7.1
• Fix Version/s: 1.5.3, 1.6.2, 1.7.2, 1.8.0
• Component/s: master
• Labels: foundations mesosphere partition registry
• Target Version/s: 1.5.3, 1.6.2, 1.7.2, 1.8.0
• Epic Link: PARTITION_AWARE-improvements
• Assignee: Vinod Kone
• Reporter: Gilbert Song
• Created: 07/May/18 07:05
• Updated: 16/Feb/19 00:19
• Resolved: 16/Feb/19 00:19

Description

Unreachable agents will be gc-ed by the master registry after `-registry_max_agent_age` duration or `-registry_max_agent_count`. When the GC happens, the agent will be removed from the master's unreachable agent list, but its corresponding tasks are still in UNREACHABLE state in the framework struct (though removed from `slaves.unreachableTasks`). We should instead remove those tasks from everywhere or transition those tasks to a terminal state, either TASK_LOST or TASK_GONE (further discussion is needed to define the semantic).

This improvement relates to how do we want to couple the update of task with the GC of agent. Right now they are somewhat decoupled.

Issue Links

• blocks: MESOS-8888 Support manual UNREACHABLE task GC.
• relates to: MESOS-9545 Marking an unreachable agent as gone should transition the tasks to terminal state