[MESOS-4048] Consider unifying slave timeout behavior between steady state and master failover

• Type: Improvement
• Status: Accepted
• Priority: Minor
• Resolution: Unresolved
• Component/s: agent, master
• Labels: mesosphere
• Assignee: Megha Sharma
• Reporter: Neil Conway
• Created: 02/Dec/15 19:54
• Updated: 26/Nov/18 13:36

Description

Currently, there are two timeouts that control what happens when an agent is partitioned from the master:

1. max_slave_ping_timeouts + slave_ping_timeout controls how long the master waits before declaring a slave to be dead in the "steady state"
2. slave_reregister_timeout controls how long the master waits for a slave to reregister after master failover.

It is unclear whether these two cases really merit being treated differently – it might be simpler for operators to configure a single timeout that controls how long the master waits before declaring that a slave is dead.

Issue Links

• relates to: MESOS-4049 Allow user to control behavior of partitioned agents/tasks
• relates to: MESOS-5396 After failover, master does not remove agents with same UPID.