Marcus Larsson created MESOS-3703: ------------------------------------- Summary: Give frameworks more control when agents fail health checks Key: MESOS-3703 URL: https://issues.apache.org/jira/browse/MESOS-3703 Project: Mesos Issue Type: Improvement Components: master Reporter: Marcus Larsson Priority: Minor
Allow frameworks to be notified and possibly decide how to deal with health check failures on agents running the framework's task. This would allow frameworks to make better scheduling decisions when agents are failing. Longer running tasks might be given more time to finish, while other tasks could just be rescheduled immediately when an agent stops responding for a while. -- This message was sent by Atlassian JIRA (v6.3.4#6332)