[ https://issues.apache.org/jira/browse/MESOS-3507?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14907342#comment-14907342 ]
Qian Zhang commented on MESOS-3507: ----------------------------------- Just curious about how we achieve this in a *uniform* way. 1. Will we ask each framework's scheduler to implement a well-known endpoint (/task_state) over its own http? 2. How will operator collect such info from each scheduler? Will we show in it Mesos Web GUI? I think here the expectation is frameworks are cooperative. > As an operator, I want a way to inspect queued tasks in running schedulers > -------------------------------------------------------------------------- > > Key: MESOS-3507 > URL: https://issues.apache.org/jira/browse/MESOS-3507 > Project: Mesos > Issue Type: Story > Reporter: Niklas Quarfot Nielsen > > Currently, there is no uniform way of getting a notion of 'awaiting' tasks > i.e. expressing that a framework has more work to do. This information is > useful for auto-scaling and anomaly detection systems. Schedulers tend to > expose this over their own http endpoints, but the format across schedulers > are most likely not compatible. -- This message was sent by Atlassian JIRA (v6.3.4#6332)