[Feature Request] Latency metrics of task execution in cluster manager

[dhwanilpatel]

Is your feature request related to a problem? Please describe

Today we have visibility on the total number of pending tasks and time in queue. We should have visibility on latency of the tasks during execution phase as well.

This latency metric will help determine if any task is taking long time in execution. We can even add more granular metrics of execution as well like time taken during cluster state computation/publication/committing/etc.

Describe the solution you'd like

One possible solution could be to expose in _cluster/pending_task API, where along with time_in_queue we can expose time_in_execution.

We can explore other solutions as well if there is any other better way to expose it.

Related component

Cluster Manager

Describe alternatives you've considered

No response

Additional context

No response

[peternied]

[Triage - attendees 1 2]
@dhwanilpatel Thanks for filing, look forward to a PR that adds this feature