Hello!
We are trying to run in Mesos a job which will launch its own cluster (as opposed to launching the cluster and then submitting jobs to it)
We have a couple of questions/issues:
- Is there any easier way to achieve this without having to generate a graph file before submitting the job? We wonder why this is not part
of the whole execution as it currently is with Yarn.
- If we cancel the job (or this one fails for whatever reason) the cluster doesn’t die but ends up in a weird state where we cannot submit
jobs to it.
Considering that with Yarn the cluster shuts down when the job dies, we would expect the same in Mesos. Is this happening just to us?
Thank you,
Juan G.
Locked
