[
https://issues.apache.org/jira/browse/FLINK-7087?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16104725#comment-16104725
]
Till Rohrmann commented on FLINK-7087:
--------------------------------------
These are all good questions you're raising here. I think most of them are open
and you're already naming some alternatives. We could make this issue the
umbrella issue for the above mentioned issues.
If we opt for the {{JobGraph}} inclusion in the image, then a job upgrade would
entail a redeployment of a Mesos cluster and then resuming from a savepoint. If
the {{JobGraph}} can be obtained from an external service, then we could reuse
the existing Mesos cluster.
Concerning the different types of pods, the specialization of pods can only
work if some kind of leader election works (e.g. using ZooKeeper). At the end
of the day, I think the most important part is to provide proper tooling for
the user to scale up and down a Mesos cluster.
> Implement Flip-6 container entry point
> --------------------------------------
>
> Key: FLINK-7087
> URL: https://issues.apache.org/jira/browse/FLINK-7087
> Project: Flink
> Issue Type: Bug
> Components: Cluster Management
> Reporter: Till Rohrmann
> Labels: flip-6
>
> In order to support Docker and K8 we have to provide a container entry point.
> In a first version, the container entry point could be similar to the
> standalone session mode just with the difference that we don't submit a job
> to the cluster. The job has to be contained in the container image or being
> at least retrievable by the entry point. In that sense the container entry
> point acts like a per-job standalone mode.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
