GitHub user sachingoel0101 opened a pull request:
https://github.com/apache/flink/pull/979
[FLINK-2472]Make JobClientActor poll JobManager continuously for updates on
submitted job
This PR adds functionality for the `JobClientActor` to poll the
`JobManager` after a set interval `JOB_CLIENT_HEARTBEAT_INTERVAL`(5s) about the
status of submitted job.
Furthermore, there is a inherent timeout on the `JobClientActor` for not
receiving any messages(only `JobManager` is likely to send any messages) for
`JOB_CLIENT_JOB_MANAGER_TIMEOUT`(10s).
If the job status comes as `CREATED` or `RESTARTING` for more than
`JOB_CLIENT_JOB_STATUS_TIMEOUT`(30s), the `JobClient` is sent a
`Status.Failure` message, which to equivalent to the idea mentioned in the jira
ticket.
This work isn't complete, and I'd love feedback about whether different job
status messages are handled properly, and what is the best place to place the
time interval configuration. It can be placed in the usual place, i.e.,
`ConfigConstants` but there is no `Configuration` accessible in the
`JobClientActor`.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/sachingoel0101/flink flink-2472
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/flink/pull/979.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #979
----
commit 9cf23c5e397bc7a8a8faf7427d2ae6dbd0d73143
Author: Sachin Goel <[email protected]>
Date: 2015-08-03T23:14:29Z
[FLINK-2472]Make JobClientActor poll JobManager continuously for updates
and fail if job manager is dead or task hasn't been running
----
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
