[
https://issues.apache.org/jira/browse/FLINK-22935?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17366647#comment-17366647
]
Piotr Nowojski commented on FLINK-22935:
----------------------------------------
I would suggest to ask such questions on the user mailing list first, as this
is most likely not a bug, but some configuration/environment problem and you
would get a help there much faster
https://flink.apache.org/community.html#how-do-i-get-help-from-apache-flink.
For example there was recently at least one similar question:
https://lists.apache.org/thread.html/rec73c61033b374504fa47d43ec946f64c3bc685a4ee92a177db80667%40%3Cuser.flink.apache.org%3E
In short I would expect this to be a problem with configuration
(`flink-conf.yaml`) being not distributed among all Task Managers properly.
One machine has picked up the correct file, maybe two other local ones managed
to connect with some incorrect configuration, and the last one (the remote one)
was not able to connect at all. Alternatively from the attached screenshot it
looks like you have some lingering processes (warning about instance already
running), so I would try to make sure everything is correctly shutdown (stop
them cleanly or manually kill processes or reboot machines) and re-try again.
If this doesn't help, could you ask this question on the user mailing list,
attaching your configuration and logs from all of the task managers and job
manager?
> Can not start standalone cluster
> ---------------------------------
>
> Key: FLINK-22935
> URL: https://issues.apache.org/jira/browse/FLINK-22935
> Project: Flink
> Issue Type: Bug
> Components: Runtime / Coordination
> Affects Versions: 1.11.2
> Environment: centos
> Reporter: Lyubing Qiang
> Priority: Major
> Attachments: image-2021-06-09-10-05-21-481.png,
> image-2021-06-09-10-06-08-418.png, image-2021-06-09-10-06-48-825.png
>
>
> I tried to start the standalone mode (following the instructions to
> configure) with 4 slaves and master. When I run the start-cluster.sh it
> seemed all the 4 slaves were opened and running, yet actually there're only 3
> taskmanagers and all of them are on the master machine. Also the number of
> task slots are deviated from the settings in the flink-conf.yaml.
> !image-2021-06-09-10-05-21-481.png!
> !image-2021-06-09-10-06-08-418.png!
> !image-2021-06-09-10-06-48-825.png!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
