Hello Till, I submitted a PR(#8084) for this issue. Could you help review it?
Many thanks! Jacky Yin 发件人: Till Rohrmann <trohrm...@apache.org> 日期: 2019年3月29日 星期五 下午11:06 收件人: Jacky Yin 殷传旺 <jacky....@eoitek.com> 抄送: "user@flink.apache.org" <user@flink.apache.org> 主题: Re: How to run a job with job cluster mode on top of mesos? Thanks a lot Jacky. Cheers, Till On Fri, Mar 29, 2019 at 1:50 PM Jacky Yin 殷传旺 <jacky....@eoitek.com<mailto:jacky....@eoitek.com>> wrote: Hello Till, Thanks very much for the clear steps. And I noticed that you have already created a jira ticket for this issue and I would like to contribute it. (I have already assigned it to me). 😊 Jacky Yin 发件人: Till Rohrmann <trohrm...@apache.org<mailto:trohrm...@apache.org>> 日期: 2019年3月26日 星期二 下午6:31 收件人: Jacky Yin 殷传旺 <jacky....@eoitek.com<mailto:jacky....@eoitek.com>> 抄送: "user@flink.apache.org<mailto:user@flink.apache.org>" <user@flink.apache.org<mailto:user@flink.apache.org>> 主题: Re: How to run a job with job cluster mode on top of mesos? Hi Jacky, you're right that we are currently lacking documentation for the `mesos-appmaster-job.sh` script. I've added a JIRA issue to cover this [1]. In order to use this script you first need to store a serialized version of the `JobGraph` you want to run somewhere where the Mesos appmaster can read it. Moreover, you need to make sure that the user code jars are on the classpath (e.g. putting them in the lib directory). See how the AbstractYarnClusterDescriptor serializes the `JobGraph` [2] for some details. The last thing before you can start the MesosJobClusterEntrypoint is to specify the path to the serialized job graph via the `internal.jobgraph-path` configuration option which you can specify in your flink-conf.yaml or pass in via dynamic property. I know that this is not super streamlined and needs to be improved. E.g. one could do it similarly to the `StandaloneJobClusterEntrypoint` that one includes the user code jar and specifies the class name of the user code to load. That way one would not need to generate the JobGraph yourself and then serialize it. I hope that I could help you a little bit. [1] https://issues.apache.org/jira/browse/FLINK-12020 [2] https://github.com/apache/flink/blob/master/flink-yarn/src/main/java/org/apache/flink/yarn/AbstractYarnClusterDescriptor.java#L846 Cheers, Till On Fri, Mar 22, 2019 at 7:24 AM Jacky Yin 殷传旺 <jacky....@eoitek.com<mailto:jacky....@eoitek.com>> wrote: Anybody can help? I found the help documentation about how to do with the job cluster mode for yarn, docker and k8s. However only the help of session cluster mode is provided in flink web site for mesos. It looks like the shell “mesos-appmaster-job.sh” should be the right one to run a job with the mode of job cluster on top of mesos. However I cannot find any help or example for this shell. Any help will be greatly appreciated. Thanks! Jacky Yin 发件人: Jacky Yin 殷传旺 <jacky....@eoitek.com<mailto:jacky....@eoitek.com>> 日期: 2019年3月21日 星期四 下午2:31 收件人: "user@flink.apache.org<mailto:user@flink.apache.org>" <user@flink.apache.org<mailto:user@flink.apache.org>> 主题: Documentation of mesos-appmaster-job.sh Hello All, I cannot find any documentation or help about how to use $flin_home/bin/mesos-appmaster-job.sh. Anybody help? Thanks! Jacky Yin
