Github user ohal commented on the pull request:
https://github.com/apache/spark/pull/4960#issuecomment-142045298
@tnachen
I'm trying to get roles implementation working without any luck, can you
please point me what should be configured?
Have a cluster, Mesos 0.22.1, hadoop 2.5.0, 3 masters, 8 slaves, which are
3 - role **fast**, 3 - role **"*"**, 2 other roles.
Spark 1.5.0 is deployed to use cluster mode with dispatcher, it looks
working, however I can not get it working on the particular nodes with role
configured.
Job submitted as `# bin/spark-submit --class
org.apache.spark.examples.SparkPi --master master.novalocal:7077 --deploy-mode
cluster --conf spark.mesos.role=fast http://www/spark.jar 10`
Also, I've tried to configure Spark defaults for submitted jobs and added
to `/opt/spark/conf/spark-defaults.conf` the line `spark.mesos.role fast`
In any case after submitting job is staging/running on the **"*"** nodes
not on the **fast** nodes.
Can you please describe how to configure it properly?
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
