[
https://issues.apache.org/jira/browse/MESOS-7362?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
liliang updated MESOS-7362:
---------------------------
Description:
when start the mesos agent with flag --containerizers=docker,mesos
--image_providers=docker
--isolation="filesystem/linux,docker/runtime,cgroups/devices,gpu/nvidia", and
submit spark job, the spark job can't get resource from mesos, as long as I
remove the gpu/nvidia, and restart the agent, spark task can run.
So i think i must be the agent bug
was:when start the mesos agent with flag --containerizers=docker,mesos
--image_providers=docker
--isolation="filesystem/linux,docker/runtime,cgroups/devices,gpu/nvidia", and
submit spark job, the spark job can't get resource from mesos
> GPU support can't work when run spark
> --------------------------------------
>
> Key: MESOS-7362
> URL: https://issues.apache.org/jira/browse/MESOS-7362
> Project: Mesos
> Issue Type: Bug
> Components: agent
> Affects Versions: 1.2.0
> Environment: ubuntu 16.04.
> Reporter: liliang
> Priority: Critical
>
> when start the mesos agent with flag --containerizers=docker,mesos
> --image_providers=docker
> --isolation="filesystem/linux,docker/runtime,cgroups/devices,gpu/nvidia", and
> submit spark job, the spark job can't get resource from mesos, as long as I
> remove the gpu/nvidia, and restart the agent, spark task can run.
> So i think i must be the agent bug
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
