[jira] [Updated] (MESOS-5555) Always provide access to NVIDIA control devices within containers (if GPU isolation is enabled).

["Benjamin Mahler (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MESOS-5555?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Benjamin Mahler updated MESOS-5555:
-----------------------------------
    Summary: Always provide access to NVIDIA control devices within containers (if GPU isolation is enabled).  (was: Change semantics for granting access to /dev/nvidiactl, etc)

> Always provide access to NVIDIA control devices within containers (if GPU isolation is enabled).
> ------------------------------------------------------------------------------------------------
>
>                 Key: MESOS-5555
>                 URL: https://issues.apache.org/jira/browse/MESOS-5555
>             Project: Mesos
>          Issue Type: Improvement
>            Reporter: Kevin Klues
>            Assignee: Kevin Klues
>              Labels: gpu, mesosphere
>             Fix For: 1.0.0
>
>
> Currently, access to `/dev/nvidiactl` and `/dev/nvidia-uvm` is only granted to / revoked from a container as GPUs are added and removed from them. On some level, this makes sense because most jobs don't need access to these devices unless they are also using a GPU. However, there are cases when access to these files is appropriate, even when not making use of a GPU. Running `nvidia-smi` to control the global state of the underlying nvidia driver, for example.
>     
> We should add `/dev/nvidiactl` and `/dev/nvidia-uvm` to the default whitelist of devices to include in every container when the `gpu/nvidia` isolator is enabled. This will allow a container to run standard nvidia driver tools (such as `nvidia-smi`) without failing with abnormal errors when no GPUs have been granted to it. As such, these tools will now report that no GPUs are installed instead of failing abnormally.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)