Evan Lezar

Hi @ai-john. Have you checked the behaviour if you run the two smaller apps outside of a container? I would not expect containerization to affect this specifically expect that it...

@HYL-Dave from the log it seems that you're using `v1.3.0` of the NVIDIA Container CLI. Would it be possible to repeat the test with a more up to date version?...

The `nvidia-smi` output show persistence mode as being disabled. Does the behaviour still exist when this is enabled?

@kevin-bockman the experimental mode is still a work in progress and we don't have a concrete timeline on when this will be available for testing. I will update the issue...

Hi @carlos-garcia-returnly. Your `nvidia-container-cli -V` outptu shows version `1.11.0~rc.2` whereas your `nvidia-container-cli -k -d /dev/tty info` output shows version `1.10.0`. Were thes generated by the same CLI? Would it be...

Thanks for the additional logs. The `1.11.0-rc.2` logs clearly show: ``` W0824 16:55:19.394903 29690 nvc_info.c:425] missing binary nvidia-smi ``` whereas the earlier `info` output shows: ``` I0824 11:46:06.154184 12009 nvc_info.c:299]...

@soulseen we have released [v0.5.1](https://github.com/container-orchestrated-devices/container-device-interface/releases/tag/v0.5.1) of CDI which includes a fix for the data race. Please let us know if there are any further problems.

@saschagrunert @rhatdan what is still required to get these changes merged?

@searce-aditya I see from the list of pods that you are adeploying the `nvidia-container-toolkit-daemonset`, however your comments seem to indicate that the driver and the NVIDIA Container Toolkit are installed...

@hoangtnm the document that was linked by @shivamerla describes the options and their behaviour. If you want to prevent the use of the `NVIDIA_VISIBLE_DEVICES` environment variable in unprivileged containers, you...