Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add support for "--device" option #2938

Closed
jaychris opened this issue Jul 31, 2017 · 3 comments
Closed

Add support for "--device" option #2938

jaychris opened this issue Jul 31, 2017 · 3 comments

Comments

@jaychris
Copy link

jaychris commented Jul 31, 2017

We need an option that implements the "--device" Docker cli option, to mount usb devices and other serial control devices for hardware that container based applications are managing.

Per comment from @schmichael on #2671:

Allowing mounting arbitrary devices makes sense. I think the distinction
between GPUs (#406) and other devices is we intend for the scheduler to
have first class knowledge of GPU resources. Other devices will just be
supported in driver configs.

Jay or Pieter: Would one of you mind opening a Github issue for your
"--device" use cases? It shouldn't be blocked by #406.

@pieterbreed
Copy link

On my side, we need to support the --device /dev/fuse docker cli option to support mounting an AWS s3 bucket using riofs.

@kmalec
Copy link
Contributor

kmalec commented Oct 20, 2017

I might post a small PR for this at some point, since we're currently PoCing stuff that runs GPU-enabled containers on CoreOS instances, and I need to use --device /dev/nvidiactl --device /dev/nvidia0 --device /dev/nvidia-uvm.
I've seen the GPU-specific issue, not sure it's needed - in our case, I took inspiration from The Internet(TM) and I've built the drivers and kernel module for CoreOS inside Ubuntu docker container, so to load them, I run privileged docker container first, then create the devices inside CoreOS with a script, and then expose those devices to the normal containers. Fingerprinting in that case would require making sure Nomad agent starts after the devices have been initialized.

In the long run, it's probably pretty cool, but for now, in my case we can make do by just tagging the instances' metadata and scheduling by a node class.

@github-actions
Copy link

github-actions bot commented Dec 6, 2022

I'm going to lock this issue because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active issues.
If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.

@github-actions github-actions bot locked as resolved and limited conversation to collaborators Dec 6, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

4 participants