-
Notifications
You must be signed in to change notification settings - Fork 4
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
Docker update #172
Docker update #172
Conversation
@bentsherman would you mind looking this over? If you don't have time for a functional review if you can just check the changes and see if you are okay with them? The only outstanding item on this is clarification about the |
Also, once this PR is merged and after we do the 3.4.2 release we can rebuild the images for DockerHub. For the 3.4.2 release we just have this PR and the other #171 . Then we can release! |
Oh, one other note.... If you try to build the images you'll see some warnings when building KINC about unused variables. That has been fixed in #171 so no need to try to clean that up. It will go away once all the PRs are merged. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I left my review via inline comments. Just a few things that need to be changed. The docker image will likely be a lot bigger because of the additional R dependencies, but hopefully that will all be removed once the R scripts are migrated to Python. A few extra notes:
- Make sure that you can actually use the KINC GUI from the docker image. I know you can run web servers from a docker image but I have never tried native GUI applications. If KINC GUI doesn't work then you should leave the
GUI=no
option in the build and specify that in the documentation. - The CPU builds are for CPU-only machines, and it installs the NVIDIA headless drivers instead of the actual drivers so that KINC loads correctly. You may recall users in the past reporting issues with running KINC without a GPU, and this is the workaround that I developed for that.
Hi @bentsherman. I believe I have resolved all of the issues you raised on this PR. I did update the Dockerfile a bit more to make it easier to build just a cpu or gpu version. Also I added a new Can you review again? I'm hoping to do a release on Monday for Iris. |
Thanks @bentsherman |
This PR updates the Docker file to support the RScripts and the 3D python viewer so that users who want to use the Docker image to avoid installing all of the dependencies can do so.
This PR makes the following changes:
lastest
images now have ACE hardcoded to 3.2.0 rather thandevelop
. This is because the develop branch can have, at any point in time, experimental code. I think thelatest
image should be the latest of development version of KINC and not the latest of of ACE.-cpu
and-gpu
tagged images. That should be included as the current image name isn't correct in the commands.nvidia-docker
to justdocker
with the--gpus all
option. I wasn't able to run the image withnvidia-docker
as that didn't exist on my machine.Unfortunately, all of these changes makes the image rather large and take a long time to build.