fix: Improve efficiency and apply overall fixes to the role processes#22
Merged
spetrosi merged 5 commits intolinux-system-roles:mainfrom Oct 2, 2025
Merged
Conversation
Reviewer's guide (collapsed on small PRs)Reviewer's GuideRefreshed the README to provide clearer guidance on installing and using the system OpenMPI package and ensure consistent phrasing for loading environment modules. File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
* Install kernel-headers and -devel for all installed kernels * Restart dkms after installing nvidia drivers
openmpi needs to be configured with the rebuilt, PMIx aware UCX library, not the default on e shipped in HPCX. Make sure the lmod for the library has the correct library paths set up, too.
Testing of the openmpi-5.0.8 library resulted in this warning being emitted: [tmp-chhhgt6r-gpu-1:25690] SET UCX_TLS=tcp [LOG_CAT_SBGP] libnuma.so: cannot open shared object file: No such file or directory [LOG_CAT_SBGP] Failed to dlopen libnuma.so. Fallback to GROUP_BY_SOCKET manual. [LOG_CAT_SBGP] libnuma.so: cannot open shared object file: No such file or directory [LOG_CAT_SBGP] Failed to dlopen libnuma.so. Fallback to GROUP_BY_SOCKET manual. The lib64/libnuma.so link is provided by the numactl-devel package, not numactl-libs. Add the -devel package to the depedency list for early install.
b29de92 to
d0b0f4a
Compare
Rename the openmpi-5.0.8 module to indicate that it only supports CUDA-based applications on hardware with NVidia GPUs. Add new wrapper scripts for the NVidia HPCX openmpi libraries. There are two versions of this - hpcx and hpcx-pmix - that point to the original library and the one rebuilt with pmix. This provides conflicts with the other MPI libraries, as well as allows us to remove the /etc/lmod/.modulespath file that points to the base HPCX module files.
d0b0f4a to
296b704
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Improvements after QE tests