NVIDIA / NVIDIA/MagnumIO

Build container fails due to libnccl versions

Open
#12 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
122
Forks
27
PR merge metrics
No merged PRs in 30d

Description

When running ./installer.sh build-container the container build fails due to libnccl versions:

Building dependency tree...
Reading state information...
The following held packages will be changed:
libnccl-dev libnccl2
The following packages will be DOWNGRADED:
libnccl-dev libnccl2
E: Packages were downgraded and -y was used without --allow-downgrades.
0 upgraded, 0 newly installed, 2 downgraded, 0 to remove and 15 not upgraded.
Error: error building at STEP "RUN wget -qO - https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/7fa2af80.pub | apt-key add - && echo "deb https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 /" >> /etc/apt/sources.list.d/hpccm.list && apt-get update -y && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends libnccl-dev=2.10.3-1+cuda11.4 libnccl2=2.10.3-1+cuda11.4 && rm -rf /var/lib/apt/lists/*": error while running runtime: exit status 100
##NV## Fri Apr 8 17:46:05 EDT 2022 ## END Building Container

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start in installer.sh at the build-container path and inspect the container step that adds the NVIDIA repository and installs the pinned libnccl-dev and libnccl2 versions. Run ./installer.sh build-container to reproduce the apt downgrade error; done means the container builds successfully without the libnccl version failure.

Written by the indexing model from the issue text.

Assessment

Tech stack
docker, shell
Domain
build-system, devops
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
42/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.