add additional packages via apt for better multi-node support (#2949)

* cleanup in Dockerfile and add infiniband packages

* fixes for ci

* fix nightly too
This commit is contained in:
Wing Lian
2025-07-20 21:19:23 -04:00
committed by GitHub
parent b986f7c7cb
commit 31a15a49b6
6 changed files with 26 additions and 20 deletions

View File

@@ -11,7 +11,7 @@ ENV NIGHTLY_BUILD="{{ NIGHTLY_BUILD }}"
ENV HF_HOME="{{ HF_HOME }}"
RUN apt-get update && \
apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev
apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev ibverbs-providers ibverbs-utils infiniband-diags librdmacm-dev librdmacm1 rdmacm-utils slurm-wlm
WORKDIR /workspace

View File

@@ -12,7 +12,7 @@ ENV HF_HOME="{{ HF_HOME }}"
ENV AXOLOTL_DATASET_PROCESSES="8"
RUN apt-get update && \
apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev
apt-get install -y --allow-change-held-packages vim curl nano libnccl2 libnccl-dev ibverbs-providers ibverbs-utils infiniband-diags librdmacm-dev librdmacm1 rdmacm-utils slurm-wlm
WORKDIR /workspace