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:
@@ -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
|
||||
|
||||
|
||||
@@ -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
|
||||
|
||||
|
||||
Reference in New Issue
Block a user