ssss

training/cambricon/docker_image/flagscale_2409/Dockerfile

@@ -1,4 +1,4 @@
-FROM cambricon_flagscale_2409_pt21
+FROM flagperf:cambricon-v24.08.02-torch2.1.0-catch1.22.1-ubuntu22.04-py310-megatron-patch-flagscale
 #shell
 SHELL ["/bin/bash", "-c"]
 RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -yq --no-install-recommends tzdata && apt-get install -y openssh-server && mkdir -p /run/sshd

training/cambricon/docker_image/flagscale_2409/flagscale_2409_install.sh

@@ -5,3 +5,13 @@ pip3 config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
 pip3 install regex==2024.5.15 schedule==1.2.2 accelerate==0.31.0 transformers==4.40.1 protobuf==3.20.0
 pip3 install pybind11 hydra-core s3fs braceexpand webdataset wandb loguru sentencepiece datasets
 pip3 install megatron-energon==2.2.0
+
+# 配置免密
+echo 'root:123456' | chpasswd
+sed -i '/StrictHostKeyChecking/c StrictHostKeyChecking no' /etc/ssh/ssh_config
+sed -i 's/#Port 22/Port 9876/g' /etc/ssh/sshd_config
+sed -i 's/#PubkeyAuthentication yes/PubkeyAuthentication yes/g' /etc/ssh/sshd_config
+sed -i 's/#PermitRootLogin prohibit-password/PermitRootLogin yes/g' /etc/ssh/sshd_config
+/etc/init.d/ssh restart
+sleep 10
+sshpass -p "123456" ssh-copy-id -i /root/.ssh/id_rsa.pub -p 9876 root@`hostname -i | awk '{print $1}'`