Fix ssh

Dockerfile

@@ -1,19 +1,18 @@
 FROM atlanticcrypto/cuda-ssh-server:10.2-cudnn
 
-RUN echo 'AuthorizedKeysFile .ssh/authorized_keys' >> /etc/ssh/sshd_config && \
- echo 'PasswordAuthentication no' >> /etc/ssh/sshd_config && \
- apt-get update && \
+RUN apt-get update && \
  apt-get install -y git python3.8 python3.8-dev python3-pip sudo pdsh && \
  update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.8 1 && \
  update-alternatives --install /usr/bin/python python /usr/bin/python3.8 1 && \
- update-alternatives --install /usr/bin/python python /usr/bin/python3.8 1
-
-RUN python3 -m pip install --upgrade pip && \
+ python3 -m pip install --upgrade pip && \
  pip3 install torch pipx && \
- python3 -m pipx ensurepath && \
- mkdir /app && \
- mkdir ~/.ssh
+ python3 -m pipx ensurepath
 
+RUN mkdir -p ~/.ssh /app && \
+ echo 'Host *' > ~/.ssh/config && \
+ echo ' StrictHostKeyChecking no' >> ~/.ssh/config && \
+ echo 'AuthorizedKeysFile .ssh/authorized_keys' >> /etc/ssh/sshd_config && \
+ echo 'PasswordAuthentication yes' >> /etc/ssh/sshd_config
 
 WORKDIR /app
 
@@ -23,4 +22,4 @@ RUN sh ./install_deepspeed.sh
 COPY requirements.txt /app
 RUN pip install -r requirements.txt
 
-COPY . /app
+COPY . /app

deploy_k8s.sh

@@ -19,5 +19,6 @@ do
 done
 rm authorized_keys hosts
 rm id_rsa*
-#echo 'git remote set-url origin https://github.com/EleutherAI/gpt-neox/ && git pull' | kubectl exec --stdin --tty $MASTER_ID -- /bin/bash
+echo 'chmod 600 ~/.ssh/authorized_keys && chmod 700 ~/.ssh && chown -R root /root/.ssh' | kubectl exec --stdin --tty $MASTER_ID -- /bin/bash
+
 kubectl exec --stdin --tty $MASTER_ID -- /bin/bash