Cluster environment

Manage the cluster environment with Ansible.

Prerequisites

Master machine

Install Ansible with pipx

# Install pipx
sudo apt install pipx
pipx ensurepath

# Reload ${PATH}
source .bashrc

# Install Ansible
pipx install --index-url "https://mirrors.ustc.edu.cn/pypi/simple" --include-deps ansible

Generate SSH key and copy it to the all machines

# Generate SSH key
ssh-keygen -t ed25519

# Add SSH key to ssh-agent
eval "$(ssh-agent -s)"
ssh-add ~/.ssh/id_ed25519

# Copy SSH key to the all machines, don't forget to copy the key to the master machine
ssh-copy-id <username>@<target-machine>

Target machines

Required packages:

• python-apt

Usage

!!! WARNING !!!

The following command will alter the target machines. Add --check to perform a dry-run to confirm what will be changed.

ansible-playbook -i hosts.ini playbook.yml -K --tags "clone-repo, <tag>, ..."

Tags

Required	Action	Tag
Yes	Clone this repository	clone-repo
No	Slurm	slurm
No	Copy slurm configuration	copy-slurm-config
No	Intel oneAPI HPC Toolkit	intel-oneapi-hpc-toolkit
No	Install docker	install-docker