SSH

SSH Login

To log in to a cluster, use the following command, replacing <username> with your username and <mycluster> with the fully qualified domain name (FQDN) of your cluster:

 ssh <username>@<mycluster>

Passwordless Login

If you want to bypass password authentication, you can use SSH key authentication. SSH keys allow you to connect to a server or multiple servers without entering your password each time.

To generate an SSH key pair, follow these steps:

1. On your local machine, generate an SSH key:

 ssh-keygen -t ed25519 -C "[$(whoami)@$(uname -n)-$(date -I)]"

2. Automatically add your public key to the server by running the following command (use your ANDREW password):

 ssh-copy-id -i ${HOME}/.ssh/id_ed25519.pub <username>@<mycluster>.

3. You can also configure your ~/.ssh/config file to avoid typing the full cluster address every time. Add the following lines to the file:

 Host <mycluster>
 HostName <mycluster's hostname>
 User <username>
 IdentityFile ${HOME}/.ssh/id_ed25519

4. Then, you can log in with the simple command ssh mycluster.

Connecting to a Compute Node

You can also ssh directly to a compute node where you have a job already running by doing:

ssh -J babel babel-X-XX

More on Port Forwarding

Copying Data to Compute Nodes

To copy data to compute nodes when the data directories are not directly accessible from the login nodes, you can utilize the login node as a JumpHost, allowing direct connections to the compute nodes. Follow these steps to configure SSH and establish the connection:

1. Launch an interactive job on a compute node.
2. Determine the IP address of the compute node by executing the ip a command on the compute node itself.
3. Update your SSH configuration file located at ~/.ssh/config with the following entries:

 Host <mycluster-short-name>
   HostName <mycluster-fqdn>
   User <username>
   IdentityFile <identityfile>

 Host <mycluster>-compute-node
   HostName <node ip addr>
   User <username>
   ProxyJump <mycluster>

Replace the relevant items in < > with your actual username, cluster name and the path to your SSH private key file. For example, to connect to a node in the Babel cluster:

Host babel
  HostName babel.lti.cs.cmu.edu
  User <username>
  IdentityFile /path/to/identityfile

Host babel-compute-node
  HostName <node ip addr>
  User <username>
  ProxyJump babel

The updated configuration allows you to establish a connection to the compute node through the login node, acting as a JumpHost.

Refer to your cluster's dedicated page in the Cluster's section for more detailed information.

Why is there an SSH key pair in the home directory upon first login?

When a user runs a job on two nodes, their script might need to SSH from node-1 to node-2. If the SSH connection fails or hangs, it is often because the fingerprint of the target node is not in the ~/.ssh/known_hosts file. To prevent this issue, we automatically create SSH keys and add the necessary fingerprints to ~/.ssh/authorized_keys when the user's home directory is created.

This process also eliminates password prompts when connecting to a node where you have a job running.

For more information on generating these keys, see: Connecting to the Cluster - Passwordless Login

Troubleshooting

Windows

SSH Connection Errors

When attempting to connect to the HPC cluster using Visual Studio Code and SSH, Windows users may encounter the following error:

Corrupted MAC on input.
ssh_dispatch_run_fatal: Connection to 128.2.205.73 port 22: message authentication code incorrect

Solution

To resolve this issue, add the following configuration to your local SSH config file (~/.ssh/config):

Host babel babel-login

   HostName babel.lti.cs.cmu.edu
   Ciphers aes128-ctr,aes192-ctr,aes256-ctr
   MACs hmac-sha2-256,hmac-sha2-512,hmac-sha1
   User <USERNAME>
   ForwardAgent yes

Replace <USERNAME> with your actual username.

If you are using an SSH key, add the following line, replacing the placeholder with your private SSH key's file path:

   IdentityFile C:/Users/<USERNAME>/.ssh/<private_ssh_key_name>