Login Node¶

Users accessing CCR's HPC resources will be connected to a login node. A login node is a outward facing node within CCR's HPC environment that users can connect to from their local machines. Once on a login node, users can perform a limited number of tasks:

• Edit files
• Transfer Data
• Submitting Jobs
• Access storage resources

This page covers connecting to a login node using SSH (Secure Shell Protocol). When you connect via SSH, you authenticate using a private key file on your local machine. For more information about SSH, see Secure Shell on Wikipedia. You can also connect to CCR's login nodes using a web browser with the OnDemand Portal.

To get started connecting to CCR's HPC clusters and shared storage you need the following:

• A CCR user account with two factor authentication enabled and an active allocation to HPC resources. For more information see our getting access guide.
• An SSH Key added to your account
• An SSH client. Linux & MacOS include a terminal app. For Windows users, we recommend using Git Bash

CCR login nodes hostnames:

• vortex.ccr.buffalo.edu - production login nodes
• vortex-future.ccr.buffalo.edu - includes new features and serves as a preview of the next deployment of production login nodes

Watch this virtual workshop to learn more about monitoring your jobs:

Connecting with SSH¶

Using the SSH protocol, you can connect and authenticate to CCR login nodes without supplying your password. To set up SSH, you will need to generate a new SSH key pair that contains a private key (stored only on your personal computer) and a public key (uploaded to your CCR account using the IDM portal). During the login process the CCR servers attempt to match the public key that you've uploaded to your CCR account with the private key you have stored on your computer. If they match, the login is successful.

Generate new SSH key¶

Prior to using SSH to connect to CCR's login servers, you'll need to generate a new SSH key pair on your local machine. If you connect to CCR from multiple devices, we recommend generating a separate SSH key pair for each device. After you generate the key pair(s), you must add the public key(s) to your account using CCR's IDM portal.

Follow these easy steps:

1. Open your terminal or Git Bash if you're on windows

2. Run the following command, substituting in your email address:

   $ ssh-keygen -t ed25519 -C "myemailaddress@institution.edu"
   

3. At the prompt, type a secure passphrase

4. Copy the contents of the public key to your clipboard. The file is located here:

   ~/.ssh/id_ed25519.pub
   

5. Login to the CCR IDM portal and click on SSH Keys in the left nav menu

6. Click on the "New SSH Key" button, paste the contents of your public key in the text box, and click "Add".

Using the SSH Agent¶

Running an SSH agent process on your local machine allows you to load your SSH private key one time and it will be used for every SSH login attempt. This allows you to skip entering your SSH key passphrase each time you login and you will not have to specify your private key in the SSH login command. Follow these steps to add your key to the ssh-agent:

1. Open your terminal or Git Bash if you're on windows

2. Start the ssh-agent in the background:

   $ eval "$(ssh-agent -s)"
   > Agent pid 12345
   

3. Add your SSH private key to the ssh-agent using the command below. If you created your key with a different name or location, use the correct path and filename to your private key. You will be prompted to enter your passphrase for the key. NOTE: The SSH key passphrase is NOT your CCR password. This is the passphrase you set when creating your SSH key pair:

   $ ssh-add ~/.ssh/id_ed25519
   

4. List the SSH keys currently loaded in your running SSH agent:
   

   $ ssh-add -L
   

Logging in¶

Once you've uploaded your SSH public key to your CCR account you will be able to use the SSH protocol to connect to CCR's login nodes. Follow these steps:

1. Open your terminal or Git Bash if you're on Windows and, if you're running a SSH agent as described above, enter the following:
   

   ssh [CCRusername]@vortex.ccr.buffalo.edu
   

   Not using ssh-agent?

   If you're not running a SSH agent (see above) you will need to specify the location and filename of your PRIVATE key using the -i option in the SSH command and you'll be prompted for your SSH key passphrase. This is NOT your CCR password. This is the passphrase you set when creating your SSH key pair. For example:
   

   ssh -i /path-to-key/id_ed25519 [CCRusername]@vortex.ccr.buffalo.edu
   Enter passphrase for key 'id_ed25519':
   

2. When logging into CCR's login nodes, you may see a warning message similar to this. Verify the fingerprint in the message you see matches CCR's public key fingerprint. If it does, then type yes.

   The authenticity of host 'vortex.ccr.buffalo.edu (128.205.41.24)' can't be established.
   ED25519 key fingerprint is SHA256:qYT1DzrHv8yTlHiNGV2td29309oXPHdN4OPj/KptFYg.
   Are you sure you want to continue connecting (yes/no/[fingerprint])?
   

3. If your login was successful, you should now be at a shell prompt on the login node:

   CCRusername@login:~$
   

Compute Node Logins¶

You will only be able to login to compute nodes that your jobs are running on. However, you can not use rsh/ssh to access the compute nodes. You can use the Slurm srun command to get on the node. If the job is running on one node use:
srun --jobid=jobid --pty /bin/bash

If the job is running on more than one node, specify the node you want to login to:
srun --jobid=jobid --nodelist=node_name -N1 --pty /bin/bash

• If your job is allocated all of the resources on the node, you will need to include the --overlap option.
• If your job is running on the faculty cluster, you will need to specify the --clusters=faculty option.

OnDemand for Web-Based Cluster Access¶

Open OnDemand provides access to CCR's clusters, storage, visualization servers, and interactive apps. CCR's OnDemand portal offers Linux desktops for GUI-based applications, software applications like MatLab, RStudio Desktop, Jupyter Notebook, and vscode, an interactive development environment. OnDemand can be used in any browser from almost any device. More information about OnDemand can be found here.