Skip to content

Remote Access to Research Computing Resources§

UCL's Research Computing services are accessible from inside the UCL firewall. If you wish to connect from outside, you need to either connect through a VPN or use SSH to log in to a machine accessible from outside and use that to "jump" through into the UCL network.

Connecting to the jump boxes§

You can connect to the jump boxes by connecting with your SSH client to:

Once connected you can then log on to the UCL RC service you are using as normal.

You can configure your ssh client to automatically connect via these jump boxes so that you make the connection in one step.

Single-step logins using tunnelling§

Linux / Unix / Mac OS X§

On the command line§
# Log in to Kathleen, jumping via jump box
# Replace ccxxxxx with your own username.
ssh -o


# Copy 'my_file', from the machine you are logged in to, into your Scratch on Kathleen
# Replace ccxxxxx with your own username.
scp -o my_file

This tunnels through the jump box service in order to get you to your destination - you'll be asked for your password twice, once for each machine. You can use this to log in or to copy files.

You may also need to do this if you are trying to reach one cluster from another and there is a firewall in the way.

Useful resources can be found here:

Using a config file§

You can create a config which does this without you needing to type it every time.

Inside your ~/.ssh directory on your local machine, add the below to your config file (or create a file called config if you don't already have one).

Generically, it should be of this form where <name> can be anything you want to call this entry. You can use these as short-hand names when you run ssh.

Host <name>
   User <remote_user_id>
   HostName <remote_hostname>
   proxyCommand ssh -W <remote_hostname>:22 <remote_user_id>

This proxyCommand option causes the commands you type in your client to be forwarded on over a secure channel to the specified remote host.

On newer versions of OpenSSH, you can use ProxyJump <remote_user_id> instead of this proxyCommand line.

Here are some examples - you can have as many of these as you need in your config file.

Host myriad
   User ccxxxxx
   proxyCommand ssh -W

Host kathleen01
   User ccxxxxx
   proxyCommand ssh -W

Host aristotle
   User ccxxxxx
   proxyCommand ssh -W

You can now just type ssh myriad or scp file1 aristotle:~ and you will go through the jump box. You'll be asked for login details twice since you're logging in to two machines, the jump box and your endpoint.

File storage on the Gateway servers§

The individual servers in the pool for the Gateway service have extremely limited file storage space, intentionally, and should not be used for storing files - if you need to transfer files you should use the two-step process above. This storage should only be used for SSH configuration files.

This storage is not mirrored across the jump boxes which means if you write a file to your home directory, you will not be able to read it if you are allocated to another jump box next time you log in.

Key management§


If you use SSH keys you absolutely MUST NOT STORE UNENCRYPTED PRIVATE KEYS ON THIS OR ANY OTHER MULTI-USER COMPUTER. We will be running regular scans of the filesystem to identify and then block unencrypted key pairs across our services.

There are currently two servers in the pool, internally named ejp-gateway01 and ejp-gateway02.

Because the /home filesystem is not shared across the jump boxes, you need to sync SSH configuration files like ~/.ssh/authorized_keys across all the available jump boxes so that the change takes effect whichever jump box you are allocated to.

You can see which machine you are logged into by the bash prompt.

So for example, if on ejp-gateway02 then do:

[ ~]$ scp -r ~/.ssh ejp-gateway01:

known_hosts 100% 196 87.1KB/s 00:00
authorized_keys 100% 0 0.0KB/s 00:00
[ ~]$

and similarly if on ejp-gateway01 do scp -r ~/.ssh ejp-gateway02: