Skip to content

Access

In order to access the server, you need to contact the datalab with your basic information and SODAS affiliation, and they will help setup up an account for you. The process currently takes a bit of time, so please be patient while we look into ways of optimizing this.

Connecting to the server

First of all, you need to be connected to UCPH's network through the VPN. See the guides on the intranet. Our friends at DIKU also have a guide for using a VPN with UCPH's network for linux and mac users.

Warning

It is (currently) not enough to be on the cabled network at UCPH, you need to be connected to the VPN.

You connect to the server using SSH. See tips on SSH here. An SSH connection to the server can be established either through your terminal of choice or e.g. VSCode.

Note

Windows users can use the build in command prompt to run the following commands, or download Windows Terminal for a better experience.

Terminal

SSH

ssh <username>@sodashead01fl.unicph.domain

To connect locally to the server with the user abc123, do:

ssh abc123@sodashead01fl.unicph.domain

Note

You can specify a specific address and port to connect to, eg:

ssh -L 8000:localhost:8000 abc123@sodashead01fl.unicph.domain

Transferring files

Transfer from local to server
scp /path/to/file <username>@sodashead01fl.unicph.domain:/path/to/destination

To move a text file in documents to your H-drive which is mounted on the server by default, do:

scp /documents/test.txt abc123@sodashead01fl.unicph.doman:/ucph/hdir
Transfer file from server to local
scp <username>@sodashead01fl.unicph.domain:/path/to/file /path/to/destination

To move a text file from the H-drive on the server to your local documents folder, do:

scp abc123@sodashead01fl.unicph.doman:/ucph/hdir/test.txt /documents

VSCode

You can access the server through VSCode by installing the Remote - SSH extension. Follow the instructions in the documentation to connect to the server.

Then, when connected, install the python and juyter extensions on the remote.

To make the compute resources from slurm available in vscode, see under usage.