User Management

To add users to the server, please follow the following steps:

1. Add username (UID) to the JupyterHub allowed list under admin tab ToDo: picture

2. Add username along with the requested resource to the user-list.csv

   # list of users
   vim /srv/jupyterhub/users/user-list.csv

   Username	Node	RAM	CPU	GPU	GPU_shared	Start_time	End_time	Staff	Image
   muenst2s	node0+node1	64	24	3	1	14.03.2023-12:00	14.04.2024-19:00	1	ghcr.io/b-it-bots/docker/gpu-notebook:11.3.1-cudnn8-runtime-ubuntu20.04+ghcr.io/digiklausur/docker-stacks/notebook-dev:latest
   mmuens2s	node0	8	4	2	0	14.02.2023-12:00	14.02.2024-19:00	0
   mmuensd2s	dynamic_node	64	24	3	1	14.03.2023-12:00	14.04.2024-19:00	1	ghcr.io/b-it-bots/docker/gpu-notebook:11.3.1-cudnn8-runtime-ubuntu20.04+ghcr.io/digiklausur/docker-stacks/notebook-dev:latest

• Username: UID of the user
• Node: Nodes availabe to users (options: node0, node1, node2, node3, node4, and dynamic_node). Sign + indicates that the user has access to multiple nodes that are labelled a2s.cluster.gpu/node-label=<label>, and dynamic_node indicates that the user can be allocated dynamically (depending on the resource availability) to nodes which are labelled a2s.cluster.gpu/node-allocation=dynamic.
• RAM: Allocated RAM in GB (options: 8, 16, 32, 64)
• CPU: Number of allocated CPU cores (options: 8, 16, 32)
• GPU: Number of allocated GPUs (dependent on the number of GPUs available on the selected node)
• GPU_shared: Set to 1 to enable shared GPU
• Start_time: Earliest access date for the user
• End_time: Latest access date for the user
• Staff: Set to 1 if the user is a staff member with access to server configuration
• Image: Image selection for the user. Sign + indicates that the user has access to multiple images.