Skip to content

Latest commit

 

History

History
96 lines (60 loc) · 2.51 KB

README.md

File metadata and controls

96 lines (60 loc) · 2.51 KB
Raw

Slicer Compose

SlicerCompose

Docker-compose implementations for running 3D Slicer in cloud environments.

Project structure

.
├── Slicer          <- Slicer app container
├── SlicerJupyter   <- Slicer Jupyter server container
└── docs            <- Documentation and images

Create a new VM

As an example GCE is used as the cloud provider.

Instructions of setting up a new machine are pretty basic. They go through installing basic packages, git and docker engine.

See detailed walkthrough here

GCE VM Setup

By the time you are done you should have a running machine and you are connected to the machine via ssh.

Clone the SlicerCompose repository

git clone https://github.com/piiq/SlicerCompose.git

Run slicer dockers

Run Slicer "in the browser"

The compose file in this repository uses pre-built images from SlicerDockers repository. If you need to customize the image check out the Dockerfiles and how they're built.

Navigate to the Slicer subfolder and launch docker-compose:

cd Slicer
docker-compose up -d

This will launch the container in deattached mode and make it auto-start on machine boot.

Slicer Dockers

Credentials for basic authorization

A file with slicer:researcher login and password is bundled with the repository by default.

You can create an htpasswd file with user access credentials on the remote machine to be used with the container.

For a user slicer with a password researcher the following command will create an htpasswd file:

echo "researcher" | htpasswd -c htpasswd slicer

Run Jupyter Notebook server

Launch the container

cd SlicerJupyter
docker-compose up

After the container starts you will see a log message that contains text similar to this:

slicer-notebook_1  |     Or copy and paste one of these URLs:
slicer-notebook_1  |         http://57ff061693bc:8888/?token=0821a59da174e587a4f3ec7ecef303805f2149ad8b335e33
slicer-notebook_1  |      or http://127.0.0.1:8888/?token=0821a59da174e587a4f3ec7ecef303805f2149ad8b335e33

Copy the token and navigate to

http://your-machine-ip-address:8888

Enter the token and set the password.

After setting a password for the notebook server stop the container by pressing Ctrl+C and start the container in deatached mode:

docker-compose up -d

Slicer jupyter