
This docker container has all the components you need to run the landlab. It is made by Simon Mudd and not the landlab team so any bugs in the container are not their fault!
It is built within a miniconda parent container and also includes ipython so that you can host and jupyter notebook to run examples.
These are the bare bones instructions. For a bit more detail and potential bug fixes, scroll down to the section on Docker notes.
brew cask install dockersudo usermod -a -G docker $USERC:\landlab.Preamble: Once you have downloaded docker, you can control how much memory you give the docker containers. The default is 3Gb. If you have even moderate sized DEM data, this will not be enough. You can go into the docker settings (varies by operating system, use a search engine to figure out where they are) and increase the memory.
$ docker pull muddpile/landlab_docker
$ docker run -it -v C:\landlab:/landlab muddpile/landlab_docker
-it means "interactive".-v stands for "volume" and in practice it links the files in the docker container with files in your host operating system.-v you need to tell docker where the directories are on both the host operating system (in this case C:\landlab) and the container (in this case /landlab). These are separated by a colon (:).# symbol showing that you are inside the container. If you don't have the example notebooks, grab them with# grab_landlab_notebooks.sh
The landlab container can also serve as a host for jupyter notebooks
You need to open your docker container with a port (using the -p flag, 8888:8888 is a common port):
> docker run -it -v C:\landlab:/landlab -p 8888:8888 muddpile/landlab_docker
C:\landlab to reflect the directory structure on your local machine.#. If you don't have the example notebooks, grab them with# grab_landlab_notebooks.sh
# jupyter notebook --ip 0.0.0.0 --port 8888 --no-browser --allow-root
token: you can see it in the address after token=.token that was shown in the docker container.If you want to know all about Docker, make sure to read the docker documentation. A note of warning: Docker documentation is similar to documentation for the turbo encabulator. Below are some brief notes to help you with the essentials.
Here are some shortcuts if you just need a reminder of how docker works.
List all containers
$ docker ps -a
List containers with size
$ docker ps -as
Remove all unused containers
$ docker system prune
After you install docker on Linux, you will need to add users to the docker permissions:
$ sudo usermod -a -G docker $USER
Once you have done this you will need to log out and log back in again.
I have not made any scientific study of this but many scientific users are on Windows operating systems.
Firstly, you need to have Windows 10 Enterprise. It will not work otherwise (well, that is not exactly true but getting it to work on Windows 7 is a massive pain). If you don't have Windows 10 Enterprise but are on Windows you probably should use Vagrant; see our vagrant documentation. If you do have Windows 10 enterprise then you can download and install Docker for Windows CE. After you install this you will need to restart your computer not once but twice: once after install and a second time to activate the hyper-V feature, which allows you to have 64 bit guest operating systems.
Second, if you have that and have it installed, you might also need to add yourself to the docker-users group. To do that, do this (instructions from here: https://github.com/docker/for-win/issues/868):
Content type
Image
Digest
Size
1.2 GB
Last updated
about 6 years ago
docker pull muddpile/landlab_docker