Installation

The Mission Support System (MSS) including a Web Map Service, a Collaboration Server and a Graphical User Interface is available as conda-forge package.

This channel conda-forge has builds for osx-64, linux-64, win-64

The conda-forge github organization uses various automated continuos integration build processes.

We provide an automatic installation and a manual installation.

We recommend to use Mamba for an installation.

Automatic

  • For Windows, use Windows.bat

    1. Right click on the webpage and select “Save as…” to download the file

    2. Double click the downloaded file and follow further instructions

    • For fully automatic installation, open cmd and execute it with /Path/To/Windows.bat -a

  • For Linux/Mac, use LinuxMac.sh

    1. Right click on the webpage and select “Save as…” to download the file

    2. Make it executable via chmod +x LinuxMac.sh

    3. Execute it and follow further instructions ./LinuxMac.sh

    • For fully automatic installation, run it with the -a parameter ./LinuxMac.sh -a

Manual

Mamba based installation

We strongly recommend to start from Mambaforge, a community project of the conda-forge community.

As Beginner start with an installation of Mambaforge - Get mambaforge for your Operation System

Install MSS

You must install mss into a new environment to ensure the most recent versions for dependencies.

$ mamba create -n mssenv
$ mamba activate mssenv
(mssenv) $ mamba install mss=8.3.2 python
(mssenv) $ msui

Mamba Server based installation example

For a wms server setup or mscolab setup you may want to have a dedicated user for the apache2 wsgi script. We suggest to create a mss user.

  • create a mss user on your system

  • login as mss user

  • create a src directory in /home/mss

  • cd src

  • get mambaforge

  • set execute bit on install script

  • execute script, enable environment in .bashrc

  • login again

  • mamba create -n mssenv

  • mamba activate mssenv

  • mamba install mss=8.3.2 python

For a simple test you could start the builtin standalone mswms and mscolab server:

$ mswms &
$ mscolab start

Point a browser for the verification of both servers installed on

Further details in the components section on http://mss.rtfd.io

Conda based installation

Anaconda provides an enterprise-ready data analytics platform that empowers companies to adopt a modern open data science analytics architecture.

Warning

Installing Mamba in Anaconda setup is not recommended. We strongly recommend to use the Mambaforge method (see above).

Please add the channel conda-forge to your defaults:

$ conda config --add channels conda-forge

The conda-forge channel must be on top of the list before the anaconda default channel.

Install MSS

You must install mss into a new environment to ensure the most recent versions for dependencies.

$ conda install -n base conda-libmamba-solver
$ conda create -n mssenv
$ conda activate mssenv
(mssenv) $ conda install mss=8.3.2 python --solver=libmamba
(mssenv) $ msui

Update

Builtin Update

Since version 5.0 we provide a feature for updating MSS by the UI or the command line After you started the MSS UI it informs you after a while if there is a new update available. From the command line you can trigger this update feature by

(mssenv) $ msui --update

Other Methods

For updating an existing MSS installation to the current version, it is best to install it into a new environment. If your current version is not far behind the new version you could try the mamba update mss as described.

Important

mamba is under development. All dependencies of MSS and MSS itselfs are under development. Sometimes this update feature of mamba can’t resolve from existing to new dependencies.

search for MSS what you can get

(mssenv) $ mamba search mss
...
mss                            8.3.2  py38h578d9bd_0  conda-forge
mss                            8.3.2  py39hf3d152e_0  conda-forge

compare what you have installed

(mssenv) $ mamba list mss

  mss                            7.0.2     py310hff52083_0    conda-forge

We have reports that often an update suceeds by using the install option and the new version number, in this example 8.3.2 and python as second option

(mssenv) $ mamba install mss=8.3.2 python

All attemmpts show what you get if you continue. Continue only if you get what you want.

The alternative is to use a new environment and install mss.

For further details of configurating mss Configuration of MSUI

Docker Instance

You can use images from the docker hub. based on our repository

Build settings are based on the stable branch. Our openmss/mss:latest has any update in the stable branch.

You can start server and client by loading the image

$ xhost +local:docker
$ docker run -ti --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix --network host openmss/mss:latest  /bin/bash
$ /opt/conda/envs/mssenv/bin/msui &
$ /opt/conda/envs/mssenv/bin/mswms --port 80 &
$ /opt/conda/envs/mssenv/bin/mscolab start &
$ curl http://localhost/?service=WMS&request=GetCapabilities&version=1.1.1
$ curl http://localhost:8083/status

The WMS server initialized by demodata, and the mscolab server and the userinterface can be started by

$  xhost +local:docker
$  docker run -d -ti --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix --network host openmss/mss:latest MSS

If you want only to start the msui do this by

$  xhost +local:docker
$  docker run -d -ti --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix --network host openmss/mss:latest msui

We have not setup keyring in the docker container. When a login is needed you will get a message:

INFO: Can’t use Keyring on your system: No recommended backend was available. Install a recommended 3rd party backend package; or, install the keyrings.alt package if you want to use the non-recommended backends. See https://pypi.org/project/keyring for details.

For using keyring in a the openmss/mss container you need to start the container with different options and after installing gnome-keyring you have to configure it.

$ xhost +local:docker
$ docker run -ti --ulimit nofile=65536:65536 --cap-add=IPC_LOCK --rm -e DISPLAY=$DISPLAY -v /tmp/.X11-unix/:/tmp/.X11-unix --network host openmss/mss:latest  /bin/bash
$ apt update
$ apt install gnome-keyring
$ conda activate mssenv
$ dbus-run-session -- sh # start a new D-bus shell, prompt changes to a hash
# echo 'credpass' | gnome-keyring-daemon --unlock # unlock the systems keyring
# msui # starts msui

Singularity

You can use images from the docker hub. based on our repository by converting them to singularity or build from our singularity definition

Build settings are based on the stable branch. Our openmss/mss:latest has any update in the stable branch.

You can start server and client by loading the image

$ host +
$ singularity build -f mss.sif Singularity.def
$ singularity shell mss.sif
$ Singularity > msui # starts the ui
$ Singularity > mswms_demodata --seed  # creates in your $HOME a mss/ folder with testdata
$ Singularity > export PYTHONPATH=$HOME/mss; mswms # starts the development server
$ Singularity > mscolab db --init; mscolab start # starts the mscolab development server