This will build a docker image that runs mapproxy .
This recipe is adapted from the one provided by Tim Sutton https://github.com/kartoza/docker-mapproxy
Python 3.4 Mapproxy 1.9.0
To build this image yourself you need to clone this repo locally first. Then build using a local url instead of directly from github.
git clone git://github.com/thinkwhere/mapproxy-docker
Then build using the command:
docker build -t thinkwhere/mapproxy .
or simply:
./build.sh
To run the mapproxy container do:
./run.sh
This will create and mount a "mapproxy" folder in your current working directory as a volume in the container. This is used to hold the config .yaml files. Mounting this volume allows you to create the config files without having to rebuild the image.
mkdir mapproxy
docker run --name "mapproxy" -p 8080:8080 -d -t -v \
`pwd`/mapproxy:/mapproxy thinkwhere/mapproxy-docker
The first time your run the container, mapproxy basic default configuration
files will be written into ./mapproxy. You should read the mapproxy documentation
on how to configure these files and create appropriate service definitions for
your WMS services. Then restart the container to activate your changes.
The cached wms tiles will be written to ./mapproxy/cache_data.
Note that the mapproxy containerised application will run as the user that owns the /mapproxy folder.
The mapproxy container 'speaks' uwsgi so you need to put NGINX in front of it.
This can be achived using Docker Compose, a tool for defining and running multi-container
Docker applications.
A docker-compose file is included, which will start both the MapProxy container and
and an NGINX container together, with the appropriate ports exposed and links between
the two containers.
Note docker-compose ships with the Docker Toolbox for Win/Mac users. Linux users will need to install seperately.
To start the containers with Compose:
docker-compose up
In your browser navigate to:
http://localhost/demo