gsienkopl / kartoza-docker-mapproxy

A docker recipe for http://mapproxy.org/

Mapproxy Dockerfile

This will build a docker image that runs mapproxy .

Getting the image

There are various ways to get the image onto your system:

The preferred way (but using most bandwidth for the initial image) is to get our docker trusted build like this:

docker pull kartoza/mapproxy

To build the image yourself do:

docker build -t kartoza/mapproxy git://github.com/kartoza/docker-mapproxy

To build using a local url instead of directly from github.

git clone git://github.com/kartoza/docker-mapproxy

docker build -t kartoza/mapproxy .

Note: We do not use tagged versions as we install the latest version of mapproxy.

Environment variables

The image specifies a couple of environment variables

• MAPPROXY_DATA_DIR=path to store configuration files when running single app mode
• MULTI_MAPPROXY_DATA_DIR=path to store configuration files when running multi app mode
• PROCESSES=number of processes to run uwsgi in. Only available when running the production version.
• CHEAPER=Minimum number of workers allowed. This should always be lower than the env PROCESSES
• THREADS=maximum number of parallel threads to run production instance with.
• PRODUCTION=Boolean value to indicate if you need to run develop server or using uwsgi
• MULTI_MAPPROXY=Boolean value to indicate if you need to run multi mapproxy. Defaults to false
• ALLOW_LISTING=Allows listing all config files in multi map mode
• LOGGING=Boolean value to indicate if you need to activate logging. Useful when using uwsgi (not in multi-app mode)
• ENABLE_S3_CACHE=Boolean value to indicate if support for the S3 storage backend should be enabled.
• AWS_ACCESS_KEY_ID=The S3 backend username value
• AWS_SECRET_ACCESS_KEY=The S3 backend password value
• CREATE_DEFAULT_S3_BUCKETS=Boolean value to indicate if mapproxy should try to create buckets when the container starts. If MinIO starts up with the MapProxy container, it will not have a bucket available by default, so S3 based caching will not work until a bucket is created. The default value is false.
• S3_BUCKET_LIST=A list of bucket names to check for on the S3 endpoint, and create if they are not available. Supports space, comma, and semi-colon separated values. Default value is mapproxy.
• S3_BUCKET_ENDPOINT=The endpoint for your S3 service. Unless you are using S3 from AWS directly, you most likely want to set this to http://minio:9000/ or the URL of your external S3 service.

Mounting Configs

if running in production you can specify any uwsgi parameters.

You can mount the uwsgi.ini to a path inside the container thus overriding a lot of the uwsgi default settings.

-v /data/uwsgi.ini:/settings/uwsgi.ini

Run

To run a mapproxy container do:

docker run --name "mapproxy" -p 8080:8080 -d -t kartoza/mapproxy

Typically, you will want to mount the mapproxy volume, otherwise you won't be able to edit the configs:

In single app mode

mkdir mapproxy
docker run --name "mapproxy" -p 8080:8080 -d -t -v `pwd`/mapproxy:/mapproxy kartoza/mapproxy

In multi mode app

mkdir multi_mapproxy
docker run --name "mapproxy" -p 8080:8080 -d -t -v `pwd`/multi_mapproxy:/multi_mapproxy kartoza/mapproxy

The first time your run the container, mapproxy basic default configuration files will be written into /mapproxy or multi_mapproxy volumes. 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 ./configuration/cache_data externally or any other path that is defined by the mapproxy.yaml.

Note that the mapproxy containerised application will run as the user that owns the /mapproxy folder. The UID:GID of the process will be 1000:1000. If you are mounting existing config directory i.e. ./configuration folder, you need to set the folder permission with chown -R 1000:1000 ./configuration from this directory.

docker-compose

You can set up the services using the docker-compose. The docker-compose sets up the QGIS server container and links it to the mapproxy container and nginx for reverse proxy.

An index.html is provided in the web folder to preview the layers in mapproxy.

Reverse proxy

The mapproxy container 'speaks' uwsgi protocol so you can also put nginx in front of it to receive http request and translate it to uwsgi (try the nginx docker container). However, our sample configuration by default make uwsgi uses http socket instead of socket parameter (uwsgi protocol). A sample configuration (via linked containers) that will forward traffic into the uwsgi container, adding the appropriate headers as needed is provided via docker-compose

Take a look at the docker-compose to look at linking two or more containers

Once the service is up and running you can connect to the default demo mapproxy service by pointing QGIS' WMS client to the mapproxy service. In the example below the nginx container is running on localhost on port 8080.

http://localhost/mapproxy/service/?

S3 storage backend

MapProxy supports the S3 storage backend for data caching. This provides a number of benefits, including the ability to decouple MapProxy and more readily scale your solution with multiple instances of MapProxy sharing the same storage backend without having to concern yourself with io locks or access collisions.

We provide an example implementation with mapproxy-s3.yaml, which is used in the docker-compose-s3.yml implementation, to configure an S3 backend for certain services.

docker compose -f ${pwd}docker-compose-s3.yml up -d

Then review the example service at http://localhost/s3sample.html. Ensure that minio_admin and secure_minio_secret are stored and used as environment variables in production deployments.

This implementation mounts MinIO - an S3 Compatible container service - as a storage backend for MapProxy

MinIO can be accessed from http://localhost:9001 or using the minio api endpoint from http://localhost:9000

Note that the MinIO service does not support subpaths or routes on the web server and any reverse proxy will need to be implemented at the web root, using a dedicated subdomain. Note as well that MinIO provides a Console (web-ui), and an API endpoint as distinct services which serve different functions... api calls to the console will return errors.

You can use this methodology to serve as a proxy for other storage solutions, for example, using MinIO as a proxy for Microsoft Azure Blob Storage.

---

Tim Sutton ([email protected]) Admire Nyakudya ([email protected]) March 2022