Skip to content

docker.rst

Latest commit

 

History

History
98 lines (64 loc) · 3.23 KB

docker.rst

File metadata and controls

98 lines (64 loc) · 3.23 KB
.. index::
    single: docker

Docker

Docker is popular container platform. With help of Docker, we can easily distribute VTS* services and split the whole needed infrastructure into more microservices

In this document, we are going to discuss, how to setup VTSD and Mapproxy servers along with Nginx server for production.

Note

Before you start, you should be familiar with basic concepts of Docker and have Docker installed.

Getting the configuration

Docker configuration is distributed along with the vts-backend package. You have to git clone the project and dpkg-buildpackage the content, after that, all needed files are available in the vts-backend project directory and we can proceed to generate the Docker images and start te Containers.

git clone --recursive https://github.com/melowntech/vts-backend.git
cd vts-backend
make

Note

You may obtain dpkg-buildpackage: error: failed to sign .changes file error after your run make. That is ok you can still continue

Build and run the VTSD Docker Container

To get :ref:`vtsd` container up and running, use:

docker build -t vts-vtsd:latest -f docker/vtsd/Dockerfile .

If you were trying to follow one of our :ref:`backend tutorials <backend-tutorials>`, you may have the storage directory already in your host computer. You can therefore start the container with existing directory mounted as volume:

docker run -ti -p 3060:3060 -v /path/to/vtsd-storage:/var/vts/store:rw --name vtsd vts-vtsd

Build and run the Mapproxy Docker Container

First build the Docker image:

docker build -t vts-mapproxy:latest -f docker/mapproxy/Dockerfile .

You can now run the container directly, there are already examples being part of the deployment.

But again, if you followed some :ref:`backend tutorials <backend-tutorials>`, you have have the datasets directory already available along with the resources.json file.:

`- project/
      `- resources.json
      `- datasets/

You can mount it as Bind volume.

docker run -ti -p 3070:3070 -v /my/projects/project:/vts/datasets/project --name mapproxy vts-mapproxy

VTS-Mapproxy is configured, so that

  • it is excpeting /vts/datasets/.../resources.json files to be at the image available.
  • The resources.json shall cotain absolute paths to existing datasets.

Download and run containers from Docker Hub directly

Melown Technologies have account at Docker Hub, you can download and use images directly, e.g.:

docker run -ti -p 3070:3070 --name mapproxy --rm \
    -v /projects/mapproxy-wmts:/vts/datasets/mapproxy-wmts:rw \
    melown/vts-mapproxy