docker-geoserver

A simple docker container that runs GeoServer influenced by this docker recipe: https://github.com/eliotjordan/docker-geoserver/blob/master/Dockerfile

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/geoserver

To build yourself with a local checkout using the build script:

Edit the build script to change the following variables:

• The variables below represent the latest stable release you need to build. i.e 2.15.2

  BUGFIX=2
  MINOR=15
  MAJOR=2
  

git clone git://github.com/kartoza/docker-geoserver
cd docker-geoserver
./build.sh

Ensure that you look at the build script to see what other build arguments you can include whilst building your image.

If you do not intend to jump between versions you need to specify that in the build script.

Building with war file from a URL

If you need to build the image with a custom GeoServer war file that will be downloaded from a server, you can pass the war file url as a build argument to docker, example:

docker build --build-arg WAR_URL=http://download2.nust.na/pub4/sourceforge/g/project/ge/geoserver/GeoServer/2.13.0/geoserver-2.13.0-war.zip --build-arg GS_VERSION=2.13.0

Note: war file version should match the version number provided by GS_VERSION argument otherwise we will have a mismatch of plugins and GeoServer installed.

Building with Oracle JDK

Download jdk-8u201-linux-x64.tar.gz or the latest version from Oracle Java and save the contents into the resources folder. This used to be done by the setup scripts but no longer works due to the changes in the licencing terms from Oracle which require users to login to their site.

To replace OpenJDK Java with the Oracle JDK, set build-arg ORACLE_JDK=true:

docker build --build-arg ORACLE_JDK=true --build-arg GS_VERSION=2.13.0 -t kartoza/geoserver .

Building with plugins

Inspect setup.sh to confirm which plugins (community modules or standard plugins) you want to include in the build process, then add them in their respective sections in the script.

You should ensure that the plugins match the version for the GeoServer WAR zip file.

Removing Tomcat extras during build

To remove Tomcat extras including docs, examples, and the manager webapp, set the TOMCAT_EXTRAS build-arg to false:

docker build --build-arg TOMCAT_EXTRAS=false --build-arg GS_VERSION=2.13.0 -t kartoza/geoserver .

Building with specific version of Tomcat

To build using a specific tagged release for tomcat image set the IMAGE_VERSION build-arg to 8-jre8: See the dockerhub tomcat to choose which tag you need to build against.

docker build --build-arg IMAGE_VERSION=8-jre8 --build-arg GS_VERSION=2.13.0 -t kartoza/geoserver:2.13.0 .

Building with file system overlays (advanced)

The contents of resources/overlays will be copied to the image file system during the build. For example, to include a static Tomcat setenv.sh, create the file at resources/overlays/usr/local/tomcat/bin/setenv.sh.

You can use this functionality to write a static GeoServer directory to /opt/geoserver/data_dir, include additional jar files, and more.

Overlay files will overwrite existing destination files, so be careful!

Build with CORS Support

The contents of resources/overlays will be copied to the image file system during the build. For example, to include a static web xml with CORS support web.xml, create the file at resources/overlays/usr/local/tomcat/conf/web.xml.

Run (manual docker commands)

Note: You probably want to use docker-compose for running as it will provide a repeatable orchestrated deployment system.

You probably want to also have PostGIS running too. To create a running container do:

docker run --name "postgis" -d -t kartoza/postgis:9.4-2.1
docker run --name "geoserver"  --link postgis:postgis -p 8080:8080 -d -t kartoza/geoserver

You can also use the following environment variables to pass a user name and password to PostGIS:

• -e USERNAME=<PGUSER>
• -e PASS=<PGPASSWORD>

You can also use the following environment variables to pass arguments to GeoServer:

• GEOSERVER_DATA_DIR=<PATH>

• ENABLE_JSONP=<true or false>

• MAX_FILTER_RULES=<Any integer>

• OPTIMIZE_LINE_WIDTH=<false or true>

• FOOTPRINTS_DATA_DIR=<PATH>

• GEOWEBCACHE_CACHE_DIR=<PATH>

• GEOSERVER_ADMIN_PASSWORD=<password>

• Tomcat properties:

  • You can change the variables based on geoserver container considerations. These arguments operate on the -Xms and -Xmx options of the Java Virtual Machine
  • INITIAL_MEMORY=<size> : Initial Memory that Java can allocate, default 2G
  • MAXIMUM_MEMORY=<size> : Maximum Memory that Java can allocate, default 4G

Note:

Changing GeoServer password on runtime

The default GeoServer user is 'admin' and the password is 'geoserver'. You can pass the environment variable GEOSERVER_ADMIN_PASSWORD to change it on runtime.

docker run --name "geoserver"  -e GEOSERVER_ADMIN_PASSWORD=myawesomegeoserver -p 8080:8080 -d -t kartoza/geoserver

Run (automated using docker-compose)

We provide a sample docker-compose.yml file that illustrates how you can establish a GeoServer + PostGIS + GeoGig orchestrated environment with nightly backups that are synchronised to your backup server via btsync.

If you are not interested in the backups, GeoGig and btsync options, comment out those services in the docker-compose.yml file.

If you start the stack using the compose file make sure you login into GeoServer using username:admin and password:myawesomegeoserver as specified by the env file geoserver.env

Please read the docker-compose documentation for details on usage and syntax of docker-compose - it is not covered here.

If you are interested in btsync backups, install [Resilio sync] on your desktop NAS or other backup destination and create two folders:

• one for database backup dumps
• one for geoserver data dir

Then make a copy of each of the provided EXAMPLE environment files e.g.:

cp docker-env/btsync-db.env.EXAMPLE docker-env/btsync-db.env
cp docker-env/btsync-media.env.EXAMPLE docker-env/btsync-media.env

Then edit the two env files, placing your Read/Write Resilio keys in the place provided.

To run the example do:

docker-compose up

Which will run everything in the foreground giving you the opportunity to peruse logs and see that everything spins up nicely.

Once all services are started, test by visiting the GeoServer landing page in your browser: http://localhost:8600/geoserver.

To run in the background rather, press ctrl-c to stop the containers and run again in the background:

docker-compose up -d

Note: The docker-compose.yml uses host based volumes so when you remove the containers, all data will be kept. Using host based volumes ensures that your data persists between invocations of the compose file. If you need to delete the container data you need to run docker volume prune. Pruning the volumes will remove all the storage volumes that are not in use so users need to be careful of such a move. Either set up btsync (and test to verify that your backups are working, we take no responsibility if the examples provided here do not produce a reliable backup system).

Run (automated using rancher)

An even nicer way to run the examples provided is to use our Rancher Catalogue Stack for GeoServer. See http://rancher.com for more details on how to set up and configure your Rancher environment. Once Rancher is set up, use the Admin -> Settings menu to add our Rancher catalogue using this URL:

https://github.com/kartoza/kartoza-rancher-catalogue

Once your settings are saved open a Rancher environment and set up a stack from the catalogue's 'Kartoza' section - you will see GeoServer listed there.

If you want to synchronise your GeoServer settings and database backups (created by the nightly backup tool in the stack), use Resilio sync to create two Read/Write keys:

• one for database backups
• one for GeoServer media backups

Note: Resilio sync is not Free Software. It is free to use for individuals. Business users need to pay - see their web site for details.

You can try a similar approach with Syncthing or Seafile (for free options) or Dropbox or Google Drive if you want to use another commercial product. These products all have one limitation though: they require interaction to register applications or keys. With Resilio Sync you can completely automate the process without user intervention.

Storing data on the host rather than the container.

Docker volumes can be used to persist your data.

If you need to use geoserver data directory that contains sample examples and configurations download it from geonode site as indicated below:

# Example - ${GS_VERSION} is the geoserver version i.e 2.13.0
wget http://build.geonode.org/geoserver/latest/data-2.13.x.zip
unzip data-2.13.x.zip -d ~/geoserver_data
cp scripts/controlflow.properties ~/geoserver_data
chmod -R a+rwx ~/geoserver_data
docker run -d -p 8580:8080 --name "geoserver" -v $HOME/geoserver_data:/opt/geoserver/data_dir kartoza/geoserver:${GS_VERSION}

Create an empty data directory to use to persist your data.

mkdir -p ~/geoserver_data && chmod -R a+rwx ~/geoserver_data
docker run -d -v $HOME/geoserver_data:/opt/geoserver/data_dir kartoza/geoserver

Control flow properties

The control flow module is installed by default and it is used to manage request in geoserver. In order to customise it based on your resources and use case read the instructions from documentation. Modify the file scripts/controlflow.properties before building the image.

Credits

• Tim Sutton (tim@kartoza.com)
• Shane St Clair (shane@axiomdatascience.com)
• Alex Leith (alexgleith@gmail.com)
• Admire Nyakudya (admire@kartoza.com)
• Gavin Fleming (gavin@kartoza.com)