The scripts under gsky/docker
are meant for building GSKY docker images from
scratch. To build a docker image, please run the following command:
cd docker
docker build --build-arg sample_data_dir=<geoglam sample data dir> \
--build-arg gsky_repo="<gsky git repo branch>" .
-
sample_data_dir
is the directory path for the sample data. To obtain sample data for reproducible builds, please pull thelatest
tagged image using the following command:docker pull gjmouse/gsky:latest
. The sample data files are under/gpath
of thelatest
image. Please also be aware thatsample_data_dir
must be a relative path undergsky/docker
directory. This is a restriction imposed by dockerADD
command. -
gsky_repo
is an optional build argument to specify any git repository/branch other thanhttp://github.com/nci/gsky.git
. For example, one might want to build an image for a feature branch using the following build arguments:--build-arg gsky_repo="-b feature_branch http://github.com/<your own repo>
The Dockerfile
publishes port 8080 and 8888 for GSKY ows and MAS API services.
To access these ports from the host OS, one needs to use the -p
option for
docker run
. For example, docker run --rm -it -p 8080:8080 <GSKY image>
The sample data were taken from a subset of Geoglam Fractional Cover http://www.geo-rapp.org/rapp-monitor/, which covers the entire Australia. The sample data consist of 16 data files whose total size is about 311MB.
The Dockerfile
also bundles TerriaJS https://github.com/TerriaJS/terriajs
for the purpose of visually demostrating GSKY WMS services on the sample data.
To access TerriaJS from your web browser, please do the following:
-
Run
docker run --rm -it -p 8080:8080 <GSKY image>
to bring up a GSKY container. -
TerriaJS can be accessed from
http://127.0.0.1:8080/terria
from your web browser.