Docker Container for Jena-Fuseki

Docker container build workflow for Jena-Fuseki forked from blankdots configuration.

SHACL service operation

In Jena-Fuseki 3.13.0 SHACL validation has been added as an option for a service endpoint. Instructions are available for configuring the endpoint in the Fuseki configuration file. The /ds endpoint is configured for SHACL validation.

To upload data set fu:

curl -XPOST --data-binary @fu-data.ttl    \  
     --header 'Content-type: text/turtle' \  
     'http://localhost:3030/ds?default'

Validate with shapes in fu-shapes.ttl and get back a validation report:

curl -XPOST --data-binary @fu-shapes.ttl  \  
     --header 'Content-type: text/turtle' \  
     'http://localhost:3030/ds/shacl?graph-default'

Original Readme updated for Jena-Fuseki 3.13.0

Current container is based on https://registry.hub.docker.com/u/stain/jena-fuseki/

In order to make it more production friendly it adds:

• gosu (running with limited permissions as non-root user) - https://github.com/tianon/gosu
• configuration loader from file - and it provides two endpoints; one for test and one for working with
• FUSEKI 3.13.0

Run image

Skip the build and just run with plain configuration:

• docker pull blankdots/jena-fuseki
• docker run -p 3030:3030 blankdots/jena-fuseki vanilla run
• docker run -p 3030:3030 -v /data/fuseki:/data/fuseki/fuseki_DB blankdots/jena-fuseki run with volume attached
• docker run -p 3030:3030 -v /data/fuseki:/data/fuseki/fuseki_DB -d blankdots/jena-fuseki run volume attached in detached mode
• docker run -p 3030:3030 -v /data/fuseki:/data/fuseki/fuseki_DB -e ADMIN_PASSWORD=pass blankdots/jena-fuseki preset the password.

A password will be generated if one not provided at runtime (see container logs).

For other parameters see: https://hub.docker.com/r/stain/jena-fuseki/

Build and Configuration Instructions

The configuration of the store is done via the config.ttl file. which contains two datasets /test (in memory) and /ds (persistent on disk if the image is run with a volume attached) and provides two named graphs.

TO DO: provide other configuration scenarios. Some envisioned scenarios could be envisioned where:

• an endpoint is used for just the sparql queries and there is one for uploading the data and materialising the results in the SPARQL endpoint;
• there is a unuion model type of endpoint which provides via SPARQL materialised results of an inference type of endpoint (which only works in memory);
• an endpoint that provides search based like functionalities;
• etc.

Data asscoiated to the configuration is present in the data folder.

Once the configuration and associated data building the image is as follows:

• docker build -t blankdots/jena-fuseki .