Docker Pentaho Data Integration

Introduction

DockerFile for Pentaho Data Integration (a.k.a kettel / PDI)

This image is intendend to allow execution os PDI transformations and jobs throught command line and run PDI's UI (Spoon). PDI server (Carter) is available on this image.

Quick start

Basic Syntax

$ docker container run --rm andrespp/pdi

Usage:	/entrypoint.sh COMMAND

Pentaho Data Integration (PDI)

Options:
  runj filename		Run job file
  runt filename		Run transformation file
  spoon			    Run spoon (GUI)
  help		         Print this help

Running Transformations

$ docker container run --rm -v $(pwd):/jobs andrespp/pdi runt sample/dummy.ktr

Running Jobs

$ docker container run --rm -v $(pwd):/jobs andrespp/pdi runj  sample/dummy.kjb

Running Spoon (UI)

Using docker run

$ docker run -it --rm -v /tmp/.X11-unix/:/tmp/.X11-unix/:ro \
        -v $(pwd):/jobs \
        -e XAUTH=$(xauth list|grep `uname -n` | cut -d ' ' -f5) -e "DISPLAY" \
        --name spoon \
        andrespp/pdi spoon

Using startup script (Installing)

In order to run the container as if the application was installed locally, download the spoon script to a directory in you $PATH, for example:

$ sudo curl -fsSL https://raw.githubusercontent.com/andrespp/docker-pdi/master/spoon \
       -o /usr/local/bin/spoon
$ sudo chmod +x /usr/local/bin/spoon

Then you'll be able to run JupyterLab in the current directory simply by calling jlab:

$ spoon

Custom kettle.properties

In order to use a custom kettle.properties, you need to leave the file available in /jobs/kettle.properties.

$ # Custom properties in $(pwd)/kettle.properties
$ docker container run --rm -v $(pwd):/jobs andrespp/pdi runj  sample/dummy.kjb

Environment variables

This image uses several environment variables in order to control its behavior, and some of them may be required

Environment variable	Default value	Note
PDI_VERSION	7.1

Issues

If you have any problems with or questions about this image, please contact me through a GitHub issue.