Awesome Open Source
Awesome Open Source
Sponsorship

LinkedPipes ETL

Build Status

LinkedPipes ETL is an RDF based, lightweight ETL tool.

Requirements

For building locally

Installation and startup

You can run LP-ETL in Docker, or build it from the source.

Docker

To start LP-ETL master branch on http://localhost:8080, you can use a one-liner:

curl https://raw.githubusercontent.com/linkedpipes/etl/master/docker-compose.yml | docker-compose -f - up

Alternatively, you can clone the entire repository

git clone https://github.com/linkedpipes/etl.git

and run

docker-compose up

Note that this uses just the docker-compose.yml file, so the rest of the cloned repository is useless.

You may need to run the commands as sudo or be in the docker group.

Configuration

Each component (executor, executor-monitor, storage, frontend) has separate Dockerfile.

Environment variables:

  • LP_ETL_BUILD_BRANCH - The Dockerfiles are designed to run build from the github repository, the branch is set using this property, default is master.
  • LP_ETL_BUILD_JAVA_TEST - Set to empty to allow to run Java tests, this will slow down the build.
  • LP_ETL_DOMAIN - The URL of the instance, this is used instead of the domain.uri from the configuration.
  • LP_ETL_FTP - The URL of the FTP server, this is used instead of the executor-monitor.ftp.uri from the configuration.

For Docker Compose, there are additional environment variables:

  • LP_ETL_PORT - Specify port mapping for frontend, this is where you can connect to your instance. This does NOT have to be the same as port in LP_ETL_DOMAIN in case of reverse-proxying.

For example to run LP-ETL from develop branch on http://localhost:9080 use can use following command:

curl https://raw.githubusercontent.com/linkedpipes/etl/develop/docker-compose.yml | LP_ETL_PORT=9080 LP_ETL_DOMAIN=http://localhost:9080 docker-compose -f - up

docker-compose utilizes several volumes that can be used to access/provide data. See docker-compose.yml comments for examples and configuration. You may want to create your own docker-compose.yml for custom configuration.

From source on Linux

Installation

$ git clone https://github.com/linkedpipes/etl.git
$ cd etl
$ mvn install

Configuration

The configuration file deploy/configuration.properties can be edited, mainly changing paths to working, storage, log and library directories.

Startup

$ cd deploy
$ ./executor.sh >> executor.log &
$ ./executor-monitor.sh >> executor-monitor.log &
$ ./storage.sh >> storage.log &
$ ./frontend.sh >> frontend.log &

Running LP-ETL as a systemd service

See example service files in the deploy/systemd folder.

From source on Windows

Note that it is also possible to use Bash on Ubuntu on Windows or Cygwin and proceed as with Linux.

Installation

git clone https://github.com/linkedpipes/etl.git
cd etl
mvn install

Configuration

The configuration file deploy/configuration.properties can be edited, mainly changing paths to working, storage, log and library directories.

Startup

In the deploy folder, run

  • executor.bat
  • executor-monitor.bat
  • storage.bat
  • frontend.bat

Plugins - Components

The components live in the jars directory. Detailed description of how to create your own is coming soon, in the meantime, you can copy an existing component and change it.

Update notes

Update note 5: 2019-09-03 breaking changes in the configuration file. Remove /api/v1 from the executor-monitor.webserver.uri, so it loolks like: executor-monitor.webserver.uri = http://localhost:8081. You can also remove executor.execution.uriPrefix as the value is derived from domain.uri.

Update note 4: 2019-07-03 we changed the way frontend is run. If you do not use our script to run it, you need to update yours.

Update note 3: When upgrading from develop prior to 2017-02-14, you need to delete {deploy}/jars and {deploy}/osgi.

Update note 2: When upgrading from master prior to 2016-11-04, you need to move your pipelines folder from e.g., /data/lp/etl/pipelines to /data/lp/etl/storage/pipelines, update the configuration.properites file and possibly the update/restart scripts as there is a new component, storage.

Update note: When upgrading from master prior to 2016-04-07, you need to delete your old execution data (e.g., in /data/lp/etl/working/data)


Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
java (29,660
etl (82
rdf (76
linked-data (36

Find Open Source By Browsing 7,000 Topics Across 59 Categories