Apache Griffin

Apache Griffin is a model driven data quality solution for modern data systems. It provides a standard process to define data quality measures, execute, report, as well as an unified dashboard across multiple data systems. You can access our home page here. You can access our wiki page here. You can access our issues jira page here.

Contact us

Dev List





How to run in docker

  1. Install docker.
  2. Pull our built docker image.
    docker pull bhlx3lyx7/griffin_demo:0.0.1
  3. Increase vm.max_map_count of your local machine, to use elasticsearch.
    sysctl -w vm.max_map_count=262144
  4. Run this docker image, wait for about one minute, then griffin is ready.
    docker run -it -h sandbox --name griffin_demo -m 8G --memory-swap -1 \
    -p 32122:2122 -p 37077:7077 -p 36066:6066 -p 38088:8088 -p 38040:8040 \
    -p 33306:3306 -p 39000:9000 -p 38042:8042 -p 38080:8080 -p 37017:27017 \
    -p 39083:9083 -p 38998:8998 -p 39200:9200 bhlx3lyx7/griffin_demo:0.0.1
  5. Now you can visit UI through your browser, login with account “test” and password “test” if required.
    http://<your local IP address>:38080/
    You can also follow the steps using UI here.

How to deploy and run at local

  1. Install jdk (1.8 or later versions).
  2. Install mysql.
  3. Install npm (version 6.0.0+).
  4. Install Hadoop (2.6.0 or later), you can get some help here.
  5. Install Spark (version 1.6.x, griffin does not support 2.0.x at current), if you want to install Pseudo Distributed/Single Node Cluster, you can get some help here.
  6. Install Hive (version 1.2.1 or later), you can get some help here. You need to make sure that your spark cluster could access your HiveContext.
  7. Install Livy, you can get some help here. Griffin need to schedule spark jobs by server, we use livy to submit our jobs. For some issues of Livy for HiveContext, we need to download 3 files, and put them into Hdfs.
  8. Install ElasticSearch. ElasticSearch works as a metrics collector, Griffin produces metrics to it, and our default UI get metrics from it, you can use your own way as well.
  9. Modify configuration for your environment. You need to modify the configuration part of code, to make Griffin works well in you environment. service/src/main/resources/application.properties
    spring.datasource.url = jdbc:mysql://<your IP>:3306/quartz?autoReconnect=true&useSSL=false
    spring.datasource.username = <user name>
    spring.datasource.password = <password>
    hive.metastore.uris = thrift://<your IP>:9083
    hive.metastore.dbname = <hive database name>    # default is "default"
    sparkJob.file = hdfs://<griffin measure path>/griffin-measure.jar
    sparkJob.args_1 = hdfs://<griffin env path>/env.json
    sparkJob.jars_1 = hdfs://<datanucleus path>/datanucleus-api-jdo-3.2.6.jar
    sparkJob.jars_2 = hdfs://<datanucleus path>/datanucleus-core-3.2.10.jar
    sparkJob.jars_3 = hdfs://<datanucleus path>/datanucleus-rdbms-3.2.9.jar
    sparkJob.uri = http://<your IP>:8998/batches
    ES_SERVER = "http://<your IP>:9200"
    Configure measure/measure-batch/src/main/resources/env.json for your environment, and put it into Hdfs /
  10. Build the whole project and deploy.(NPM should be installed , on mac you can try ‘brew install node’)
    mvn install
    Create a directory in Hdfs, and put our measure package into it.
    cp /measure/target/measure-0.1.3-incubating-SNAPSHOT.jar /measure/target/griffin-measure.jar
    hdfs dfs -put /measure/target/griffin-measure.jar <griffin measure path>/
    After all our environment services startup, we can start our server.
    java -jar service/target/service.jar
    After a few seconds, we can visit our default UI of Griffin (by default the port of spring boot is 8080).
    http://<your IP>:8080
  11. Follow the steps using UI here.

Note: The front-end UI is still under development, you can only access some basic features currently.


See CONTRIBUTING.md for details on how to contribute code, documentation, etc.