Add Hive

[piotr-kalanski]

https://hub.docker.com/r/bde2020/hive/

[piotr-kalanski]

https://github.com/big-data-europe/docker-hive-metastore-postgresql

[piotr-kalanski]

• Option 1 - resign from Hive and instead use metastore created by Spark. Consider using the same directory and bind mount to easily add data form host machine: https://stackoverflow.com/questions/45819568/why-there-are-many-spark-warehouse-folders-got-created

• Option 2 - use Hive https://hub.docker.com/r/bde2020/hive/ image and copy hive-site.xml to spark container

  • Challenge is to make sure two containers have the same hive-site.xml

• Option 3 - build custom Docker image with Spark and Hive to share the same hive-site.xml

  • Drawback: quite a lot of work for dev and maintenance of image