Trino Alluxio Iceberg Minio

Trino + Alluxio + Iceberg + Minio setup doc

Successful setup from community user (Roman Klimenko) on 01/18/2024, kept here as reference.

follow this doc: https://github.com/bitsondatadev/trino-getting-started/blob/main/iceberg/trino-alluxio-iceberg-minio/README.md

image versions in original docker-compose.yaml needs to be updated with

• image: 'trinodb/trino:422' (instead of 'trinodb/trino:latest' under trino-coordinator)
• image: 'alluxio/alluxio:2.9.3' (instead of 'alluxio/alluxio:latest' under alluxio-leader)
• entrypoint: all "2.9.0" -> "2.9.3" (under hive-metastore)

Unfortunately starting with Trino:423 will get an error

• Invalid location URI: alluxio://alluxio-leader:19998/lakehouse
• Trino shows it as: org.apache.hadoop.fs.UnsupportedFileSystemException: No FileSystem for scheme "alluxio" while trying to create a schema.

The working docker-compose.yaml is as follows:

version: '3.9'
services:
  trino-coordinator:
        image: 'trinodb/trino:422'
        hostname: trino-coordinator
        ports:
              - '8080:8080'
        volumes:
              - ./etc:/etc/trino
        networks:
              - trino-network

  alluxio-leader:
        image: 'alluxio/alluxio:2.9.3'
        hostname: alluxio-leader 
        ports:
              - '19998:19998'
              - '19999:19999'
        environment:
              ALLUXIO_JAVA_OPTS: " \
               -Dalluxio.master.hostname=alluxio-leader \
               -Dalluxio.master.mount.table.root.ufs=s3://alluxio/ \
               -Dalluxio.underfs.s3.endpoint=http://minio:9000 \
               -Dalluxio.underfs.s3.disable.dns.buckets=true \
               -Dalluxio.underfs.s3.inherit.acl=false \
               -Daws.accessKeyId=minio \
               -Daws.secretKey=minio123 \
               -Dalluxio.security.authorization.permission.enabled=false" # This is only set for demonstration purposes
        command: master
        networks:
              - trino-network

  alluxio-follower:
        image: 'alluxio/alluxio:latest'
        hostname: alluxio-follower
        ports:
              - '29999:29999'
              - '30000:30000'
        shm_size: 1G
        environment:
              ALLUXIO_JAVA_OPTS: " \
               -Dalluxio.worker.ramdisk.size=1G \
               -Dalluxio.master.hostname=alluxio-leader \
               -Dalluxio.worker.hostname=alluxio-follower \
               -Dalluxio.underfs.s3.endpoint=http://minio:9000 \
               -Dalluxio.underfs.s3.disable.dns.buckets=true \
               -Dalluxio.underfs.s3.inherit.acl=false \
               -Daws.accessKeyId=minio \
               -Daws.secretKey=minio123 \
               -Dalluxio.security.authorization.permission.enabled=false" # This is only set for demonstration purposes
        command: worker
        networks:
              - trino-network

  mariadb:
        image: 'mariadb:latest'
        hostname: mariadb
        ports:
              - '3306:3306'
        environment:
              MYSQL_ROOT_PASSWORD: admin
              MYSQL_USER: admin
              MYSQL_PASSWORD: admin
              MYSQL_DATABASE: metastore_db
        networks:
              - trino-network

  hive-metastore:
        hostname: hive-metastore
        image: 'bitsondatadev/hive-metastore:latest'
        ports:
              - '9083:9083' # Metastore Thrift
        volumes:
              - ./conf/metastore-site.xml:/opt/apache-hive-metastore-3.0.0-bin/conf/metastore-site.xml:ro
        environment:
              METASTORE_DB_HOSTNAME: mariadb
        depends_on:
              - mariadb
        networks:
              - trino-network
        entrypoint: > 
              /bin/sh -c "wget https://repo1.maven.org/maven2/org/alluxio/alluxio-shaded-client/2.9.3/alluxio-shaded-client-2.9.3.jar -O /opt/apache-hive-metastore-3.0.0-bin/lib/alluxio-2.9.3-client.jar; /entrypoint.sh"


  minio:
        hostname: minio
        image: 'minio/minio:latest'
        container_name: minio
        ports:
              - '9000:9000'
              - '9001:9001'
        volumes:
              - minio-data:/data
        environment:
              MINIO_ACCESS_KEY: minio
              MINIO_SECRET_KEY: minio123
        command: server /data --console-address :9001
        networks:
              - trino-network

  minio-setup:
        depends_on:
              - minio
        image: minio/mc
        container_name: mc
        environment:
              - MINIO_ACCESS_KEY=minio
              - MINIO_SECRET_KEY=minio123
        networks:
              - trino-network
        volumes:
              - ./data:/tmp/data
        entrypoint: >
              /bin/sh -c "
              until (/usr/bin/mc config host add minio http://minio:9000 minio minio123) do echo '...waiting...' && sleep 1; done;
              /usr/bin/mc rm -r --force minio/alluxio;
              /usr/bin/mc mb minio/alluxio;
              /usr/bin/mc policy set public minio/alluxio;
              exit 0;
              "       

volumes:
  minio-data:
        driver: local

networks:
  trino-network:
        driver: bridge