ibis Documentation

docker run -d --name ibis -p 8123:8123 --restart always ibis-postgres

services:
  clickhouse:
    image: clickhouse/clickhouse-server:26.1.2.11-alpine
    ports:
      - 8123:8123 # http port
      - 9000:9000 # native protocol port
    environment:
      CLICKHOUSE_USER: ibis
      CLICKHOUSE_PASSWORD: ""
    healthcheck:
      interval: 1s
      retries: 10
      test:
        - CMD-SHELL
        - wget -qO- 'http://ibis@clickhouse:8123/?query=SELECT%201' # SELECT 1
    volumes:
      - clickhouse:/var/lib/clickhouse/user_files/ibis
    networks:
      - clickhouse

  mysql:
    environment:
      MYSQL_ALLOW_EMPTY_PASSWORD: "true"
      MYSQL_DATABASE: ibis-testing
      MYSQL_PASSWORD: ibis
      MYSQL_USER: ibis
    healthcheck:
      interval: 1s
      retries: 20
      test:
        - CMD
        - mariadb-admin
        - ping
    image: mariadb:12.3.3
    ports:
      - 3306:3306
    networks:
      - mysql
    volumes:
      - mysql:/data
      - $PWD/docker/mysql:/docker-entrypoint-initdb.d:ro

  singlestoredb:
    environment:
      ROOT_PASSWORD: ibis_testing
      SINGLESTORE_LICENSE: "" # Optional license key
      SINGLESTOREDB_DATABASE: ibis_testing
      SINGLESTOREDB_PASSWORD: ibis_testing
      SINGLESTOREDB_USER: root
      SINGLESTOREDB_PORT: 3307
    healthcheck:
      interval: 1s
      retries: 20
      test:
        - CMD-SHELL
        - sdb-admin query --host 127.0.0.1 --user root --password ibis_testing --port 3306 --sql 'select 1'
    image: ghcr.io/singlestore-labs/singlestoredb-dev:0.2.86
    ports:
      - 3307:3306 # Use 3307 to avoid conflict with MySQL
      - 9089:9000 # Data API (use 9089 to avoid conflicts)
      # - 9088:8080 # SingleStore Studio UI (use 9088 to avoid conflicts)
    networks:
      - singlestoredb
    volumes:
      - singlestoredb:/data
      - $PWD/docker/singlestoredb/init.sql:/init.sql:ro

  postgres:
    environment:
      POSTGRES_PASSWORD: postgres
      POSTGRES_DB: ibis_testing
      POSTGRES_USER: postgres
    build: ./docker/postgres
    image: ibis-postgres
    healthcheck:
      interval: 1s
      retries: 20
      test:
        - CMD
        - pg_isready
    ports:
      - 5432:5432
    networks:
      - postgres
    volumes:
      - postgres:/data

  mssql:
    image: mcr.microsoft.com/mssql/server:2022-latest
    environment:
      MSSQL_SA_PASSWORD: 1bis_Testing!
      ACCEPT_EULA: "Y"
      # The default collation in MSSQL is SQL_Latin1_General_CP1_CI_AS
      # where the CI stands for Case Insensitive.
      # We use a case-sensitive collation for testing so that we don't
      # break users with case-sensitive collations.
      MSSQL_COLLATION: Latin1_General_100_BIN2_UTF8
    healthcheck:
      interval: 1s
      retries: 20
      test:
        - CMD-SHELL
        - $(find /opt -name sqlcmd -type f -executable) -C -S localhost -U sa -P "$$MSSQL_SA_PASSWORD" -Q "SELECT 1"
    ports:
      - 1433:1433
    volumes:
      - mssql:/data
    networks:
      - mssql

  hive-metastore-db:
    image: postgres:18.6-alpine
    environment:
      POSTGRES_USER: admin
      POSTGRES_PASSWORD: admin
      POSTGRES_DB: metastore
    healthcheck:
      interval: 1s
      retries: 20
      test:
        - CMD
        - pg_isready
        - --port=23456
    command: -c port=23456
    networks:
      - trino

  minio:
    image: bitnamilegacy/minio:2025.7.23
    environment:
      MINIO_ROOT_USER: accesskey
      MINIO_ROOT_PASSWORD: secretkey
      MINIO_SKIP_CLIENT: "yes"
    healthcheck:
      interval: 1s
      retries: 20
      test:
        - CMD-SHELL
        - mc ready data && mc mb --ignore-existing data/trino data/risingwave
    networks:
      - trino
      - risingwave
    volumes:
      - $PWD/docker/minio/config.json:/.mc/config.json:ro

  hive-metastore:
    # TODO: healthcheck?
    image: starburstdata/hive:3.1.3-e.4
    environment:
      HIVE_METASTORE_DRIVER: org.postgresql.Driver
      HIVE_METASTORE_JDBC_URL: jdbc:postgresql://hive-metastore-db:23456/metastore
      HIVE_METASTORE_USER: admin
      HIVE_METASTORE_PASSWORD: admin
      HIVE_METASTORE_WAREHOUSE_DIR: s3://trino/
      HIVE_METASTORE_USERS_IN_ADMIN_ROLE: "admin"
      S3_ENDPOINT: http://minio:9000
      S3_ACCESS_KEY: accesskey
      S3_SECRET_KEY: secretkey
      S3_PATH_STYLE_ACCESS: "true"
      REGION: ""
      GOOGLE_CLOUD_KEY_FILE_PATH: ""
      AZURE_ADL_CLIENT_ID: ""
      AZURE_ADL_CREDENTIAL: ""
      AZURE_ADL_REFRESH_URL: ""
      AZURE_ABFS_STORAGE_ACCOUNT: ""
      AZURE_ABFS_ACCESS_KEY: ""
      AZURE_WASB_STORAGE_ACCOUNT: ""
      AZURE_ABFS_OAUTH: ""
      AZURE_ABFS_OAUTH_TOKEN_PROVIDER: ""
      AZURE_ABFS_OAUTH_CLIENT_ID: ""
      AZURE_ABFS_OAUTH_SECRET: ""
      AZURE_ABFS_OAUTH_ENDPOINT: ""
      AZURE_WASB_ACCESS_KEY: ""
    depends_on:
      hive-metastore-db:
        condition: service_healthy
      minio:
        condition: service_healthy
    networks:
      - trino

  trino:
    depends_on:
      - hive-metastore
    healthcheck:
      interval: 2s
      retries: 15
      test:
        - CMD-SHELL
        - trino --output-format null --execute 'show schemas in hive; show schemas in memory'
    image: trinodb/trino:479
    ports:
      - 8080:8080
    networks:
      - trino
    volumes:
      - $PWD/docker/trino/catalog/memory.properties:/etc/trino/catalog/memory.properties:ro
      - $PWD/docker/trino/catalog/hive.properties:/etc/trino/catalog/hive.properties:ro
      - $PWD/docker/trino/jvm.config:/etc/trino/jvm.config:ro

  druid-postgres:
    image: postgres:18.6-alpine
    container_name: druid-postgres
    environment:
      POSTGRES_PASSWORD: FoolishPassword
      POSTGRES_USER: druid
      POSTGRES_DB: druid
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - pg_isready
    networks:
      - druid

  # Need 3.5 or later for container nodes
  druid-zookeeper:
    hostname: zookeeper
    container_name: zookeeper
    image: zookeeper:3.9
    environment:
      ZOO_MY_ID: 1
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - nc -z 127.0.0.1 2181
    networks:
      - druid

  druid-coordinator:
    image: apache/druid:38.0.0
    hostname: coordinator
    container_name: coordinator
    volumes:
      - druid:/opt/shared
      - druid:/opt/druid/var
    depends_on:
      druid-zookeeper:
        condition: service_healthy
      druid-postgres:
        condition: service_healthy
    command:
      - coordinator
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - nc -z 127.0.0.1 8081
    env_file:
      - ./docker/druid/environment
    networks:
      - druid

  druid-broker:
    image: apache/druid:38.0.0
    hostname: broker
    container_name: broker
    volumes:
      - druid:/opt/druid/var
    depends_on:
      druid-zookeeper:
        condition: service_healthy
      druid-postgres:
        condition: service_healthy
      druid-coordinator:
        condition: service_healthy
    command:
      - broker
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - nc -z 127.0.0.1 8082
    ports:
      - 8082:8082
    env_file:
      - ./docker/druid/environment
    networks:
      - druid

  druid-historical:
    image: apache/druid:38.0.0
    hostname: historical
    container_name: historical
    volumes:
      - druid:/opt/shared
      - druid:/opt/druid/var
    depends_on:
      druid-zookeeper:
        condition: service_healthy
      druid-postgres:
        condition: service_healthy
      druid-coordinator:
        condition: service_healthy
    command:
      - historical
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - nc -z 127.0.0.1 8083
    env_file:
      - ./docker/druid/environment
    networks:
      - druid

  druid-middlemanager:
    image: apache/druid:38.0.0
    hostname: middlemanager
    container_name: middlemanager
    volumes:
      - druid:/data
      - druid:/opt/shared
      - druid:/opt/druid/var
    depends_on:
      druid-zookeeper:
        condition: service_healthy
      druid-postgres:
        condition: service_healthy
      druid-coordinator:
        condition: service_healthy
    command:
      - middleManager
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - nc -z 127.0.0.1 8091
    env_file:
      - ./docker/druid/environment
    networks:
      - druid

  druid:
    image: apache/druid:38.0.0
    hostname: router
    container_name: router
    volumes:
      - druid:/opt/druid/var
    depends_on:
      druid-zookeeper:
        condition: service_healthy
      druid-postgres:
        condition: service_healthy
      druid-coordinator:
        condition: service_healthy
      druid-middlemanager:
        condition: service_healthy
      druid-historical:
        condition: service_healthy
      druid-broker:
        condition: service_healthy
    ports:
      - 8888:8888
    command:
      - router
    healthcheck:
      interval: 2s
      retries: 30
      test:
        - CMD-SHELL
        - nc -z 127.0.0.1 8888
    env_file:
      - ./docker/druid/environment
    networks:
      - druid

  oracle:
    image: gvenzl/oracle-free:23.9-slim
    environment:
      ORACLE_PASSWORD: ibis
      ORACLE_DATABASE: IBIS_TESTING
      APP_USER: ibis
      APP_USER_PASSWORD: ibis
    ports:
      - 1521:1521
    healthcheck:
      interval: 2s
      retries: 25
      test:
        - CMD-SHELL
        - ./healthcheck.sh
    restart: on-failure
    networks:
      - oracle
    volumes:
      - oracle:/opt/oracle/data

  exasol:
    image: exasol/docker-db:2025.2.1
    privileged: true
    ports:
      - 8563:8563
    healthcheck:
      interval: 10s
      retries: 9
      timeout: 90s
      test:
        - CMD-SHELL
        - /opt/exasol/db-*/bin/Console/exaplus -q -x -c 127.0.0.1:8563 -u sys -p exasol -jdbcparam validateservercertificate=0 -sql 'SELECT 1;'
    networks:
      - exasol
    volumes:
      - exasol:/data

  flink-jobmanager:
    image: flink:${FLINK_VERSION}
    environment:
      FLINK_PROPERTIES: |
        jobmanager.rpc.address: flink-jobmanager
    ports:
      - 8081:8081
    command: jobmanager
    networks:
      - flink

  flink:
    image: ibis-flink-${FLINK_VERSION}
    build:
      context: ./docker/flink
      args:
        FLINK_VERSION: ${FLINK_VERSION}
    environment:
      FLINK_PROPERTIES: |
        jobmanager.rpc.address: flink-jobmanager
        taskmanager.numberOfTaskSlots: 2
        taskmanager.memory.process.size: 2048m
        taskmanager.memory.network.fraction: 0.4
        taskmanager.memory.network.min: 512mb
        taskmanager.memory.network.max: 2gb
    depends_on:
      - flink-jobmanager
    command: taskmanager
    networks:
      - flink

  kudu:
    cap_add:
      - SYS_TIME
    image: apache/kudu:1.17.0
    networks:
      - impala
    command: kudu master run --fs_wal_dir=/var/lib/kudu/master --fs_data_dirs=/var/lib/kudu/master
    healthcheck:
      interval: 1s
      retries: 60
      test:
        - CMD-SHELL
        - kudu cluster ksck kudu:7051

  kudu-tserver:
    cap_add:
      - SYS_TIME
    image: apache/kudu:1.17.0
    depends_on:
      kudu:
        condition: service_healthy
    networks:
      - impala
    command: kudu tserver run --fs_wal_dir=/var/lib/kudu/master --fs_data_dirs=/var/lib/kudu/master --tserver_master_addrs=kudu
    healthcheck:
      interval: 1s
      retries: 60
      test:
        - CMD-SHELL
        - kudu cluster ksck kudu:7051

  impala-hive-metastore:
    image: apache/impala:4.5.3-impala_quickstart_hms
    container_name: impala-hive-metastore
    command: hms
    volumes:
      # Volume used to store Apache Derby database.
      - impala:/var/lib/hive
      # Warehouse directory. HMS does file operations so needs access to the
      # shared volume.
      - impala:/user/hive/warehouse
      - ./docker/impala/conf:/opt/hive/conf:ro
    healthcheck:
      interval: 3s
      retries: 20
      test:
        - CMD-SHELL
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/9083; exit $$?;'
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/impala-hive-metastore/9083; exit $$?;'
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/impala-hive-metastore.impala/9083; exit $$?;'
    networks:
      impala:
        aliases:
          - impala-hive-metastore
          - impala-hive-metastore.impala

  statestored:
    image: apache/impala:4.5.3-statestored
    ports:
      - 25010:25010 # Web debug UI
    command:
      - -redirect_stdout_stderr=false
      - -logtostderr
      - -v=1
    volumes:
      - ./docker/impala/conf:/opt/impala/conf:ro
    healthcheck:
      interval: 3s
      retries: 20
      test:
        - CMD-SHELL
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/25010; exit $$?;'
    networks:
      - impala

  catalogd:
    image: apache/impala:4.5.3-catalogd
    depends_on:
      impala-hive-metastore:
        condition: service_started
      statestored:
        condition: service_healthy
    ports:
      - 25020:25020 # Web debug UI
    command:
      - -redirect_stdout_stderr=false
      - -logtostderr
      - -v=1
      - -hms_event_polling_interval_s=1
      - -invalidate_tables_timeout_s=999999
    volumes:
      # Warehouse directory. Catalog does file operations so needs access to the
      # shared volume.
      - impala:/user/hive/warehouse
      - ./docker/impala/conf:/opt/impala/conf:ro
    healthcheck:
      interval: 3s
      retries: 20
      test:
        - CMD-SHELL
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/25020; exit $$?;'
    networks:
      - impala

  impala:
    image: apache/impala:4.5.3-impalad_coord_exec
    depends_on:
      statestored:
        condition: service_healthy
      catalogd:
        condition: service_healthy
      kudu:
        condition: service_healthy
      kudu-tserver:
        condition: service_healthy
    ports:
      - 21050:21050 # HS2 endpoint
    healthcheck:
      interval: 3s
      retries: 20
      test:
        - CMD-SHELL
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/21050; exit $$?;'
    command:
      - -v=1
      - -redirect_stdout_stderr=false
      - -logtostderr
      - -kudu_master_hosts=kudu:7051
      - -mt_dop_auto_fallback=true
      - -default_query_options=mt_dop=4,default_file_format=parquet,default_transactional_type=insert_only
      - -mem_limit=4gb
    environment:
      # Keep the Java heap small to preserve memory for query execution.
      JAVA_TOOL_OPTIONS: -Xmx1g
    volumes:
      - impala:/user/hive/warehouse
      - ./docker/impala/conf:/opt/impala/conf:ro
    networks:
      - impala

  materialize:
    image: materialize/materialized:v26.10.2
    ports:
      - 6875:6875 # SQL port
      - 6876:6876 # HTTP port
      - 6874:6874 # Internal port
    environment:
      MZ_LOG: info
      MZ_TELEMETRY: "false"
      MZ_SYSTEM_PARAMETER_DEFAULT: "max_tables=1000"
    healthcheck:
      test: ["CMD", "curl", "-f", "localhost:6878/api/readyz"]
      interval: 1s
      start_period: 60s
    volumes:
      - materialize:/data
    networks:
      - materialize

  risingwave:
    image: ghcr.io/risingwavelabs/risingwave:v2.1.3
    command: "standalone --meta-opts=\" \
      --advertise-addr 0.0.0.0:5690 \
      --backend mem \
      --state-store hummock+minio://accesskey:secretkey@minio:9000/risingwave \
      --data-directory hummock_001\" \
      --compute-opts=\"--advertise-addr 0.0.0.0:5688 --role both\" \
      --frontend-opts=\"--listen-addr 0.0.0.0:4566 --advertise-addr 0.0.0.0:4566\" \
      --compactor-opts=\"--advertise-addr 0.0.0.0:6660\""
    ports:
      - 4566:4566
    depends_on:
      minio:
        condition: service_healthy
    volumes:
      - risingwave:/data
    environment:
      RUST_BACKTRACE: "1"
      ENABLE_TELEMETRY: "false"
    healthcheck:
      test:
        - CMD-SHELL
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/6660; exit $$?;'
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/5688; exit $$?;'
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/4566; exit $$?;'
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/5690; exit $$?;'
      interval: 1s
      retries: 20
    restart: on-failure
    networks:
      - risingwave

  spark-connect:
    image: bitnamilegacy/spark:4.0.0
    ports:
      - 15002:15002
    command:
      - /opt/bitnami/spark/sbin/start-connect-server.sh
      - --name
      - ibis_testing
      - --packages
      - org.apache.spark:spark-connect_2.13:4.0.0,org.apache.iceberg:iceberg-spark-runtime-4.0_2.13:1.10.1,io.delta:delta-spark_2.13:4.0.0
    healthcheck:
      test:
        - CMD-SHELL
        - bash -c 'printf \"GET / HTTP/1.1\n\n\" > /dev/tcp/127.0.0.1/15002; exit $$?;'
      interval: 5s
      retries: 6
    volumes:
      - spark-connect:/data
      - $PWD/docker/spark-connect/conf.properties:/opt/bitnami/spark/conf/spark-defaults.conf:ro
      - $PWD/docker/spark-connect/log4j2.properties:/opt/bitnami/spark/conf/log4j2.properties:ro
    networks:
      - spark-connect

networks:
  impala:
    # docker defaults to naming networks "$PROJECT_$NETWORK" but the Java Hive
    # Metastore clients don't accept underscores in the thrift URIs and
    # something is too-aggressively supplanting the specified thrift metastore
    # URI with $SPECIFIED_URI.$NETWORK so rename it to something acceptable
    name: "impala"
  mysql:
  mssql:
  clickhouse:
  postgres:
  singlestoredb:
  trino:
  druid:
  oracle:
  exasol:
  flink:
  materialize:
  risingwave:
  spark-connect:

volumes:
  clickhouse:
  druid:
  mssql:
  mysql:
  oracle:
  postgres:
  singlestoredb:
  exasol:
  impala:
  materialize:
  risingwave:
  spark-connect: