No way to detect full initialization of the database, healthcheck.sh always returns code 0

[olivierboudet]

Name and Version

bitnami/mariadb:10.11

What architecture are you using?

None

What steps will reproduce the bug?

1. Run mariadb with a lot of sql files which slow down initialization with a compose file like this :

  db:
    image: bitnami/mariadb:10.11
    environment:
      BITNAMI_DEBUG: true
      ALLOW_EMPTY_PASSWORD: "true"
      MARIADB_SKIP_TEST_DB: "yes"
      MARIADB_USER: root
      MARIADB_PASSWORD: password
      MARIADB_DATABASE: bitnami
      MARIADB_ROOT_PASSWORD: password
    ports:
      - 3306:3306
    healthcheck:
      test: [ "CMD", "/opt/bitnami/scripts/mariadb/healthcheck.sh", "--connect", "--innodb_initialized" ]
      start_period: 10s
      interval: 10s
      timeout: 5s
      retries: 10
    volumes:
      - ./sql/:/docker-entrypoint-initdb.d/

2. we expect the container to be ready only when initialization is completed. In the official mariadb image, initialization of the db is executed with the server running with --skip-networking flag. In this case, healthcheck.sh --connect fails and the container is not in a ready state.
   With the bitnami image, the --skip-networking flag is not used, which prevents to detect the full initialization of the database

What is the expected behavior?

healthcheck.sh --connect should allow to detect the full initialization of the database

What do you see instead?

There is no way to detect the real ready state of the database

Additional information

If this bug is valid and the proposal to use --skip-networking during setup is accepted, I am willing to contribute to this, please tell me.

[alvneiayu]

hi @olivierboudet

I think that I am reproducing the error. Could you send me logs reproducing the problem, please? Just to be aligned.

Thanks a lot

Álvaro