Some containers could stop faster

[heyyo-droid]

I noticed some containers are stopping after around 10.5 seconds like cron and adminer(maybe other).
10 seconds is the maximal time docker is waiting before stopping the container if unresponsive.

It looks like those containers don't receive SIGTERMs.

Adding init: true to each container fixed it for me:

crond_mfa:
    image: wodby/drupal-php:$PHP_TAG
    container_name: "${PROJECT_NAME}_crond"
    init: true
    environment:
      CRONTAB: "0 * * * * drush -r /var/www/html/web cron"
    command: sudo -E crond -f -d 0

https://docs.docker.com/compose/compose-file/compose-file-v2/#init

What do you think of adding it ?

[csandanov]

yes, we can add it for crond and adminer, it looks like crod doesn't handle it well when running in foreground
adminer is running on a built-in php server which I guess is very limited

[csandanov]

Adding it to adminer helped but crond throws the following error

[FATAL tini (1)] Unexpected error when forwarding signal: 'Operation not permitted'

but still works – container shuts down instantly

let's keep it like that for now

[heyyo-droid]

Right I didn't pay attention to this error.
Thanks