Uncontrolled shutdown of the PostgreSQL container

Community Containers
21

Finally I did some investigation. The problem is within the shutdown of the pg-gvm container.

When docker itself is restarted or shutdown, for example during a reboot, the containers don’t get that much time as during normal stop. In case of the pg-gvm container this results in an insufficient shutdown of the postgres database which leaves a dangling lock file in the file system.

22

This should fix the issue

6 Likes
23

The same…. I run docker-compose down, docker rmi registry.community.greenbone.net/community/pg-gvm , docker rmi registry.community.greenbone.net/community/pg-gvm-migrator:stable, docker volume rm greenbone-community-edition_psql_socket_vol. What I’m doing wrong? Or I need reinstall all on new VM ? :slight_smile:

24

No

  1. shut down all containers
  2. remove the postgres socket volume
  3. pull the newest images
  4. start the containers

This should work.

1 Like
25

delete greenbone-community-edition_psql_socket_vol/.s.PGSQL.5432.lock and restart.

use “docker-compose -f $DOWNLOAD_DIR/docker-compose.yml stop” before shudown the docker.