docker-compose up doesn't exit when containers stopped

Question

I'm running some services through docker-compose, and launching them with systemd:

$ systemctl status nextcloud.service
● nextcloud.service - HomelabOS nextcloud Service
     Loaded: loaded (/etc/systemd/system/nextcloud.service; enabled; vendor preset: disabled)
     Active: deactivating (stop-sigterm) since Wed 2021-12-08 23:49:07 EST; 5min ago
    Process: 402260 ExecStop=/usr/bin/docker-compose -f /var/homelabos/docker-compose.nextcloud.yml -p nextcloud stop (code=exited, status=0/SUCCES>
   Main PID: 3128 (docker-compose)
      Tasks: 22 (limit: 38376)
     Memory: 12.8M
        CPU: 8.704s
     CGroup: /system.slice/nextcloud.service
             └─3128 /usr/bin/docker-compose -f /var/homelabos/docker-compose.nextcloud.yml -p nextcloud up --build
...

As you can see, the containers have been killed successfully, and other commands say the same thing:

$ sudo /usr/bin/docker-compose -f /var/homelabos/docker-compose.nextcloud.yml -p nextcloud ps
NAME                       COMMAND                  SERVICE             STATUS              PORTS
nextcloud-nextcloud-1      "/entrypoint.sh /usr…"   nextcloud           exited (137)        
nextcloud-nextcloud_db-1   "docker-entrypoint.s…"   nextcloud_db        exited (0)  
$ sudo docker ps
CONTAINER ID   IMAGE             COMMAND     CREATED         STATUS         PORTS      NAMES
363f0b6c5e24   grafana/grafana   "/run.sh"   2 minutes ago   Up 2 minutes   3000/tcp   grafana-grafana-1

The problem here is that docker-compose up is still running! How do I get it to stop doing that, and instead stop when the containers it manages are stopped, i.e. the original behavior?

Answer 1

This is a bug in docker compose v2.2.0. Downgrading to v2.1.1 should fix the issue.

It now also appears to be fixed in v2.2.3.