Unstable container

I’ve used Pi Gallery 2 for a week or to, but my Docker-host crashed two times, and become unavailable. CPU load was stuck at 100%, and i couldn’t get in contact with the machine by SSH.

I rebooted it, and everything was good.

After two days, the same thing happened again.

This was my docker-compose file:

pigallery2:
  image: bpatrik/pigallery2:nightly-stretch
  container_name: pigallery2
  environment:
    - NODE_ENV=production
  volumes:
    - ./pigallery2/config:/app/data/config
    - /home/USERNAME/pigallery2/db:/app/data/db
    - /media/pictures:/app/data/images
    - ./pigallery2/tmp:/app/data/tmp
  ports:
    - "CUSTOMPORT:80"
  restart: unless-stopped

After that, i removed the container - and everything has been stable hereafter.

Am I the only one, with that issue?

I would like to dump more details and so, however, with all the crashed, i can’t get it back up, before i have a few days here within my Home-network, cause everything becomes unavailable, at the time, of the CPU is going to 100%.

I did setup some logging for the docker containers too, but at the time of the Pi Gallery 2 is starting to cook my CPU, the logging is stuck - and no useful information is logged.

My docker software is not the newest (Docker version 18.03.0-ce, build 0520e24), but EVERYTHING ELSE works just fine, with no issues, what so ever 😃

[bpatrik]

Hi,

thanks for following up. Unfortunately I could not reproduce this and no one else reported anything like this, so I do not really have a clue about this issue.

[Suika]

Limiting runtime resources is also an option https://docs.docker.com/config/containers/resource_constraints/