I have a problem where stopping a docker from the UnRaid UI forces a kill after 10 seconds on all processes, meaning the service doesn't finish doing what it should be doing:

 

*** Shutting down runit daemon (PID 12)...
*** Killing all processes...
2015-10-17 14:22:26.321 [iNFO ] [penhab.io.rest.RESTApplication] - Stopped REST API
*** Not all processes have exited in time. Forcing them to exit.

 

Any idea of the best way to approach this problem so the docker doesn't just kill all the processes, but waits for them to finish?

Thanks

C

No-body else has encountered this problem?

Where are you seeing this? In the log of a specific docker? Which docker?

I'm seeing it when I exec bash into the docker, then shut it down from the unraid UI.

It's my own docker:

https://github.com/christianwaite/dockerapp-openhab

 

Hasn't been causing problems, but I want to use a db4o database and that will corrupt it if it's not given enough time to shut down. There's a post about it here:

https://labs.ctl.io/gracefully-stopping-docker-containers/

 

But I don't know how unraid handles shutting down dockers.

Thanks

Christian

I'm seeing it when I exec bash into the docker, then shut it down from the unraid UI...

I think that's your problem. Your bash session is hanging the docker. You need to exit your bash session before shutting the docker down from unRAID.

Erm.. I think the whole reason I knew about this was that it was quitting before the application I'm running had a chance to remove the lock file, which was happening without me exec'd into bash. I'll test this tonight, but I'm pretty confident it's still doing it if I'm not in the session.

 

Thanks for the help

Christian