Got through the parity check and no errors. Uptime 1 day 18 hours.

 

At this point I think it's safe to say it wasn't a hardware issue but a software one. Now to prevent it from occurring again.

 

I'm thinking a Docker caused the system to use too much RAM. I've read this is possible but I am not sure if there is anything I can do to limit a Docker's ability to do this. I read something about file calls from an app doing this. If I limit RAM in a Docker will this prevent that? 

 

These are the Dockers I've been running for a while, only CAdvisor has access to rootfs, I disabled it for now.

cadvisor

collabora

duckdns

duplicati

letsencrypt

mariadb

nextcloud

ombi

plex

radarr

sabnzbd

sonarr

tautulli

transmission - mostly turned off

zoneminder

 

If there are any that you think I should pay special attention to let me know. I've had them on auto update for the last few weeks but I think I'm going to turn that off for now.

 

I was digging through log files when the parity disk was showing missing and I guess rootfs was full and came across this line:

Mar 11 04:26:20 unraid6 rc.diskinfo[31541]: PHP Warning: file_put_contents(): Only 4096 of 7912 bytes written, possibly out of free disk space in /etc/rc.d/rc.diskinfo on line 266

 

Can someone shed some light on this? How severe is this?

Edited March 13, 20197 yr by ptirmal

I think I figured out what was going on. I believe when I upgraded to 6.6.7 either my unassigned device drive defaulted to not auto mount or it was never set as such and didn't mount. That drive is used for my security cameras via the Zoneminder docker. Since I was calling for /mnt/disks... and it wasn't mounted it was storing those files in rootfs. I noticed rootfs went from 25% to 50% after 14 or so hours and figured that out.

 

Hopefully that was it.