I replaced a hdd with an ssd for my cache driver recently and everything is gone to heck since.  Today execution errors on fresh dockers with a new rebuild docker.img. 

 

Get stuff like this in docker logs level=error msg="Handler for POST /v1.37/containers/6e4ee040ad10/start returned error: error while creating mount source path '/mnt/user/appdata/plex': mkdir /mnt/user/appdata/plex: no space left on device"  There plently of space available thats not true....  I though it might be something with shares but i can't find it.  VM'd had issues too, rebuilt the libvirt a few times and work be ok for a while?  Seeing a lot of stuff in the syslogs that where never there before as well.  At this point I might just wipe the appdata, domains, and system shares, reformat the cache drive entirely and start again.  I only use the cache drive for a small handleful of dockers and vms, not as a staging point for data going to array.   I did or did not do something when i replaced the drive this time, but i don't know what.  I've replace several over the years never had a problem.  Must be and old cfg somewhere like on the flash causing all this but i can't find it.

p67-diagnostics-20211020-1110.zip syslog-192.168.5.17.log syslog-192.168.5.172.log

 

syslog.txt

 

 

 

EDIT:  I appears there were conflicting mounts points and the user/appdata share was reporting it was full to unraid.  Deleted the share and readded it.  If that doesn't work i guess i could trying mounting as a share /mnt/cache/appdata which is prob better anyway.

Edited October 20, 20214 yr by baracas
moreinfo

Deleting and recreating the share did fix it, but only because it reset the minimum free space.  You had the appdata share set to cache only with a value of 100000000 which is 100GB.  The diagnostics at the time showed you only had 89Gig free on the drive at the time, so the error message was correct.