November 15, 20169 yr Community Expert Plex Requests wasn't launching from my browser at work, so I VPN'd to my home network and attempted to stop and then restart the Docker. Upon trying to restart it I got an error message in WebUI with a big red x saying that the execution had failed. Then I looked in the Syslog and it was complaining that the folder for the Docker was read-only. I tried restarting some other Dockers and got the same message in the UI (red x with execution error) and the Syslog (read-only error). At this point I decided to stop the Docker engine entirely, so I went to Settings > Docker and set Enabled to No. Then I set Enabled to Yes, but it showed that the Docker engine was Stopped, and the Docker tab was now missing from UI. Then I stopped the array and restarted the server. Upon reboot all of my array disks are there, but my cache disk shows as Unmountable. I've attached diagnostics from after the restart. What should I do? Is the drive toast? I'm happy to run SpinRite or similar on the SSD to see if I can resurrect it (I think SpinRite Level 2 is permissable for SSD use, I'll check before doing anything though). I'm fairly certain that I haven't been backing up my appdata folder. No big deal, it'll just be a pain at first to recreate everything tower-diagnostics-20161115-1040.zip
November 15, 20169 yr Community Expert Cache filesystem needs repairing, see wiki: https://lime-technology.com/wiki/index.php/Check_Disk_Filesystems#Drives_formatted_with_BTRFS Good luck, If unsuccessful you may need to format it and start over.
November 15, 20169 yr Author Community Expert I'm at a loss on how to proceed. I think the filesystem is toast unless someone has any other ideas: root@Tower:/mnt# btrfs restore /dev/sdb /mnt/disk6/restore/ No valid Btrfs found on /dev/sdb Could not open root, trying backup super No valid Btrfs found on /dev/sdb Could not open root, trying backup super Superblock bytenr is larger than device size Could not open root, trying backup super On second thought, redoing all my Dockers is going to be a pain. Getting SickRage to scrape my TV share isn't a big deal, but re-setting up the search providers, nzbToMedia, CouchPotatoe, SABnzbd is going to take a whole night, not to mention Plex etc. My fault for not running in a cache pool I guess. Is BTRFS the recommended FS for cache drives, or should I use a different filesystem?
November 15, 20169 yr Community Expert you need to specify the partition, eg: btrfs restore /dev/sdb1 /mnt/disk6/restore/
November 15, 20169 yr Community Expert Also the destination folder has to exist, so create restore on disk6 before hand.
November 15, 20169 yr Author Community Expert you need to specify the partition, eg: btrfs restore /dev/sdb1 /mnt/disk6/restore/ Worked! Thank you so much! Should I leave the cache drive as BTRFS or format as something else? And is a cache pool preferred?
November 15, 20169 yr Community Expert I would use XFS for a single device, if you plan to use a pool in the future you need BTRFS. A pool offers protection in case of a device failing, in a case like this it may help, because you could scrub it, but if the filesystem corruption was mirrored it won't help.
Archived
This topic is now archived and is closed to further replies.