[SOLVED] Cache Drive Corrupted, Recoved and Remounted, Next Steps?

[XanXic]

I'm not super proficient with unraid/linux but I'm fairly computer literate. But I tried doing too much with everything and hardlocked unraid and had to do a hard shutdown. When it came back the cache drive said unmounted file not found or something. Anyways after googling a lot I came across the steps here 

 

I was able to mount it as readable, copied my appdata folder off to another disk because I desperately didn't want to have to rebuild all my docker apps. Restarted just to be safe, I got confused on the instructions and ran "btrfs check --repair /dev/sdX1" listed last on the instructions on my cache dir. After it started running it seemed odd, looked over instructions and apparently I wasn't supposed to run that? Let it go because, why not? Afterwards restarted the array to get to my disk with the app data I copied on it and the SSD cache remounted, all the files where there and my unraid was totally fine. My plex database was corrupted but I restored it from a DB backup on the cache. Lost two days of changes but overall is fine. It's been perfectly fine for the last few hours. I've been working on getting appdata backups setup so if this happens I can just reformat, remount and copy onto the cache again.

 

But for now I clearly didn't understand the --repair command or the instructions. My understanding was that it should've deleted all the files but my SSD seems totally fine. I've ran a SMART test and unraid says it's a-okay in the summary but if I'm being honest I don't totally understand how to read the results. By everything I can think to check it all seems good....idk if I need to do anything else or since I just did a repair if it can corrupt again or something? Since it's all working it'd be easy to copy appdata off, reformat and put it back but idk if that's over kill?

 

I honestly feel like I learned a lot through this process (Like encrypted cloud backups are useless when rclone is unaccessable) but I feel like I clumsily lucked through the process

Edited May 21, 2020 by XanXic

[JorgeB]

10 hours ago, XanXic said:

But for now I clearly didn't understand the --repair command or the instructions. My understanding was that it should've deleted all the files but my SSD seems totally fine.

No, that's not what it says in the FAQ, if says to be careful with --repair because some times it can make things even worse, if you had already backed up the data using one of the other methods it's OK to try --repair (or if none of the other methods worked) since you have nothing to lose, but there's also a chance that a repaired filesystem might run into issues again in the near future, depends on what the problem was, so make sure backups are up to date and if it gets corrupt again soon best to re-format.

[XanXic]

9 hours ago, johnnie.black said:

No, that's not what it says in the FAQ, if says to be careful with --repair because some times it can make things even worse, if you had already backed up the data using one of the other methods it's OK to try --repair (or if none of the other methods worked) since you have nothing to lose, but there's also a chance that a repaired filesystem might run into issues again in the near future, depends on what the problem was, so make sure backups are up to date and if it gets corrupt again soon best to re-format.

Alright, it was the destructive tag on the repair command that implied to me it would delete everything.

 

Already got unencrypted backups of appdata going! Thanks, I'll probably do a reformat soon just to be safe.

Edited May 21, 2020 by XanXic