ldog88 Posted July 13, 2021 Share Posted July 13, 2021 (edited) Hi, I've been trying to understand the below post but am new to unraid/linux so am struggling to do so. Please can someone confirm what the exact terminal line entries I should be tryping to recover a unmountable cache pool as seen in the attached screenshot. To clarify, I was trying to seperate the two cache drives into seperate pools when they were prevously one combined pool. Edit: I have worked out the correct terminal entry for the BTRFS filecheck however am now getting the following error: Starting repair. Opening filesystem to check... No valid Btrfs found on /dev/nvme1n1 ERROR: cannot open file system Thanks Edited July 13, 2021 by ldog88 Quote Link to comment
JorgeB Posted July 13, 2021 Share Posted July 13, 2021 Note, split from FAQ thread, @ldog88, please don't post questions in that thread, it's reserved for FAQ entries, also please post the diags (tools -> diagnostics) after array start to see current pool problem. Quote Link to comment
ldog88 Posted July 13, 2021 Author Share Posted July 13, 2021 Hi, sorry I'm new to unraid and the forum protocol! I have attached the diagnostics to this post. Please let me know if there is anything else I should do. Thanks lawson-nas-diagnostics-20210713-1724.zip Quote Link to comment
JorgeB Posted July 13, 2021 Share Posted July 13, 2021 Try this: stop array, unassign both cache device from both pools, start array, then on the console type: btrfs-select-super -s 1 /dev/nvme0n1p1 Stop array, assign both devices to the SAME pool, start array, post new diags. Quote Link to comment
ldog88 Posted July 13, 2021 Author Share Posted July 13, 2021 Great, thank you for your help. It looks as though my dockers have all been restored now along with the cache funcitoning as normal. I have attached the logs. To clarify my intentions, I wish to seperate out the SSDs into seperate cache pools, one for docker images and one for downloads etc. What is the easiest way to do this? Thanks lawson-nas-diagnostics-20210713-1923.zip Quote Link to comment
JorgeB Posted July 13, 2021 Share Posted July 13, 2021 Pool is using single profile, i.e., it's not redundant, hence the problem when you separated them, first convert pool to raid1, then stop array, remove one of the devices (but leave it unassigned for now), start array to finish pool balance, once that's done you can create the other pool. Quote Link to comment
ldog88 Posted July 14, 2021 Author Share Posted July 14, 2021 So I have followed the steps you said above and it appears to have worked but there have been some errors on the log. Are you able to confirm if this has been executed as expected? Thanks lawson-nas-diagnostics-20210714-0903.zip Quote Link to comment
JorgeB Posted July 14, 2021 Share Posted July 14, 2021 Jul 14 01:17:21 Lawson-NAS kernel: BTRFS warning (device nvme1n1p1): csum failed root -9 ino 352 off 17227776 csum 0xde765087 expected csum 0xde765187 mirror 1 Jul 14 01:17:21 Lawson-NAS kernel: BTRFS error (device nvme1n1p1): bdev /dev/nvme1n1p1 errs: wr 0, rd 0, flush 0, corrupt 19, gen 0 This means data corruption was detected and the balance aborted, you can run a scrub to identify the corrupt files then delete them or restore from backups, after that repeat the procedure, not a bad idea to run a memtest before doing anything else to see if there are any RAM issues. Quote Link to comment
ldog88 Posted July 14, 2021 Author Share Posted July 14, 2021 I've just run the scrub and got the attached error. To clarify, the system appears to be working as expected other than the errors in the log. lawson-nas-diagnostics-20210714-1141.zip Quote Link to comment
JorgeB Posted July 14, 2021 Share Posted July 14, 2021 23 minutes ago, ldog88 said: To clarify, the system appears to be working as expected other than the errors in the log. Yes, but the pool didn't finish removing the device, it will try again next array start, you need to deleted/replace this file: path: appdata/Plex-Media-Server/Library/Application Support/Plex Media Server/Metadata/Movies/f/a9b85753f0f186578d14c908768c660a90cae21.bundle/Contents/_combined/posters/tv.plex.agents.movie_683eb90fe7ad3714a9a908e39011f9b73fde8f14 Quote Link to comment
ldog88 Posted July 14, 2021 Author Share Posted July 14, 2021 I have deleted this file however unfortunately this donesn't look to have fixed things. Thanks lawson-nas-diagnostics-20210714-1453.zip Quote Link to comment
JorgeB Posted July 14, 2021 Share Posted July 14, 2021 Balanced finished now, just didn't delete the missing device since it wasn't done automatically at array start, type this: btrfs device delete missing /mnt/cache Then post new diags to confirm it's done. Quote Link to comment
ldog88 Posted July 14, 2021 Author Share Posted July 14, 2021 Just done so hopefully all good now. Just to check, when re-adding the other nvme, I can just create a seperate pool and put it in there? Will it need to be formatted etc? Thanks lawson-nas-diagnostics-20210714-1521.zip Quote Link to comment
JorgeB Posted July 14, 2021 Share Posted July 14, 2021 Yes, now is OK, other one will need to be formatted before use. Quote Link to comment
ldog88 Posted July 14, 2021 Author Share Posted July 14, 2021 Perfect. Thank you so much for your help. 1 Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.