How to Fix XFS File System Corruption on an Unraid Server 

It was chilling how you seem to know every single one of the errors i was running into. Thank you so much for this video.

Just came here to say this video saved me. Had 2 drives randomly show up as XFS Unmountable right after I had to replace my parity drive so I was unable to rebuild either drive. After following your video I was able to repair the file system and both drives mounted with all my data intact. Thank you so much!

Whew! I was foolishly trying to find which drive had failed by yanking SATA cables to seeing which drives would go offline. They never did until I restarted the array. Now I had to rebuild 2 drives now, (thank goodness I had 2 parity drives!) and use this video. But that -L command worked like a charm. All my files are back and didn't even have to touch my backup.

OMG. Thank you soo much. You're a life saver. I'd accepted the loss of my data and was trying anything that might work and this did it! You rock!

More or less, due to your channel i tried Unraid, and i've used it for over 6 months now. But his particular video saved 6tb of data for me. Thank you

Great video. Just what I needed as i actually had corruption present. Been bugging me for a while, I thought it would take ages to repair. Fixed now, thanks spaceinvader!

Thanks SIO, you gave me the courage to do a "-L" when forums and documentation didn't make it clear what I was about to lose.

your my hero, have a limited understanding of unraid and have spent the last 2 years watching your videos but today you saved me

You saved my bacon!!!! array is back up again. I have massive appreciation for you for making these video's! Thank you so much and keep making these awesome video's

still works today and saved my butt. Server was locked up every other day and this solved the issue!

Hey Spaceinvader! You have made my unRAID experience absolutely amazing. Have you ever or would you ever do a guide on setting up a PXE server for network booting?

Spaceinvader, you're a legend this helped me get my NAS back up and running again. Excellent video :)

Again you saved my "life" / "data". Thank you very much for your dedication.

*WHEW* Keep 'em coming SpaceInvader One!

I say this every video, Supeb! Great work Ed!

If we don't have a monitor hooked to unraid on boot is there another location where we can see this a warning showing xfs corruption or will unRAID notify us?

Today, you saved my bacon! Thank you for this video!!!!

Lifesaver....just the info I needed. Keep up the great videos

Thanks for another awesome video. I’m looking to start my server over and wanted to know if you had any experience with BTRFS for your array. I’m curious how it works in the UnRAID format so being individual disks rather than a RAID array. I’m interested in the extra protection against errors and corruption but not sure if there are real drawbacks or extra maintenance needed. Also I remember in your PFSense video you mentioned a script to wake the physical box if your server shut down and was wondering if the reverse is possible. I plan on running a PFSense box with an old Dell Optiplex machine and wanted a backup VM that would act as a failover.

Another great video Ed! Thanks for always sharing them with us.👌😎JP

Thanks, helped me out with my cache drive!

THANK YOU! For some weird reason I could not access my unraid server I force closed and the drive with most important data was unmountable. This saved me! Also, I have a question. I have setup share in a way that all of its data is saved in Disk 3. My question is if Disk 3 fails do I lose all of my data?

Another Excellent, detailed, technical guide! Good to know general diagnostic procedures. I'm using btrfs format for cache. I'm wondering how much difference there are between XFS & BTRFS repair?

Of course I would find this video three weeks after my own XFS repair.

Thank you, saved me a massive headache. 😊

Thank you kind sir. This saved me today.

Overkill to ask/post this and I think I look good as running with -v generated no lost+found dir. But just for the sake of discussion and my curiosity: Couple of different lines in mine on Unraid 6.8.3 on Drive 1, a 2TB drive: - rebuild AG headers and trees - 32 of 32 allocation groups done - check for inodes claiming duplicate blocks - 1316416 of 1314720 inodes done (This is the only strange line...why not 1314720 of 1314720?) - verify and correct link counts - 32 of 32 allocation groups done ---The lines above are only on my 2TB data drive as the next 3 drives are smaller. Is this because of the size of the drive I presume?--- Here's the EDITED entire output with only the -v set in the webgui. I don't see any issues, but not my forte. Phase 1 - find and verify superblock... - reporting progress in intervals of 15 minutes - block cache size set to 367160 entries Phase 2 - using internal log - zero log... zero_log: head block 75338 tail block 75338 - scan filesystem freespace and inode maps... - 15:11:10: scanning filesystem freespace - 32 of 32 allocation groups done - found root inode chunk Phase 3 - for each AG... - scan and clear agi unlinked lists... - 15:11:10: scanning agi unlinked lists - 32 of 32 allocation groups done - process known inodes and perform inode discovery... - agno = 0 ( my comment : 32 of these in a different order, but there are 32, 0-31) - 15:14:38: process known inodes and inode discovery - 1316416 of 1314720 inodes done - process newly discovered inodes... - 15:14:38: process newly discovered inodes - 32 of 32 allocation groups done Phase 4 - check for duplicate blocks... - setting up duplicate extent list... - 15:14:38: setting up duplicate extent list - 32 of 32 allocation groups done - check for inodes claiming duplicate blocks... - agno = 0 # (my comment) -(goes to 31, so 0-31=32 allocation groups I guess) - 15:14:39: check for inodes claiming duplicate blocks - 1316416 of 1314720 inodes done Phase 5 - rebuild AG headers and trees... - agno = 0 ... (my edit) - agno = 31 - 15:14:40: rebuild AG headers and trees - 32 of 32 allocation groups done - reset superblock... Phase 6 - check inode connectivity... - resetting contents of realtime bitmap and summary inodes - traversing filesystem ... - agno = 0 ...... - agno = 31 - traversal finished ... - moving disconnected inodes to lost+found ... Phase 7 - verify and correct link counts... - 15:16:22: verify and correct link counts - 32 of 32 allocation groups done XFS_REPAIR Summary Fri Oct 23 15:16:28 2020 Phase Start End Duration Phase 1: 10/23 15:11:04 10/23 15:11:04 Phase 2: 10/23 15:11:04 10/23 15:11:10 6 seconds Phase 3: 10/23 15:11:10 10/23 15:14:38 3 minutes, 28 seconds Phase 4: 10/23 15:14:38 10/23 15:14:39 1 second Phase 5: 10/23 15:14:39 10/23 15:14:40 1 second Phase 6: 10/23 15:14:40 10/23 15:16:22 1 minute, 42 seconds Phase 7: 10/23 15:16:22 10/23 15:16:22 Total run time: 5 minutes, 18 seconds done --end-- Had a power blink and as I decommissioned my old server, it wasn't plugged in to the battery side of the UPS, just the surge side. Doh. So I checked the filesystem.

Great video for me a a Linux noob 😁. Saved for future disk problems! 👊

Thank you. Another helpful video.

Hi Spaceinvader One You are amazing Thank You for all of the video helping me to learn Unraid. As a video suggestion could you please consider making a video on how to undelete files on Unraid xfs filesystem? I recently deleted some of my important files on my NAS arrary, I have been trying to see if there is a way to undelete the files, however, I haven't been able to be successful, I tried to use testdisk and photorec but have been unsuccessful so far. You may have communication with Unraid developers, if yes, could you please recommend adding a built in tool to assist with these type of issues?

Thank you man you saved me 🙏

It worked!!!!! Thank you!

I love you!?! You saved my day!!!!!!!!!

I've been your long time subscriber. Lately my Unraid is giving me parity sync error (not sure why or is it unclean shutdown). I've been running memtest and SMART on the disks right now. I've got this plugin called Dynamic File Integrity for a while but it's kinda confusing to use. It would be nice if you can teach us how to use it! Thank Spaceinvader!

life saver. still works in 6.11.3

In Unraid 7 beta there doesn't seem to be a text field after the check button to change/enter arguments.

very usefull as all of your vids. THX ;)

Thank you very much

Hello Spaceinvader.. I would like to request an update on reverse proxy but with the use of the docker-container "NginxProxyManager".. Have to say you make some fantastic videos on different aspects of unRaid, much love from this newbie.

@1:54 you say that we can see that SDH1 refers to the cache drive, but you didn't show us where the "SDH1" was, or the "SDX1" or anything, soooo... I'm quite a bit lost on how to find it.

Is there a way to see the logs of the screen in any log files? It can be difficult to find the issue when the startup is really fast

Brother, i almost faint, i try to play around with proxmox and then bring the unraid boot usb thru a vm passthru, and no idea how my usb got corrupted, i had a backup, so i create teh usb again and transfer the config folder, and 1/2 of my disk could not be mounted, i tried everuthing, and before decide to let go all family pics and delete the drives i try the -L comand and boom, it wors for me , thank you!!!

Hi, I accendently ran 'clear' for a few seconds on a drive with data....is there a chance to recover?

Thanks you saved my ass :) 😀

Does this apply for BTRFS?

A better option would be to use a better NAS implementation. If data integrity is important anyway. Given the # of complaints of this issue, the fact that the responses aren't something like, "This thing is crap, let's use a proper/real solution.", maybe data integrity isn't important. We used to have phones that _might_ have had PDA, PDA-like, app(s) on them. But no matter how awesome the PDA app(s) was/were, if the thing fell over w/being a phone (making/receiving calls), it was properly shit-canned. Shortly after the iPhone debuted, and _really_ after Apple opened it to apps and the app store, we've been walking around w/PDAs w/a phone app on them, but we still call them phones. Ostensibly their primary duty. Yet, now, the tolerance to the phone experience being absolutely crap/unworkable is bend-over-backwards-far-enough-to-break-your-back high. This is unraid wrt NAS platforms. It's primary job should be data integrity. Which, even w/out the data points here wrt instances of corruption, unraid fails hard. Laughably, or "cryably" depending. I don't get why people continue to not only use, but _pay_ to use, this thing in light of this round-0 failing.

Zfs ftw

Thank you so much for making this video!