February 22, 20233 yr I've done some search before asking here. Here's my situation. I have a Unraid server which was running fine. The other day I had to close my server, and I don't know if I closed it wrong (maybe did a shutdown while the array was still started, or maybe other issue (could be hardware related, I don't know)). But when I rebooted the array, one of my drive was unmountable. All other drives showed up, except disk 2, a 10TB HDD which contains backups. I tried to do a parity check but it was painfully slow (like 1MB/s) when I normally get around 150MB/s. I thought something was wrong with my drive, so I ordered online a new 10TB HDD to replace it. While waiting for the new drive to arrive I tried to see if I could retrieve data from the drive. I tried 2 things: - Of course XFS is not recognized by windows, but I found that PowerISO could read XFS file. It does, I can load the drive, PowerISO show me the data on the drive and I could "extract it" to my local computer. But wow is that slow, I mean, for around 2TB it showed over 3000min (50hrs) ... I have around 4.5TB I want to retrieve... So that solution could work, but is not my preferred way. - I had an old computer laying around so I installed Ubuntu on it, thinking I should be able to easily mount the drive and then copy through my network, surely that would take less time and would be a good solution if that happens again on my Unraid Server. But no, I get the "wrong fs type, bad option, bad superblock ... " message. I've search a few way to do this without overwriting the data, but nothing I've tried works. Once I got the new 10TB drive, I think I made the wrong move, when I plugged it in UnRaid. I did a parity check before formating the new drive. So I now have the feeling that I can no longer retrieve my data through parity from UnRaid. What are my options here? How can I easily take one of my drive in my UnRaid Array and read it on another computer? Would've I been able to retrieve my data if I did a "format Unmounted device" of my new 10TB drive before check parity? I'm asking this in case this happens again. Thank you for your help.
February 22, 20233 yr 2 minutes ago, HighTick said: Would've I been able to retrieve my data if I did a "format Unmounted device" of my new 10TB drive before check parity No, absolutely not. Format is never a part of data recovery. It is a write operation which irrevocably removes any chances of recovering data because from that point on, the new correct state of drive is formatted and empty Did you run a correcting parity check? Did you add the drive back into the same slot in array? I suggest you don't so anything else for now and post diagnostics first so someone can suggest a next step
February 22, 20233 yr Author If you're asking about Write correction to parity, yes I did it with the new unformated hard drive. Yes I've added the New drive into the same slot array and on the same sata port. Thank you unraid-diagnostics-20230221-2225.zip
February 22, 20233 yr Community Expert Wish you had asked for help before doing anything at all. These problems are easily handled in Unraid and trying to do it otherwise will usually make things worse. And it looks like you have made things worse. Possibly nothing was wrong with any disks, bad connections are more common than bad disks. Most of the answers for your problems are in this section of the wiki: https://wiki.unraid.net/Manual/Storage_Management Nothing assigned as disk5 but you do have 1,2,3,4,6,7. No drives are currently disabled or unmountable. Disk2 has little if any data. Same for disks 1 , 4 but can't tell for sure since you are using a version of Unraid with a bug in diagnostics collection which prevents us knowing much about your user shares. SSDs in the parity array cannot be trimmed, and can only be written at parity speed. Usually you would put SSDs in pools outside the parity array and only put HDDs in the array. Current syslog is included in diagnostics, but that resets on reboot. Diagnostics can tell us a lot about how things are now, but can't tell us anything about what happened before reboot. Your descriptions of what happened and what you did aren't entirely clear and must be missing some details, but it probably doesn't matter since we will just have to go from how things are now. Disable Docker and VM Manager in Settings until things are stable and configured like you need them. Do you have backups of anything important and irreplaceable?
February 22, 20233 yr Community Expert 1 hour ago, HighTick said: How can I easily take one of my drive in my UnRaid Array and read it on another computer? Any linux can work with your Unraid disks, but putting your Unraid disks in another system will invalidate Unraid parity. Possibly you had an unmountable disk due to filesystem corruption, and that would have required filesystem repair before any files could be accessed. Not really necessary to put these in another computer anyway since Unraid is the best and easiest way to work with them and fix any problems. 1 hour ago, HighTick said: I'm asking this in case this happens again. If you have any problems, please ask for help before doing anything at all. So, what do you want to do now?
February 22, 20233 yr Author Hi, Thanks for your input. I'd say that I generally hesitate to ask question on forum as I see so often (around the internet) people bashing on newbie saying (read the f* manual or google is your friend...) So I try things on my own, but as you can see, I f* it up. What do you mean by bad connection? Bad sata wire or not well plugged in? I do have hard drive tray. Wouldn't a bad connection make the drive not recognized by the computer (my disk 2 was recognize but unmountable) Thank you for the link, I'll bookmark it. I have nothing on drive 5 simply because when I made my computer setup (before buying server components) in my excel file I wanted to arrange my drive in a logical way (for example having all my media on HDD on consecutive drive, VM drive on consecutive NVMe drive,...). I do realize now, that this was useless. Disk 1 is for my ISOs file for VM or Docker that I download, so for now there's nothing much on this drive. Disk 4, is a new NVMe drive, almost nothing for now on this one. The issue is with Disk 2, I had 4.5TB on this, but after parity check (with writing corrections) I lost the parity protection with Unraid. So currently, the newer 10TB drive is installed in Unraid on Disk 2 slot and is empty. I still have the older (disk 2) 10TB with data on it. I can see the data with PowerISO, to copy the data, I need to "extract it" and the extraction is really long, for now I managed to retrieve important data. Still have a lot of unretrieved data on the drive. Quote SSDs in the parity array cannot be trimmed, and can only be written at parity speed. Usually you would put SSDs in pools outside the parity array and only put HDDs in the array. I'm not sure I understand. I know that my SSD speed is currently limited by my parity drive, but I thought I had no choice. I'm not clear on what happened, because I'm not sure what happened (and because my basic language is french, so I sometime write french expression with english word) ... Quick resumé: - I had to move my server - I shutdown my UnRaid server - Moved my server - Started back my UnRaid server, all drives where fine, except disk 2 was unmountable... - I started a partity check but it was painfully slow (1MB/s) and the drive made some noise, so I stopped it. (That noise is no longer when I use PowerISO to extract data) - I bought a new 10TB drive because the noise was not reassuring - Tried to rebuild the parity with the new 10TB drive (this is where I made things worse). - I tried to mount the "old" 10TB drive in a linux computer to retrieve data, but linux won't let me mount the drive (can this be related to filesystem needing repair?) - The only way I can read the data is by using PowerISO in windows. This is really slow, but at least my data is not lost. I already retrieved important data through PowerISO. But still have data left. What I want to do now? - I would like to better understand the "Write corrections to parity". I'll read the link you provided - I would like to make a better setup of my UnRaid arrays (for example you said I should put my SSD in a different pool). - I would like to know if there's a way I can read the old drive faster than with PowerISO "extracting". - How to do a filesystem repair when in unraid (maybe it's in the link you provided) Thank you for your support
February 22, 20233 yr Community Expert 3 hours ago, HighTick said: hesitate to ask question on forum as I see so often (around the internet) people bashing on newbie saying (read the f* manual or google is your friend...) That generally doesn't happen here, and if it does, the basher is usually the one to feel the heat. I hesitated to refer you to the documentation because I didn't want to give that impression. 3 hours ago, HighTick said: bad connection? Some problem communicating with the disk rather than an actual bad disk. Could be data cables, could be power cables or splitters, could be controller... 3 hours ago, HighTick said: still have the older (disk 2) 10TB with data on it. We might be able to work with that in Unraid. Do you have an unused port you can use to connect that?
February 22, 20233 yr Author Yes I have a lot of free sata port, I have a SuperMicro X10DRi-T4+ I'll install the drive and also check if my cable are well connected. Thank you
February 22, 20233 yr Author Ok, the hard drive is installed and is now seen as an Unassigned device Dev 1
February 22, 20233 yr Community Expert 54 minutes ago, HighTick said: Unassigned device Dev 1 sdd From the command line xfs_repair /dev/sdd1 Capture the output and post it.
February 22, 20233 yr Author Phase 1 - find and verify superblock... Phase 2 - using internal log - zero log... ERROR: The filesystem has valuable metadata changes in a log which needs to be replayed. Mount the filesystem to replay the log, and unmount it before re-running xfs_repair. If you are unable to mount the filesystem, then use the -L option to destroy the log and attempt a repair. Note that destroying the log may cause corruption -- please attempt a mount of the filesystem before doing this.
February 22, 20233 yr Community Expert From the command line xfs_repair -L /dev/sdd1 Capture the output and post it.
February 22, 20233 yr Author Phase 1 - find and verify superblock... Phase 2 - using internal log - zero log... ALERT: The filesystem has valuable metadata changes in a log which is being destroyed because the -L option was used. - scan filesystem freespace and inode maps... clearing needsrepair flag and regenerating metadata - found root inode chunk Phase 3 - for each AG... - scan and clear agi unlinked lists... - process known inodes and perform inode discovery... - agno = 0 - agno = 1 - agno = 2 - agno = 3 - agno = 4 - agno = 5 - agno = 6 - agno = 7 - agno = 8 - agno = 9 - process newly discovered inodes... Phase 4 - check for duplicate blocks... - setting up duplicate extent list... - check for inodes claiming duplicate blocks... - agno = 1 - agno = 3 - agno = 4 - agno = 0 - agno = 6 - agno = 9 - agno = 5 - agno = 2 - agno = 7 - agno = 8 Phase 5 - rebuild AG headers and trees... - reset superblock... Phase 6 - check inode connectivity... - resetting contents of realtime bitmap and summary inodes - traversing filesystem ... - traversal finished ... - moving disconnected inodes to lost+found ... Phase 7 - verify and correct link counts... Maximum metadata LSN (1:1680997) is ahead of log (1:2). Format log to cycle 4. done
February 22, 20233 yr Community Expert That looks hopeful. Install Unassigned Device plugin and see if you can mount the disk as an Unassigned Device.
February 22, 20233 yr Author Ok that worked, I mounted the drive and can now see the data. Just need to create a new share
February 23, 20233 yr Community Expert Install Dynamix File Manager plugin and it will help you copy that data to the array. If you want to put the disk in the array while keeping its data, please ask for advice.
April 8, 20233 yr Author Hi, it happened again. Wednesday, there was a freezing rain that cut electricity in montreal area and I had to shutdown my server because I had only 60min on my smart ups and the electricity was off for a full day. On the power up, I got the same Unmountable disk on disk2, however it's the new drive, but the same sata port and sata hdd tray. I stopped the array after I saw the unmountable drive and took a diagnostic. I went to the wiki : https://wiki.unraid.net/Manual/Storage_Management but not really sure what to do... I was thinking about doing " xfs_repair -v /dev/md2 " but last time I did " xfs_repair -L /dev/sdd1 " Thank you unraid-diagnostics-20230405-2104.zip
April 8, 20233 yr Community Expert For a disk in the array, you must specify the md device or you will invalidate parity. Better if you do check filesystem from the webUI instead of the command line. That way it will use the correct command.
April 8, 20233 yr 3 hours ago, HighTick said: I had to shutdown my server because I had only 60min on my smart ups and the electricity was off for a full day. First... I know about the storm (feeling for you folks out east). Second... You shouldn't have to shut down the server; it should shut itself down, properly, when power is running out. I've had this happen three (3) times with my system for various reasons. 3 hours ago, HighTick said: I got the same Unmountable disk on disk2, however it's the new drive, If you're sure the drive is good then, maybe, it's the cable to that drive? I haven't read the entire thread properly, but I hope this helps. MrGrey.
April 8, 20233 yr Author ok, I did the check file system with option -nv at first and it found errors I had to do the -L twice, but now everything works fine. Thank you, I'll know what to do next time it happens. Meanwhile I'll check the sata connection, I thought it was ok, but the 2nd time on the same sata show something is wrond. Again thank you
April 8, 20233 yr Author @MrGrey Yeah the smart ups is plugged in with serial, so it would've turned off, however, I knew we would miss electricity for a long time, and I don't know how to turn off the alarm on my ups, easiest way for me was to turn my server off so I can also turn off my ups... Once the electricity came back, next morning it turned off again, but I wasn't aware (I was at work) and it did shut itself down, most likely my dogs didn't like the alarm during that time! I use HDD tray, so this might be the issue, I'll take a look at it.
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.