February 20, 20242 yr As it says in the title. I just turned my server on one day and everything was messed up. My disks are still partly occupied so my files seem to still be there, but I don't know what happened with everything else. I'm attaching the diagnostics. Any help would be very much appreciated as I have just the very basic knowledge on all this! tower-diagnostics-20240220-1808.zip
February 20, 20242 yr Community Expert Why do you have dual parity with only 2 data disks? No SMART or lsscsi for the nvme you have as cache. Looks like it disconnected. Disable Docker in Settings until you get that fixed.
February 20, 20242 yr Community Expert Cache device is dropping offline: Feb 20 09:03:49 Tower kernel: nvme nvme0: I/O 6 QID 0 timeout, reset controller Feb 20 09:04:22 Tower kernel: nvme nvme0: Abort status: 0x371 Feb 20 09:04:22 Tower kernel: nvme nvme0: 8/0/0 default/read/poll queues Feb 20 09:04:53 Tower kernel: nvme nvme0: I/O 65 QID 1 timeout, disable controller Feb 20 09:05:23 Tower kernel: nvme nvme0: I/O 5 QID 0 timeout, disable controller Feb 20 09:06:01 Tower kernel: nvme nvme0: Device shutdown incomplete; abort shutdown
February 20, 20242 yr Author 3 minutes ago, trurl said: Why do you have dual parity with only 2 data disks? No SMART or lsscsi for the nvme you have as cache. Looks like it disconnected. Disable Docker in Settings until you get that fixed. I have dual parity as I planned on adding more data disks in the forseeable future. Where would I enable SMART or lsscsi for it? So you're saying it looks like it's physically disconnected?
February 20, 20242 yr Community Expert Just now, krizo96 said: Where would I enable SMART or lsscsi for it? You don't enable, the fact they aren't there means 1 minute ago, krizo96 said: it's physically disconnected
February 20, 20242 yr Author 4 minutes ago, trurl said: You don't enable, the fact they aren't there means I did think SMART was something that was usually just there. I disconnected the drive and reinserted it, but nothing changed. Also tried inserting it into an expansion card instead and it's still not recognizing it. Is it possible it could be faulty or died?
February 21, 20242 yr Author 7 hours ago, JorgeB said: Cache device is dropping offline: Feb 20 09:03:49 Tower kernel: nvme nvme0: I/O 6 QID 0 timeout, reset controller Feb 20 09:04:22 Tower kernel: nvme nvme0: Abort status: 0x371 Feb 20 09:04:22 Tower kernel: nvme nvme0: 8/0/0 default/read/poll queues Feb 20 09:04:53 Tower kernel: nvme nvme0: I/O 65 QID 1 timeout, disable controller Feb 20 09:05:23 Tower kernel: nvme nvme0: I/O 5 QID 0 timeout, disable controller Feb 20 09:06:01 Tower kernel: nvme nvme0: Device shutdown incomplete; abort shutdown I disconnected the drive and reinserted it, but nothing changed. Also tried inserting it into an expansion card instead and it's still not recognizing it. Is it possible it could be faulty or died?
February 21, 20242 yr Community Expert Do you have something else you can use to test it? I have several nvme I have scavenged, and got USB3 enclosures for them, for example so I can plug them into laptops and PCs.
February 21, 20242 yr Author 14 minutes ago, trurl said: Do you have something else you can use to test it? I have several nvme I have scavenged, and got USB3 enclosures for them, for example so I can plug them into laptops and PCs. I will try to procure it to test the drive. Is there a way to restore the shares I had if the drive really is dead?
February 21, 20242 yr Community Expert 5 minutes ago, krizo96 said: restore the shares I had What do you think is missing besides appdata?
February 21, 20242 yr Author 7 minutes ago, trurl said: What do you think is missing besides appdata? I had my appdata, system and domains folder on the cache drive.
February 21, 20242 yr Community Expert Yes, it is recommended that those are not on the array. Since it was not a redundant pool, no way to recover its files if it has failed. There are plugins for backing up appdata and VMs, for example, but doesn't look like you had them. Previous Apps on the Apps page should let you reinstall your containers just as they were, but without their appdata they would be new installs. Note that if you do get nvme in an enclosure, Windows won't be able to read its filesystem. But Windows might be able to get its SMART report from something like CrystalDiskInfo, for example. And, any Linux should be able to read btrfs, including Unassigned Devices in Unraid.
February 21, 20242 yr Author 3 minutes ago, trurl said: Yes, it is recommended that those are not on the array. Since it was not a redundant pool, no way to recover its files if it has failed. There are plugins for backing up appdata and VMs, for example, but doesn't look like you had them. Previous Apps on the Apps page should let you reinstall your containers just as they were, but without their appdata they would be new installs. Note that if you do get nvme in an enclosure, Windows won't be able to read its filesystem. But Windows might be able to get its SMART report from something like CrystalDiskInfo, for example. And, any Linux should be able to read btrfs, including Unassigned Devices in Unraid. Thanks for the detailed explanation. The most important thing to me were my shares. As far as I understand and can see all that data is still there. What would be the simplest way to restore those? If I try to browse the disks in the unraid interface I can still see my photos and media folders on the drives.
February 21, 20242 yr Community Expert 5 minutes ago, krizo96 said: all that data is still there. What would be the simplest way to restore those? What do you think needs to be restored? Your diagnostics showed user shares were working appdata shareUseCache="prefer" # Share exists on cache domains shareUseCache="prefer" # Share does not exist isos shareUseCache="yes" # Share exists on disk1, disk2 M---a shareUseCache="yes" # Share exists on disk1, disk2 p----s shareUseCache="yes" # Share exists on disk1, disk2 system shareUseCache="prefer" # Share does not exist
February 21, 20242 yr Author 3 minutes ago, trurl said: What do you think needs to be restored? Your diagnostics showed user shares were working appdata shareUseCache="prefer" # Share exists on cache domains shareUseCache="prefer" # Share does not exist isos shareUseCache="yes" # Share exists on disk1, disk2 M---a shareUseCache="yes" # Share exists on disk1, disk2 p----s shareUseCache="yes" # Share exists on disk1, disk2 system shareUseCache="prefer" # Share does not exist That's weird. When I go to my shares tab there are none there nor on when I look through my PC.
February 21, 20242 yr Author 2 minutes ago, trurl said: Post new diagnostics There you go. But it looked like this when I exported the first diagnostics I posted as well. tower-diagnostics-20240221-0350.zip
February 21, 20242 yr Author 49 minutes ago, JorgeB said: Unassign the NVMe device, start array and post new diags. Okay, that's progress. When I unassigned it the shares appeared again. tower-diagnostics-20240221-1107.zip
February 21, 20242 yr Community Expert Since that device is not working leave it unassigned, you can then assign a new one if you want.
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.