December 31, 20232 yr Hi I set up a 20 drive zfs pool a while ago. Just fixed some power issues with the drives and unraid had disabled one drive. Was starting with one drive removed to try and enable the drive. However, I now can't mount because unraid says invalid config - it is expecting 30 drives now! How do I fix this pls? emhttpd: sas: invalid config: root_slots 30 num_slots 20 thanks unsas-diagnostics-20231231-2000.zip
December 31, 20232 yr Author Yes but I am worried that will lose the data. Is it safe to do that with ZFS?
December 31, 20232 yr Honestly, I can't say for certain so I apologize. I remember resetting the config in the past and never lost data. I think it just resets what drives unraid has in its memory. But if you're worried about it losing data, wait for someone who's more certain to chime in. Edited December 31, 20232 yr by rocky_mtn
January 1, 20242 yr Author Here it is: Quote root@UnSAS:~# zpool import pool: disk2 id: 5698125001356680421 state: ONLINE action: The pool can be imported using its name or numeric identifier. config: disk2 ONLINE sdo1 ONLINE pool: cache id: 7261360755002443771 state: ONLINE action: The pool can be imported using its name or numeric identifier. config: cache ONLINE mirror-0 ONLINE sds1 ONLINE sdq1 ONLINE pool: disk1 id: 7857505018516015911 state: ONLINE action: The pool can be imported using its name or numeric identifier. config: disk1 ONLINE sdp1 ONLINE root@UnSAS:~# Those three pools are unrelated to the SAS pool which has the problem. The 20 disks in that SAS pool do not show in the above import report. The main screen is attached in case that helps. unsas.pdf Edited January 1, 20242 yr by unraid newbie 55
January 1, 20242 yr Author root@UnSAS:~# zpool status no pools available root@UnSAS:~# unsas-diagnostics-20240101-1325.zip
January 1, 20242 yr Community Expert Something happened to that pool, it's not online and not showing as importable, this is independent of the GUI issues, does zpool import -D show anything?
January 1, 20242 yr Author Nope: root@UnSAS:~# zpool import -D no pools available to import Have I lost the data?
January 1, 20242 yr Community Expert I'm afraid you may have, try rebooting and then run zpool import again, if still nothing it's probably gone.
January 1, 20242 yr Author OK, thanks. I already rebooted a few times and nothing changed. The data was not critical but it still annoying to have lost it. In terms of what caused this, it all started with a hardware issue. I had used SATA power extensions to split PSU SATA cables from 3 to 8 connections. That set up is fixed now - all drives are powered either directly from the PSU SATA cables or via a MOLEX 1 to 4 splitter (4 splitters running from 4 MOLEX male connectors from one PSU MOLEX cable, 16 drives). Please let me know if that should fix the power issues I was having. Due to the power issues, Unraid has disabled one drive. I therefore started the array with one drive unassigned. Another disk was then disabled so I removed that one too. After doing that, Unraid screwed up and would not start the array due to an 'invalid config'. It thought there should be 30 disks not 20. It seems like this is at least partly an Unraid bug. I understand ZFS is new to Unraid and therefore unreliable. However, in future I want to be able to depend on the array not to lose data in this way. Should I ask for a refund for the expensive Unraid license and just use TrueNAS Scale instead? That seems to be a safer option and it is free as well. Perhaps Proxmox would be better, as the plugins did not work for me when I tried TrueNAS (also, it would not spin down the drives when idle). Thanks for your help with this
January 1, 20242 yr Community Expert 5 minutes ago, unraid newbie 55 said: Due to the power issues, Unraid has disabled one drive. I therefore started the array with one drive unassigned. Another disk was then disabled Was the pool created using the GUI or just imported? If the latter and it was not imported in the same order as zpool status it can cause issues when replacing/unassigning devices, but even so, I don't see how Unraid could have wiped all the pool devices, it's not like there one or two devices missing, no device from that pool is being found, there's nothing special about the Unraid zfs implementation, you can for example take an Unriaid created zfs pool and imported it with TrueNAS, I think something else happened here.
January 1, 20242 yr Author I created the pool using the GUI, 2 groups of 10 devices. Z-2. Should I try the above suggestion re new config?
January 1, 20242 yr Community Expert New config won't hurt, but don't see how it can help in this case, can you confirm if the server had been rebooted at least once after the pool was created and it still mounted? There's an issue currently that in some circumstances can make a device stop mounting after the first reboot, but for now only found that issue with the array, not sure if pools are affected, but if it mounted at least once after a reboot it cannot be that.
January 1, 20242 yr Author Yes - I must have rebooted at least 10 times over the last 4 months since the pool was created. It mounted every time until I now started the array after unassigning 2 drives. I see there is a 'no refunds' policy from unraid. Quite disappointing because I really don't know how I killed the ZFS pool.
January 1, 20242 yr Community Expert 31 minutes ago, unraid newbie 55 said: Yes - I must have rebooted at least 10 times OK, then it can't be that, post the output of fdisk -l /dev/sdX replace X with the correct identifier for a couple of pool devices, just to see if the partitions look correct.
January 1, 20242 yr Author root@UnSAS:~# fdisk -l /dev/sdd Disk /dev/sdd: 2.73 TiB, 3000592982016 bytes, 5860533168 sectors Disk model: ST3000NM0063 Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: gpt Disk identifier: 71A5CF4C-DCBE-4C23-A150-E6D805698D52 Device Start End Sectors Size Type /dev/sdd1 64 5860533134 5860533071 2.7T Linux filesystem
January 1, 20242 yr Author root@UnSAS:~# fdisk -l /dev/sde Disk /dev/sde: 2.73 TiB, 3000592982016 bytes, 5860533168 sectors Disk model: ST3000NM0063 Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: gpt Disk identifier: F3DDF53C-A043-40A8-B2F8-0A7E916D7D04 Device Start End Sectors Size Type /dev/sde1 64 5860533134 5860533071 2.7T Linux filesystem
January 1, 20242 yr Community Expert Partitions look normal, for those 2 devices post the output of: zdb -l /dev/sdX1 Replace X with correct letters again, but note the 1 in the end.
January 1, 20242 yr Author root@UnSAS:~# zdb -l /dev/sdd1 failed to unpack label 0 failed to unpack label 1 failed to unpack label 2 failed to unpack label 3 root@UnSAS:~# zdb -l /dev/sde1 failed to unpack label 0 failed to unpack label 1 failed to unpack label 2 failed to unpack label 3
January 1, 20242 yr Author I just finished running SMART extended self-tests on all 20 problem drives which contain the lost ZFS data. There were loads of errors on all the drives but the test corrected all the errors, reportedly. However, above issue continues after a reboot and rerunning zpool import etc. SMART reports attached. unsas-smart-20240101-1117.zip unsas-smart-20240101-0915.zip unsas-smart-20240101-0919.zip unsas-smart-20240101-1116.zip unsas-smart-20240101-0916.zip unsas-smart-20240101-0841.zip unsas-smart-20240101-1115.zip unsas-smart-20240101-1046.zip unsas-smart-20240101-0923.zip unsas-smart-20240101-0952.zip unsas-smart-20240101-1112.zip unsas-smart-20240101-1111.zip unsas-smart-20240101-0505.zip
January 2, 20242 yr Community Expert I don't see any reasons to suspect disk problems, main issue is that the superblock that should contain the zfs information is missing on all disks, hence why no pool is found with zpool import, just cannot say what caused that, don't remember ever seeing it before, especially since the partitions are still there and look normal.
January 2, 20242 yr Author OK - some good news. I took the advice of the first response above and created a 'new config', preserving all assignments. The server can start now and mount the problem pool. I was able to access the data via the SMB share. However, I have to manually import the zpool after reboot and the pool is still degraded (currently resilvering). It looks like one of the drives is listed as unavailable. Perhaps this is the one the removed from the pool after unraid disabled it. The drive still exists - how do I add it back in the pool? Thanks Console: Quote root@UnSAS:~# zpool status pool: cache state: ONLINE config: NAME STATE READ WRITE CKSUM cache ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 sds1 ONLINE 0 0 0 sdr1 ONLINE 0 0 0 errors: No known data errors pool: disk1 state: ONLINE config: NAME STATE READ WRITE CKSUM disk1 ONLINE 0 0 0 md1p1 ONLINE 0 0 0 errors: No known data errors pool: disk2 state: ONLINE config: NAME STATE READ WRITE CKSUM disk2 ONLINE 0 0 0 md2p1 ONLINE 0 0 0 errors: No known data errors root@UnSAS:~# zpool import pool: sas id: 17433224918486352194 state: DEGRADED status: One or more devices were being resilvered. action: The pool can be imported despite missing or damaged devices. The fault tolerance of the pool may be compromised if imported. config: sas DEGRADED raidz2-0 DEGRADED sdd1 ONLINE sde1 ONLINE sdj1 ONLINE sdx1 ONLINE replacing-4 DEGRADED 3335983727414034643 FAULTED corrupted data sdy1 ONLINE sdv1 ONLINE replacing-6 DEGRADED 11216585848571590061 FAULTED corrupted data sdn1 ONLINE sdb1 ONLINE sdc1 ONLINE sdf1 ONLINE raidz2-1 DEGRADED sdi1 ONLINE sdt1 ONLINE sdm1 ONLINE sdw1 ONLINE sdh1 ONLINE sdg1 ONLINE sdq1 ONLINE sdu1 ONLINE 12723775195252878681 UNAVAIL sdk1 ONLINE root@UnSAS:~# zpool import sas root@UnSAS:~# zpool status pool: cache state: ONLINE config: NAME STATE READ WRITE CKSUM cache ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 sds1 ONLINE 0 0 0 sdr1 ONLINE 0 0 0 errors: No known data errors pool: disk1 state: ONLINE config: NAME STATE READ WRITE CKSUM disk1 ONLINE 0 0 0 md1p1 ONLINE 0 0 0 errors: No known data errors pool: disk2 state: ONLINE config: NAME STATE READ WRITE CKSUM disk2 ONLINE 0 0 0 md2p1 ONLINE 0 0 0 errors: No known data errors pool: sas state: DEGRADED status: One or more devices is currently being resilvered. The pool will continue to function, possibly in a degraded state. action: Wait for the resilver to complete. scan: resilver in progress since Sun Dec 31 10:58:54 2023 2.19T scanned at 21.1G/s, 1.10T issued at 10.6G/s, 5.38T total 2.05G resilvered, 20.38% done, 00:06:54 to go config: NAME STATE READ WRITE CKSUM sas DEGRADED 0 0 0 raidz2-0 DEGRADED 0 0 0 sdd1 ONLINE 0 0 0 sde1 ONLINE 0 0 0 sdj1 ONLINE 0 0 0 sdx1 ONLINE 0 0 0 replacing-4 DEGRADED 0 0 0 3335983727414034643 FAULTED 0 0 0 was /dev/mapper/sdu1 sdy1 ONLINE 0 0 0 (resilvering) sdv1 ONLINE 0 0 0 replacing-6 DEGRADED 0 0 0 11216585848571590061 FAULTED 0 0 0 was /dev/mapper/sdy1 sdn1 ONLINE 0 0 0 (resilvering) sdb1 ONLINE 0 0 0 sdc1 ONLINE 0 0 0 sdf1 ONLINE 0 0 0 raidz2-1 DEGRADED 0 0 0 sdi1 ONLINE 0 0 0 sdt1 ONLINE 0 0 0 sdm1 ONLINE 0 0 0 sdw1 ONLINE 0 0 0 sdh1 ONLINE 0 0 0 sdg1 ONLINE 0 0 0 sdq1 ONLINE 0 0 0 sdu1 ONLINE 0 0 0 12723775195252878681 UNAVAIL 0 0 0 was /dev/mapper/sdo1 sdk1 ONLINE 0 0 0 errors: No known data error unsas-diagnostics-20240102-1438.zip
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.