October 8, 20241 yr Greetings, everyone: I recently had a drive in my unraid server go bad. I recently replaced the drive and ran a data rebuild on it and for a week the drive was operating normally. But after several days the new drive got disabled and this popped up on the web terminal: Sep 18 08:40:07 Tower kernel: BTRFS error (device loop2: state EA): bdev /dev/loop2 errs: wr 0, rd 1655, flush 1, corrupt 0, gen 0 Sep 18 08:40:07 Tower kernel: sd 8:0:0:0: [sdf] tag#20 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=DRIVER_OK cmd_age=0s Sep 18 08:40:07 Tower kernel: sd 8:0:0:0: [sdf] tag#20 CDB: opcode=0x28 28 00 05 e4 cc e0 00 00 08 00 Sep 18 08:40:17 Tower kernel: sd 8:0:0:0: [sdf] tag#14 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=DRIVER_OK cmd_age=0s I have looked on the forum for posts dealing with similar problems but the solutions always require for the SATA cables to be replaced. However, I don't think that this necessary because I installed the new drive with new SATA cables. What should I do? ST8000DM004-2U9188_ZR15X21K-20241008-1304.txt tower-diagnostics-20241008-1350.zip
October 8, 20241 yr Community Expert Disk was already disabled at boot, so we can't see what happened, but disk looks fine, so most likely a power/connection issue
October 8, 20241 yr Author 29 minutes ago, JorgeB said: Disk was already disabled at boot, so we can't see what happened, but disk looks fine, so most likely a power/connection issue What should I do?
October 9, 20241 yr Community Expert Replace the cables for that disk to rule that out, and if the emulated disk is still mounting and contents look correct, rebuild on top: https://docs.unraid.net/unraid-os/manual/storage-management#rebuilding-a-drive-onto-itself
October 15, 20241 yr Author On 10/9/2024 at 3:26 AM, JorgeB said: Replace the cables for that disk to rule that out, and if the emulated disk is still mounting and contents look correct, rebuild on top: https://docs.unraid.net/unraid-os/manual/storage-management#rebuilding-a-drive-onto-itself Ok, so I ordered new SATA cables for the drive. I replaced the cables and ran a data rebuild but the drive got disabled again. This is what showed up in the log: Oct 15 05:47:21 Tower kernel: sd 10:0:0:0: [sdg] tag#15 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=DRIVER_OK cmd_age=0s Oct 15 05:47:21 Tower kernel: sd 10:0:0:0: [sdg] tag#15 CDB: opcode=0x88 88 00 00 00 00 03 a3 81 29 c6 00 00 00 01 00 00 Oct 15 05:47:21 Tower kernel: I/O error, dev sdg, sector 15628052934 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 2 Oct 15 05:47:21 Tower kernel: Buffer I/O error on dev sdg1, logical block 15628052870, async page read What should I do now?
October 15, 20241 yr Author 8 hours ago, JorgeB said: Did you also replace the power cable? Yes, I did.
October 16, 20241 yr Author 3 hours ago, JorgeB said: In that case would recommend replacing the disk. Even though the disk is new and it has passed the smart test? Do you you think that this issue might be related to one of the parity drives?
October 16, 20241 yr Community Expert If replacing cables didn't solve the problem, the disk could be bad, despite a good SMART, but could also be a bad PSU for example, or bad splitter cable, etc.
October 22, 20241 yr Author On 10/16/2024 at 7:54 AM, JorgeB said: If replacing cables didn't solve the problem, the disk could be bad, despite a good SMART, but could also be a bad PSU for example, or bad splitter cable, etc. I notice that when the data rebuild fails. Cpu 3 is at 100%. Could this be the issue?
November 3, 20241 yr Author I ran a new one on that disk a couple of days ago and these were the results: SmartFTP open device: /dev/sdg failed: INQUIRY failed When it comes to read errors this is what pops up whenever I try streaming something off the disk: Read error, sector=8876464320 However, when I run a parity check after reconfiguring this is what shows up: Ata2: COMRESET failed and then this shows up: disk 1 write error, sector 27927576 I am starting to think that this issue is stemming from the PSU and that I will probably have to get one that can handle more wattage. What do you think I should do?
November 3, 20241 yr Community Expert 1 hour ago, cdixon said: SmartFTP open device Do you mean smartctl? Post new diags.
November 3, 20241 yr Community Expert 9 minutes ago, cdixon said: It said smartFTP. That's not a disk testing app. Diags show issues with multiple devices, looks more like a power/connection issue, check cables, and if you are using any splitters remove or replace them.
November 4, 20241 yr Author I have replaced two of my splitter cables. Should I replace the other one? And could this be an issue with my psu? Maybe I have too many drives plugged into it?
November 7, 20241 yr Author So do you think I should replace the psu with one with more wattage? On 10/16/2024 at 7:54 AM, JorgeB said: If replacing cables didn't solve the problem, the disk could be bad, despite a good SMART, but could also be a bad PSU for example, or bad splitter cable, etc. On 10/22/2024 at 10:38 AM, JorgeB said: Seems unlikely On 11/4/2024 at 8:00 AM, JorgeB said: Could be.
November 7, 20241 yr Community Expert If you have a different one you could test with, it's worth a try.
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.