May 7, 20233 yr After figuring out that my motherboard just doesn't like Seagate drives, I finally managed to have a solid/stable server for a couple of months. No more parity check errors, drives disabling, etc. I thought I had it solved. Then I decided to start upgrading my 12tb drives to 20tb drives (slowly). I started with 2 drives for my parity. I precleared them with no issues. I then swapped one in as parity 1 and let it rebuild. That went fine. Then, as luck would have it, while swapping the other into Parity 2, the parity 1 drive started having issues and then disabled. I didn't know because I was at work at the time. I came home, and had to start the whole process over again. Not really knowing what I'm doing, I just started over and precleared both drives again and this time swapped the 2 drives, so the 2nd drive was now parity 1. That seemed to work. I was then able to get parity 2 installed as well, and all was good for about 3 days. Then, Parity 2 disabled again. Maybe Parity 2 is bad? I posted the logs in discord and everybody helping agreed the problem isn't with the drive but with the SATA cables. So, I bought new ones. And actually, that didn't help (2 days pasted then they disabled again) so I bought ANOTHER set of cables. Thought I had it, then after a full week they disabled again. I've since bought 2 more 20tb drives and have swapped them into the parity. They ALSO have been having this disabling issue. So my latest attempt has been to first of all do an extended self-test on the parity 1 drive, which came back perfect. I then did an extended test on my parity 2 drive, which was perfect too. I have another 20tb now, and prior to doing it's preclear I did an extended test on it, and it was perfect as well. I then have started the preclears, this time doing the full gamut of tests, pre-read then clear then post-read. Everything has been going great on the 2nd new drive. But, just as expected, the one connected as Parity 2 has started having issues again. It's not been disabled yet, but I imagine it will soon. My current setup... The motherboard has 4 SATA ports on it. The "array" drives are connected to an HBA. The parity drives are connected to the first 2 motherboard ports, and an SSD is connected to a 3rd used for VM's. The 4th port is not used. I have tried swapping SATA ports as well (when trying to replace the cables) so they were using 3 & 4, and had the SSD disconnected. This just doesn't seem to be working. At this point, I'm out of ideas. I've tried different drives, different cables, different ports. I don't know what else to try. I'm starting to suspect the problem might be with the motherboard. I realize SATA drives are pretty much a standard. But is it possible my BIOS doesn't work well with this capacity? I checked the manufacturer's website and I do see they have newer BIOS's available, but nothing directly indicates they have improved compatibility with drives. I posted a thread in their forums but nobody has replied in a week. Am I just SOL? Should I maybe look at getting a controller card and not use the onboard? If so, any good recommendations? I really appreciate any help anybody can provide. I'm just totally lost now. Here's the recent errors my drive is showing. Diagnostics also attached. Spoiler May 7 00:38:37 Kyber kernel: ata2.00: exception Emask 0x10 SAct 0x1c SErr 0x90302 action 0xe frozen May 7 00:38:37 Kyber kernel: ata2.00: irq_stat 0x00400000, PHY RDY changed May 7 00:38:37 Kyber kernel: ata2: SError: { RecovComm UnrecovData Persist PHYRdyChg 10B8B } May 7 00:38:37 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 00:38:37 Kyber kernel: ata2.00: cmd 60/c0:10:40:8f:00/04:00:eb:00:00/40 tag 2 ncq dma 622592 in May 7 00:38:37 Kyber kernel: ata2.00: status: { DRDY } May 7 00:38:37 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 00:38:37 Kyber kernel: ata2.00: cmd 60/40:18:00:94:00/05:00:eb:00:00/40 tag 3 ncq dma 688128 in May 7 00:38:37 Kyber kernel: ata2.00: status: { DRDY } May 7 00:38:37 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 00:38:37 Kyber kernel: ata2.00: cmd 60/c0:20:40:99:00/00:00:eb:00:00/40 tag 4 ncq dma 98304 in May 7 00:38:37 Kyber kernel: ata2.00: status: { DRDY } May 7 00:38:37 Kyber kernel: ata2: hard resetting link May 7 00:38:39 Kyber kernel: ata2: found unknown device (class 0) May 7 00:38:39 Kyber kernel: ata2: SATA link down (SStatus 0 SControl 300) May 7 00:38:39 Kyber kernel: ata2: hard resetting link May 7 00:38:45 Kyber kernel: ata2: found unknown device (class 0) May 7 00:38:49 Kyber kernel: ata2: softreset failed (device not ready) May 7 00:38:49 Kyber kernel: ata2: hard resetting link May 7 00:38:54 Kyber kernel: ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 7 00:38:54 Kyber kernel: ata2.00: configured for UDMA/133 May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 Sense Key : 0x5 [current] May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 ASC=0x21 ASCQ=0x4 May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 CDB: opcode=0x88 88 00 00 00 00 00 eb 00 8f 40 00 00 04 c0 00 00 May 7 00:38:54 Kyber kernel: I/O error, dev sdb, sector 3942682432 op 0x0:(READ) flags 0x80700 phys_seg 152 prio class 0 May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 Sense Key : 0x5 [current] May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 ASC=0x21 ASCQ=0x4 May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 CDB: opcode=0x88 88 00 00 00 00 00 eb 00 94 00 00 00 05 40 00 00 May 7 00:38:54 Kyber kernel: I/O error, dev sdb, sector 3942683648 op 0x0:(READ) flags 0x84700 phys_seg 168 prio class 0 May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#4 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#4 Sense Key : 0x5 [current] May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#4 ASC=0x21 ASCQ=0x4 May 7 00:38:54 Kyber kernel: sd 2:0:0:0: [sdb] tag#4 CDB: opcode=0x88 88 00 00 00 00 00 eb 00 99 40 00 00 00 c0 00 00 May 7 00:38:54 Kyber kernel: I/O error, dev sdb, sector 3942684992 op 0x0:(READ) flags 0x80700 phys_seg 24 prio class 0 May 7 00:38:54 Kyber kernel: ata2: EH complete May 7 01:30:21 Kyber kernel: ata2.00: exception Emask 0x10 SAct 0x70000 SErr 0x90202 action 0xe frozen May 7 01:30:21 Kyber kernel: ata2.00: irq_stat 0x00400000, PHY RDY changed May 7 01:30:21 Kyber kernel: ata2: SError: { RecovComm Persist PHYRdyChg 10B8B } May 7 01:30:21 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 01:30:21 Kyber kernel: ata2.00: cmd 60/c0:80:40:e9:60/00:00:52:01:00/40 tag 16 ncq dma 98304 in May 7 01:30:21 Kyber kernel: ata2.00: status: { DRDY } May 7 01:30:21 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 01:30:21 Kyber kernel: ata2.00: cmd 60/40:88:00:ea:60/05:00:52:01:00/40 tag 17 ncq dma 688128 in May 7 01:30:21 Kyber kernel: ata2.00: status: { DRDY } May 7 01:30:21 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 01:30:21 Kyber kernel: ata2.00: cmd 60/c0:90:40:ef:60/04:00:52:01:00/40 tag 18 ncq dma 622592 in May 7 01:30:21 Kyber kernel: ata2.00: status: { DRDY } May 7 01:30:21 Kyber kernel: ata2: hard resetting link May 7 01:30:23 Kyber kernel: ata2: found unknown device (class 0) May 7 01:30:23 Kyber kernel: ata2: SATA link down (SStatus 0 SControl 300) May 7 01:30:23 Kyber kernel: ata2: hard resetting link May 7 01:30:28 Kyber kernel: ata2: found unknown device (class 0) May 7 01:30:33 Kyber kernel: ata2: softreset failed (device not ready) May 7 01:30:33 Kyber kernel: ata2: hard resetting link May 7 01:30:37 Kyber kernel: ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 7 01:30:37 Kyber kernel: ata2.00: configured for UDMA/133 May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#16 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#16 Sense Key : 0x5 [current] May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#16 ASC=0x21 ASCQ=0x4 May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#16 CDB: opcode=0x88 88 00 00 00 00 01 52 60 e9 40 00 00 00 c0 00 00 May 7 01:30:37 Kyber kernel: I/O error, dev sdb, sector 5677050176 op 0x0:(READ) flags 0x80700 phys_seg 24 prio class 0 May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#17 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#17 Sense Key : 0x5 [current] May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#17 ASC=0x21 ASCQ=0x4 May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#17 CDB: opcode=0x88 88 00 00 00 00 01 52 60 ea 00 00 00 05 40 00 00 May 7 01:30:37 Kyber kernel: I/O error, dev sdb, sector 5677050368 op 0x0:(READ) flags 0x84700 phys_seg 168 prio class 0 May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#18 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#18 Sense Key : 0x5 [current] May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#18 ASC=0x21 ASCQ=0x4 May 7 01:30:37 Kyber kernel: sd 2:0:0:0: [sdb] tag#18 CDB: opcode=0x88 88 00 00 00 00 01 52 60 ef 40 00 00 04 c0 00 00 May 7 01:30:37 Kyber kernel: I/O error, dev sdb, sector 5677051712 op 0x0:(READ) flags 0x80700 phys_seg 152 prio class 0 May 7 01:30:37 Kyber kernel: ata2: EH complete May 7 09:27:27 Kyber kernel: ata2.00: exception Emask 0x10 SAct 0x3 SErr 0x90202 action 0xe frozen May 7 09:27:27 Kyber kernel: ata2.00: irq_stat 0x00400000, PHY RDY changed May 7 09:27:27 Kyber kernel: ata2: SError: { RecovComm Persist PHYRdyChg 10B8B } May 7 09:27:27 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 09:27:27 Kyber kernel: ata2.00: cmd 60/40:00:00:2a:93/05:00:c3:04:00/40 tag 0 ncq dma 688128 in May 7 09:27:27 Kyber kernel: ata2.00: status: { DRDY } May 7 09:27:27 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 09:27:27 Kyber kernel: ata2.00: cmd 60/c0:08:40:2f:93/04:00:c3:04:00/40 tag 1 ncq dma 622592 in May 7 09:27:27 Kyber kernel: ata2.00: status: { DRDY } May 7 09:27:27 Kyber kernel: ata2: hard resetting link May 7 09:27:29 Kyber kernel: ata2: found unknown device (class 0) May 7 09:27:29 Kyber kernel: ata2: SATA link down (SStatus 0 SControl 300) May 7 09:27:29 Kyber kernel: ata2: hard resetting link May 7 09:27:34 Kyber kernel: ata2: found unknown device (class 0) May 7 09:27:39 Kyber kernel: ata2: softreset failed (device not ready) May 7 09:27:39 Kyber kernel: ata2: hard resetting link May 7 09:27:43 Kyber kernel: ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 7 09:27:43 Kyber kernel: ata2.00: configured for UDMA/133 May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#0 Sense Key : 0x5 [current] May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#0 ASC=0x21 ASCQ=0x4 May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#0 CDB: opcode=0x88 88 00 00 00 00 04 c3 93 2a 00 00 00 05 40 00 00 May 7 09:27:43 Kyber kernel: I/O error, dev sdb, sector 20461070848 op 0x0:(READ) flags 0x84700 phys_seg 168 prio class 0 May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 Sense Key : 0x5 [current] May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 ASC=0x21 ASCQ=0x4 May 7 09:27:43 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 CDB: opcode=0x88 88 00 00 00 00 04 c3 93 2f 40 00 00 04 c0 00 00 May 7 09:27:43 Kyber kernel: I/O error, dev sdb, sector 20461072192 op 0x0:(READ) flags 0x80700 phys_seg 152 prio class 0 May 7 09:27:43 Kyber kernel: ata2: EH complete May 7 10:28:16 Kyber kernel: ata2.00: exception Emask 0x10 SAct 0xe SErr 0x90202 action 0xe frozen May 7 10:28:16 Kyber kernel: ata2.00: irq_stat 0x00400000, PHY RDY changed May 7 10:28:16 Kyber kernel: ata2: SError: { RecovComm Persist PHYRdyChg 10B8B } May 7 10:28:16 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 10:28:16 Kyber kernel: ata2.00: cmd 60/30:08:00:4a:8a/09:00:28:05:00/40 tag 1 ncq dma 1204224 in May 7 10:28:16 Kyber kernel: ata2.00: status: { DRDY } May 7 10:28:16 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 10:28:16 Kyber kernel: ata2.00: cmd 60/d0:10:30:53:8a/00:00:28:05:00/40 tag 2 ncq dma 106496 in May 7 10:28:16 Kyber kernel: ata2.00: status: { DRDY } May 7 10:28:16 Kyber kernel: ata2.00: failed command: READ FPDMA QUEUED May 7 10:28:16 Kyber kernel: ata2.00: cmd 60/00:18:00:54:8a/06:00:28:05:00/40 tag 3 ncq dma 786432 in May 7 10:28:16 Kyber kernel: ata2.00: status: { DRDY } May 7 10:28:16 Kyber kernel: ata2: hard resetting link May 7 10:28:18 Kyber kernel: ata2: found unknown device (class 0) May 7 10:28:18 Kyber kernel: ata2: SATA link down (SStatus 0 SControl 300) May 7 10:28:18 Kyber kernel: ata2: hard resetting link May 7 10:28:24 Kyber kernel: ata2: found unknown device (class 0) May 7 10:28:28 Kyber kernel: ata2: softreset failed (device not ready) May 7 10:28:28 Kyber kernel: ata2: hard resetting link May 7 10:28:33 Kyber kernel: ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 7 10:28:33 Kyber kernel: ata2.00: configured for UDMA/133 May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 Sense Key : 0x5 [current] May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 ASC=0x21 ASCQ=0x4 May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#1 CDB: opcode=0x88 88 00 00 00 00 05 28 8a 4a 00 00 00 09 30 00 00 May 7 10:28:33 Kyber kernel: I/O error, dev sdb, sector 22154988032 op 0x0:(READ) flags 0x84700 phys_seg 168 prio class 0 May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 Sense Key : 0x5 [current] May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 ASC=0x21 ASCQ=0x4 May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#2 CDB: opcode=0x88 88 00 00 00 00 05 28 8a 53 30 00 00 00 d0 00 00 May 7 10:28:33 Kyber kernel: I/O error, dev sdb, sector 22154990384 op 0x0:(READ) flags 0x80700 phys_seg 13 prio class 0 May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=16s May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 Sense Key : 0x5 [current] May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 ASC=0x21 ASCQ=0x4 May 7 10:28:33 Kyber kernel: sd 2:0:0:0: [sdb] tag#3 CDB: opcode=0x88 88 00 00 00 00 05 28 8a 54 00 00 00 06 00 00 00 May 7 10:28:33 Kyber kernel: I/O error, dev sdb, sector 22154990592 op 0x0:(READ) flags 0x80700 phys_seg 97 prio class 0 May 7 10:28:33 Kyber kernel: ata2: EH complete kyber-diagnostics-20230507-1110.zip
May 8, 20233 yr Community Expert You could try swapping the parity drives with other disks in the HBA.
May 8, 20233 yr Author Thanks for the reply! Yeah swapping drives in and out with the HBA has been something I tried. Whenever a drive is connected to the HBA, it works fine. I've not had any of my drives display this kind of behavior when connected to the HBA controller. It's only been with the motherboard controller. Up until a few months ago, I was having trouble with the drives having all kinds of parity check errors, but I finally figured out it was the motherboard connected parity drives and that it didn't like Seagate drives for whatever reason. After switching to WesternDigital, the problem went away. That's why I'm thinking the problem might be the motherboard and compatibility. I'm still using WD drives, but the 20tb range just has not been cooperating. I've had 6 drives with this issue, so I know it's practically impossible for it to still be a DRIVE problem. I'm about to try updating the BIOS, since there has been numerous updates. Although nothing mentioned in the update notes indicates a fix to this type of problem, it's worth the shot right? If this doesn't fix the problem, I'm very tempted to try and get a different controller card and just not use the motherboard connectors. **** EDIT After an hour of tinkering, I've got the latest BIOS installed. Rebuilding parity now.. We'll see. (fingers crossed) Edited May 8, 20233 yr by hansolo77
May 9, 20233 yr Community Expert 10 hours ago, hansolo77 said: I'm very tempted to try and get a different controller card and just not use the motherboard connectors. That's probably a good idea, there have been some issues with Linux and the Ryzen onboard SATA controllers, usually under load.
May 9, 20233 yr Author 15 hours ago, JorgeB said: usually under load. Funny you should mention that. The more I think about it, the more I think that’s when I’ve been having trouble. Here I thought building a system with a Ryzen processor that had 12 cores would be awesome. If I ever build another machine, I think I’ll stick with tried and true Intel. lol So far so good, no issues. Only 33% complete though. **** EDIT Seems I spoke to soon. About an hour after I posted this, I started getting errors again. Grr it's so frustrating. Spoiler May 8 18:58:42 Kyber kernel: ata5: SATA max UDMA/133 abar m2048@0xfba00000 port 0xfba00100 irq 82 May 8 18:58:42 Kyber kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 8 18:58:42 Kyber kernel: ata5.00: ATA-12: WDC WD201KFGX-68BKJN0, 83.00A83, max UDMA/133 May 8 18:58:42 Kyber kernel: ata5.00: 39063650304 sectors, multi 16: LBA48 NCQ (depth 32), AA May 8 18:58:42 Kyber kernel: ata5.00: Features: NCQ-sndrcv NCQ-prio May 8 18:58:42 Kyber kernel: ata5.00: configured for UDMA/133 May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] 39063650304 512-byte logical blocks: (20.0 TB/18.2 TiB) May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] 4096-byte physical blocks May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] Write Protect is off May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] Mode Sense: 00 3a 00 00 May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] Preferred minimum I/O size 4096 bytes May 8 18:58:42 Kyber kernel: sdz: sdz1 May 8 18:58:42 Kyber kernel: sd 5:0:0:0: [sdz] Attached SCSI removable disk May 8 18:59:03 Kyber emhttpd: WDC_WD201KFGX-68BKJN0_8LG7TDXE (sdz) 512 39063650304 May 8 18:59:03 Kyber emhttpd: read SMART /dev/sdz May 8 18:59:22 Kyber unassigned.devices: Partition 'sdz1' does not have a file system and cannot be mounted. May 8 19:02:02 Kyber emhttpd: read SMART /dev/sdz May 8 19:02:44 Kyber emhttpd: WDC_WD201KFGX-68BKJN0_8LG7TDXE (sdz) 512 39063650304 May 8 19:02:45 Kyber kernel: mdcmd (30): import 29 sdz 64 19531825100 1 WDC_WD201KFGX-68BKJN0_8LG7TDXE May 8 19:02:45 Kyber kernel: md: import disk29: (sdz) WDC_WD201KFGX-68BKJN0_8LG7TDXE size: 19531825100 erased May 8 19:02:45 Kyber emhttpd: read SMART /dev/sdz May 8 19:02:50 Kyber emhttpd: WDC_WD201KFGX-68BKJN0_8LG7TDXE (sdz) 512 39063650304 May 8 19:02:50 Kyber kernel: mdcmd (30): import 29 sdz 64 19531825100 1 WDC_WD201KFGX-68BKJN0_8LG7TDXE May 8 19:02:50 Kyber kernel: md: import disk29: (sdz) WDC_WD201KFGX-68BKJN0_8LG7TDXE size: 19531825100 erased May 8 19:02:50 Kyber emhttpd: read SMART /dev/sdz May 8 19:03:13 Kyber emhttpd: writing GPT on disk (sdz), with partition 1 byte offset 32KiB, erased: 0 May 8 19:03:13 Kyber emhttpd: shcmd (353): sgdisk -Z /dev/sdz May 8 19:03:13 Kyber kernel: sdz: sdz1 May 8 19:03:14 Kyber emhttpd: shcmd (354): sgdisk -o -a 8 -n 1:32K:0 /dev/sdz May 8 19:03:14 Kyber kernel: sdz: sdz1 May 8 19:03:15 Kyber kernel: sdz: sdz1 May 9 12:51:48 Kyber kernel: ata5: found unknown device (class 0) May 9 12:51:48 Kyber kernel: ata5: SATA link down (SStatus 0 SControl 300) May 9 12:51:54 Kyber kernel: ata5: found unknown device (class 0) May 9 12:51:59 Kyber kernel: ata5: softreset failed (device not ready) May 9 12:52:03 Kyber kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 9 12:52:03 Kyber kernel: ata5.00: configured for UDMA/133 May 9 13:34:15 Kyber kernel: ata5.00: exception Emask 0x10 SAct 0x800000 SErr 0x90202 action 0xe frozen May 9 13:34:15 Kyber kernel: ata5.00: irq_stat 0x00400000, PHY RDY changed May 9 13:34:15 Kyber kernel: ata5: SError: { RecovComm Persist PHYRdyChg 10B8B } May 9 13:34:15 Kyber kernel: ata5.00: failed command: WRITE FPDMA QUEUED May 9 13:34:15 Kyber kernel: ata5.00: cmd 61/28:b8:30:27:be/01:00:a5:03:00/40 tag 23 ncq dma 151552 out May 9 13:34:15 Kyber kernel: ata5.00: status: { DRDY } May 9 13:34:15 Kyber kernel: ata5: hard resetting link May 9 13:34:17 Kyber kernel: ata5: found unknown device (class 0) May 9 13:34:17 Kyber kernel: ata5: SATA link down (SStatus 0 SControl 300) May 9 13:34:17 Kyber kernel: ata5: hard resetting link May 9 13:34:22 Kyber kernel: ata5: found unknown device (class 0) May 9 13:34:27 Kyber kernel: ata5: softreset failed (device not ready) May 9 13:34:27 Kyber kernel: ata5: hard resetting link May 9 13:34:32 Kyber kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 9 13:34:32 Kyber kernel: ata5.00: configured for UDMA/133 May 9 13:34:32 Kyber kernel: ata5: EH complete May 9 13:35:23 Kyber kernel: ata5.00: exception Emask 0x10 SAct 0x2 SErr 0x90202 action 0xe frozen May 9 13:35:23 Kyber kernel: ata5.00: irq_stat 0x00400000, PHY RDY changed May 9 13:35:23 Kyber kernel: ata5: SError: { RecovComm Persist PHYRdyChg 10B8B } May 9 13:35:23 Kyber kernel: ata5.00: failed command: WRITE FPDMA QUEUED May 9 13:35:23 Kyber kernel: ata5.00: cmd 61/00:08:88:44:c5/04:00:a5:03:00/40 tag 1 ncq dma 524288 out May 9 13:35:23 Kyber kernel: ata5.00: status: { DRDY } May 9 13:35:23 Kyber kernel: ata5: hard resetting link May 9 13:35:25 Kyber kernel: ata5: found unknown device (class 0) May 9 13:35:25 Kyber kernel: ata5: SATA link down (SStatus 0 SControl 300) May 9 13:35:25 Kyber kernel: ata5: hard resetting link May 9 13:35:30 Kyber kernel: ata5: found unknown device (class 0) May 9 13:35:35 Kyber kernel: ata5: softreset failed (device not ready) May 9 13:35:35 Kyber kernel: ata5: hard resetting link May 9 13:35:40 Kyber kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 9 13:35:40 Kyber kernel: ata5.00: configured for UDMA/133 May 9 13:35:40 Kyber kernel: ata5: EH complete May 9 15:03:20 Kyber kernel: ata5.00: exception Emask 0x10 SAct 0x200000 SErr 0x90202 action 0xe frozen May 9 15:03:20 Kyber kernel: ata5.00: irq_stat 0x00400000, PHY RDY changed May 9 15:03:20 Kyber kernel: ata5: SError: { RecovComm Persist PHYRdyChg 10B8B } May 9 15:03:20 Kyber kernel: ata5.00: failed command: WRITE FPDMA QUEUED May 9 15:03:20 Kyber kernel: ata5.00: cmd 61/f0:a8:b0:f4:6b/02:00:ed:03:00/40 tag 21 ncq dma 385024 out May 9 15:03:20 Kyber kernel: ata5.00: status: { DRDY } May 9 15:03:20 Kyber kernel: ata5: hard resetting link May 9 15:03:22 Kyber kernel: ata5: found unknown device (class 0) May 9 15:03:22 Kyber kernel: ata5: SATA link down (SStatus 0 SControl 300) May 9 15:03:23 Kyber kernel: ata5: hard resetting link May 9 15:03:28 Kyber kernel: ata5: found unknown device (class 0) May 9 15:03:33 Kyber kernel: ata5: softreset failed (device not ready) May 9 15:03:33 Kyber kernel: ata5: hard resetting link May 9 15:03:37 Kyber kernel: ata5: SATA link up 6.0 Gbps (SStatus 133 SControl 300) May 9 15:03:37 Kyber kernel: ata5.00: configured for UDMA/133 May 9 15:03:37 Kyber kernel: ata5: EH complete ** Press ANY KEY to close this window ** Edited May 9, 20233 yr by hansolo77
May 11, 20233 yr Author Quick update, got the new controller installed, parity check is running. Just for reference, the controller card is this: https://www.amazon.com/dp/B07M93P18S?psc=1&ref=ppx_yo2ov_dt_b_product_details Also.. started running the parity check last night around 7:30pm. Woke up this morning to check and the drives are all still good. Got home from work about an hour ago and just now checked. Drives are still good, showing no signs of resetting like it had been before. Unfortunately, they do have some Parity errors that occurred after midnight. The errors are from Unraid and not from the Drive logs, so that should indicate a good sign that the controller is working well; I would have seen resets by now. I'm running the check "uncorrecting". I suspect there would be issues from the drive resetting when they were first being built for parity. So I'm going to try and start over with them connected to the new controller (stop array, remove parity, start array, stop array, add parity, start array, parity rebuilds). Unless this isn't a good idea? I don't think correcting parity bits with a "dirty" drive is a bad idea.. would likely corrupt data more than protect it. Thoughts?
May 11, 20233 yr From my experience, it likely PSU issue or disk power cabling relate, problem likely happen when all disks in loading but no problem when part of disk loading. I will review / renew disk power cable first. You almost change every hardware. Edited May 11, 20233 yr by Vr2Io
May 12, 20233 yr Author Solution 1 hour ago, Vr2Io said: From my experience, it likely PSU issue or disk power cabling relate, problem likely happen when all disks in loading but no problem when part of disk loading. I will review / renew disk power cable first. You almost change every hardware. I appreciate your feedback. Unfortunately there's not anything I can do about the power. The setup of my server uses 2x server grade PSU's that combine into a PDU. That PDU only has so many cables on it. In fact (and you may feel like this is an ah-ha! moment) I had to use a splitter to get 2 SATA power connector's out of one. I've previously connected various drives (2x8tb, then 2x12tb, and now 2x20tb) to this primary cable. Also, although not mentioned, I have tried switching cables (first to a 4-pin MOLEX splitter with 4 in and 4 out, and then to a SATA splitter with 4 in and SATA out) but had no change in behavior. This has definitely been an eye-opening experience for me. I've been building computers since the 90's. I pretty much know everything on the hardware-side to try. I even at one point tried having the drives OUTSIDE the case, just in case the problem was something possibly "magnetic" because they're mounted on the side of the power supply (see the pictures in my build). At this point I don't think it's hardware. I take that back though. It looks more and more like the problem is with Ryzen and Linux. I made a post (mentioned in the OP of this thread) on the motherboard's manufacturing forum about the issue I'm having. The general consensus was to upgrade the BIOS. I did that, then started to rebuild parity again. It made no difference at all. I still ended up getting hard and soft resets (although the drives never disabled). My last ditch effort was to install a separate SATA controller. With it connected, I've not had ANY further reset issues. I started a parity check last night and noticed it having errors so I'm rebuilding a clean parity now using the new controller (the last parity was built on the motherboard controller with drives resetting, so hopefully this will solve everything).
May 12, 20233 yr 6 hours ago, hansolo77 said: I started a parity check last night and noticed it having errors so I'm rebuilding a clean parity now using the new controller (the last parity was built on the motherboard controller with drives resetting, so hopefully this will solve everything). I miss you got a new controller ASM1064 and seems great now. 6 hours ago, hansolo77 said: It looks more and more like the problem is with Ryzen and Linux. My main build all disks under HBA, but some small build use AMD / Intel onboard SATA controller haven't problem at all, at least no problem with B450, X370. PS : A newly 5600H ( Morefine S500 ) put in service.
May 21, 20233 yr Author Seems like I still have the problem, unfortunately. Future build plan is to not use Ryzen again. I'll just have to live with it. On the plus side, the resets/etc haven't caused Unraid to completely disable the drives like it did in the past. So maybe that's a good sign. I'm still concerned about it though. Oh well. Guess AMD and MSI didn't plan on having 20tb drives when they designed their components.
July 15, 20232 yr Author Just wanted to post a quick update (hopefully not jinxing). Firstly, I "had the last straw" and finally decided to move out of my dad's place and get my own. It's been a month or so that I've been doing the move. I'm done now, and decided now's as good a time as any to try and do some spring cleaning and solve the issue once and for good. While I didn't do the ULTIMATE FIX, which would involve a complete rebuild of hardware, I did accomplish a FIX just the same. The solution now is to not use ANY SATA controllers. The motherboard just doesn't seem compatible with any of my new 20tb drives. No matter what I tried, from replacing cables to replacing drives (yes, I even RMA'd a drive hoping it would fix the problem), nothing worked. I even tried that expansion card for a while, but it had the same results after awhile. I just gave up. I pulled the two bottom right drives from my case (drives 23 and 24) and put the parity drives in there. I did it properly though. I used `krusader` and moved all the data off those drives before removing them. Once removed from the array, I crossed my fingers then removed the parity drives from the array. After a "no guts no glory" reboot, I then put the parity drives in and rebuilt the array parity. I didn't have a single issue. And since then I've done 2 forced parity checks and the July 1st auto check all came back perfect. So now I'm relying entirely on the array's HBA controller. So far so good. Here's hoping it stays that way.
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.