CuriousMind Posted February 27 Share Posted February 27 I have a disk in my array that I am trying to replace due to it racking up enough SMART errors that it is being disabled. I purchased a couple of Recertified Drives from Seagate and ran preclear on them one at a time. My issue is that both of them failed preclear, so I am wondering what are the odds that they BOTH are bad? My case has limited space so I utilized the same power and SATA connection for both, swapping in the second drive after the first one failed. I did some searching of the interwebs and the forums for the errors, and it SEEMS like they might just be bad, but I was hoping for a second opinion before I go through reaching out to the vendor for a replacement. Preclear logs attached Thanks for any help! preclear_disk_ZHZ3CPEG_8792.txt Quote Link to comment
JorgeB Posted February 27 Share Posted February 27 Don't see the reason for any error there, if you didn't reboot yet post the diagnostics Quote Link to comment
CuriousMind Posted February 27 Author Share Posted February 27 23 minutes ago, JorgeB said: Don't see the reason for any error there, if you didn't reboot yet post the diagnostics Sure thing, please see that attached. Thanks tower-diagnostics-20240227-0637.zip Quote Link to comment
JorgeB Posted February 27 Share Posted February 27 Is this the disk you are trying to preclear? Feb 26 15:11:49 Tower emhttpd: ST12000NM0538-2K2101_ZHZ3CPEG (sdi) 512 23437770752 In any case there are constant errors with it: Feb 26 15:21:12 Tower kernel: I/O error, dev sdi, sector 119069616 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 3 Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 Sense Key : 0x2 [current] Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 ASC=0x4 ASCQ=0x0 Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 CDB: opcode=0x88 88 00 00 00 00 00 07 18 db b0 00 00 00 08 00 00 Feb 26 15:21:12 Tower kernel: I/O error, dev sdi, sector 119069616 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 3 Feb 26 15:21:12 Tower kernel: Buffer I/O error on dev sdi, logical block 14883702, async page read Check/replace cables, you also need to check filesystem on the disabled disk. Quote Link to comment
CuriousMind Posted February 27 Author Share Posted February 27 1 hour ago, JorgeB said: Is this the disk you are trying to preclear? Feb 26 15:11:49 Tower emhttpd: ST12000NM0538-2K2101_ZHZ3CPEG (sdi) 512 23437770752 In any case there are constant errors with it: Feb 26 15:21:12 Tower kernel: I/O error, dev sdi, sector 119069616 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 3 Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 Sense Key : 0x2 [current] Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 ASC=0x4 ASCQ=0x0 Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 CDB: opcode=0x88 88 00 00 00 00 00 07 18 db b0 00 00 00 08 00 00 Feb 26 15:21:12 Tower kernel: I/O error, dev sdi, sector 119069616 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 3 Feb 26 15:21:12 Tower kernel: Buffer I/O error on dev sdi, logical block 14883702, async page read Check/replace cables, you also need to check filesystem on the disabled disk. I've replaced the power and SATA cables twice, so I'm not sure if that could mean an issue with the board connections. The connections themselves are firmly seated. I'll try the file system check tonight Quote Link to comment
CuriousMind Posted March 2 Author Share Posted March 2 On 2/27/2024 at 6:47 AM, JorgeB said: Is this the disk you are trying to preclear? Feb 26 15:11:49 Tower emhttpd: ST12000NM0538-2K2101_ZHZ3CPEG (sdi) 512 23437770752 In any case there are constant errors with it: Feb 26 15:21:12 Tower kernel: I/O error, dev sdi, sector 119069616 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 3 Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 Sense Key : 0x2 [current] Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 ASC=0x4 ASCQ=0x0 Feb 26 15:21:12 Tower kernel: sd 1:0:4:0: [sdi] tag#2824 CDB: opcode=0x88 88 00 00 00 00 00 07 18 db b0 00 00 00 08 00 00 Feb 26 15:21:12 Tower kernel: I/O error, dev sdi, sector 119069616 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 3 Feb 26 15:21:12 Tower kernel: Buffer I/O error on dev sdi, logical block 14883702, async page read Check/replace cables, you also need to check filesystem on the disabled disk. Well I ran the filesystem check, and the system still immediately disables that disk when I try and mount it. I was able to switch to the second of the new drives to a different SATA and power cable (this would be the third one) and it cleared pre-clear just fine. Of course as soon as I swapped that drive in and started the parity rebuild, I started getting a constant spin-up-chirp-spin-down loop on what has to be either that drive or the parity somehow. I cannot believe that I would have this many bad drives, so am I looking at a bad PSU altogether or something else? SMART looks fine for all disks, system logs look fine, and the disk logs are clear. I stopped the parity sync because I am worried I am going to wreck an otherwise good drive. tower-diagnostics-20240301-2114.zip Quote Link to comment
JorgeB Posted March 2 Share Posted March 2 7 hours ago, CuriousMind said: so am I looking at a bad PSU altogether or something else? It's possible, or a bad power connection, any splitters in use? Quote Link to comment
CuriousMind Posted March 2 Author Share Posted March 2 1 hour ago, JorgeB said: It's possible, or a bad power connection, any splitters in use? Yes, I have a 4 way power splitter set up on both of my hard drive carriages. The parity used to be on a direct connection instead, but I was trying to figure out if it could be a bad power cable (not splitter) so I swapped that side to another brand new splitter. Quote Link to comment
JorgeB Posted March 2 Share Posted March 2 1 minute ago, CuriousMind said: I have a 4 way power Is that molex to SATA or SATA to SATA? For SATA to SATA never recommend using more that a 2 way splitter, and even with molex it's not great. Quote Link to comment
CuriousMind Posted March 2 Author Share Posted March 2 SATA to SATA. I didn't know about that, but with two carriages that hold 4 drives each, I'm not sure how else to make it all work. Quote Link to comment
Solution itimpi Posted March 2 Solution Share Posted March 2 2 hours ago, CuriousMind said: SATA to SATA. I didn't know about that, but with two carriages that hold 4 drives each, I'm not sure how else to make it all work. Do you have any spare Molex connectors on the PSU so you could instead use some Molex->SATA splitters? Quote Link to comment
CuriousMind Posted March 2 Author Share Posted March 2 I'll see if I cannot purchase some and see if that makes a difference. Fingers crossed Quote Link to comment
CuriousMind Posted March 11 Author Share Posted March 11 On 3/2/2024 at 9:15 AM, itimpi said: Do you have any spare Molex connectors on the PSU so you could instead use some Molex->SATA splitters? Thank you. After removing the failing drive, and then changing to a Molex->SATA splitter, things have (thank God finally) resolved. Quote Link to comment
Recommended Posts
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.