linksta7 Posted December 20, 2023 Share Posted December 20, 2023 9 minutes ago, itimpi said: You must NOT do this as it destroys the pre-clear signature. 2 minutes ago, dlandon said: No. Yes, but you need to use the "Clear Disk" option so the pre-clear signature is written to the disk. When Unraid sees the signature, it will not clear the disk which has already been done by the pre-clear plugiun. The normal steps are: Pre-clear the disk using "Clear Disk". Assign the disk to the Array. Start Array and you will be asked to confirm formatting the disk to the file format you've specified as either the default or changed for this particular disk. I see, glad I checked then. I'll make sure to just let Unraid do its thing when its building the array then. Thanks for making it simple and clear. One last is, now that I've already started erase and clear process last now (all disks are now in erase step), is it ok to stop the process and re-run by just using clear disk? Or should I just let them run the rest of the erase and clear process? Quote Link to comment
dlandon Posted December 20, 2023 Author Share Posted December 20, 2023 35 minutes ago, linksta7 said: One last is, now that I've already started erase and clear process last now (all disks are now in erase step), is it ok to stop the process and re-run by just using clear disk? Or should I just let them run the rest of the erase and clear process? I would stop all the preclears and run the "Clear Disk" as Unraid will do the equivalent when you install disks not pre-cleared this way. 1 Quote Link to comment
linksta7 Posted December 20, 2023 Share Posted December 20, 2023 25 minutes ago, dlandon said: I would stop all the preclears and run the "Clear Disk" as Unraid will do the equivalent when you install disks not pre-cleared this way. Ok thank you, appreciate your help Quote Link to comment
linksta7 Posted December 23, 2023 Share Posted December 23, 2023 On 12/20/2023 at 11:16 AM, dlandon said: No. Yes, but you need to use the "Clear Disk" option so the pre-clear signature is written to the disk. When Unraid sees the signature, it will not clear the disk which has already been done by the pre-clear plugiun. The normal steps are: Pre-clear the disk using "Clear Disk". Assign the disk to the Array. Start Array and you will be asked to confirm formatting the disk to the file format you've specified as either the default or changed for this particular disk. So my preclear finished this morning. I went ahead and assigned the disks to the array, explicitly specified xfs as the data drives format, and clicked start array. I wasn't prompted to confirm formatting the drives. It appears that it still wants me to format, but you guys made it pretty clear that I should not explicitly format the drives via the format button or it will erase the preclear signature. Should the drives appear as formatted after the parity sync is complete? As it stands, it doesn't appear as though they will be formatted. Quote Link to comment
dlandon Posted December 23, 2023 Author Share Posted December 23, 2023 Are they all new disks? Quote Link to comment
linksta7 Posted December 23, 2023 Share Posted December 23, 2023 (edited) 25 minutes ago, dlandon said: Are they all new disks? Yea all new disks. Three of them were new in packaging (had zero power on hours) and one of them (a 6tb data drive) was taken from a prebuilt (5 power on hours iirc) Edited December 23, 2023 by linksta7 additional info Quote Link to comment
linksta7 Posted December 23, 2023 Share Posted December 23, 2023 Diagnostics if needed chowdertower-diagnostics-20231223-1327.zip Quote Link to comment
itimpi Posted December 23, 2023 Share Posted December 23, 2023 1 hour ago, linksta7 said: format, but you guys made it pretty clear that I should not explicitly format the drives via the format button or it will erase the preclear signature. Should the drives appear as formatted after the parity sync is complete? As it stands, it doesn't appear as though they will be formatted It is fine to format them AFTER adding them to the array (and is in fact required). What you must not do (as many users seem to do) is format them in UD after the Preclear but BEFORE adding them to the array. Quote Link to comment
linksta7 Posted December 23, 2023 Share Posted December 23, 2023 1 hour ago, itimpi said: It is fine to format them AFTER adding them to the array (and is in fact required). What you must not do (as many users seem to do) is format them in UD after the Preclear but BEFORE adding them to the array. Ah I see, I thought the formatting would take place as part of the array building process. I will format then now that preclear is complete and all the disks have been added to the array then. Thank you Quote Link to comment
trurl Posted December 24, 2023 Share Posted December 24, 2023 A clear disk is all zeros. When you add a clear disk to a new slot in the array that already has valid parity, parity is still valid since all zeros has no effect on parity. If you format a clear disk before adding it to the array, it is no longer all zeros, so Unraid has to clear it again so parity will remain valid. Format is a write operation, it writes an empty filesystem to the disk. Unraid treats this write operation in the array just as it does any other, by updating parity. So after formatting a disk in the array, parity is in sync with the formatted disk. Please keep this in mind and don't format a disk in the array if it had data on it you wish to keep. Ask for help instead. Quote Link to comment
opiekeith Posted December 27, 2023 Share Posted December 27, 2023 (edited) Running preclear on a disk, and got the following warnings/errors attached. This is the second drive that I have gotten. I am about to order a new SFF-8087 to SFF-8482 cable. I did just order a molex to SATA power cable so I don't have to cover the 3.3v pins up, to rule that out. Here is a diagnostics also. hades-diagnostics-20231227-0032.zip preclear_disk_segate-hdd.txt Edited December 27, 2023 by opiekeith Quote Link to comment
dlandon Posted December 27, 2023 Author Share Posted December 27, 2023 5 hours ago, opiekeith said: Running preclear on a disk, and got the following warnings/errors attached. The dd files are written to the tmpfs ram file system. This can be a memory issue. Please run a memory test. Quote Link to comment
opiekeith Posted December 27, 2023 Share Posted December 27, 2023 (edited) On 12/27/2023 at 5:47 AM, dlandon said: The dd files are written to the tmpfs ram file system. This can be a memory issue. Please run a memory test. I got a pass, no errors. This is a pretty old box. DDR3-1333 memory. I was using the memory test in the unraid menu. If there is another that I need to run, jlmk. About done with the second pass with no errors. I let it run for 4 passes without any failures. Edited December 28, 2023 by opiekeith Quote Link to comment
opiekeith Posted December 28, 2023 Share Posted December 28, 2023 Switched out the data cable and running preclear again. Still getting the same error after about 4 and a half min. It does this 5 times, then fails. Dec 28 11:45:28 preclear_disk_ZAD201L30000C75286YQ_11826: Preclear Disk Version: 1.0.29 Dec 28 11:45:29 preclear_disk_ZAD201L30000C75286YQ_11826: Disk size: 6001175126016 Dec 28 11:45:29 preclear_disk_ZAD201L30000C75286YQ_11826: Disk blocks: 1465130646 Dec 28 11:45:29 preclear_disk_ZAD201L30000C75286YQ_11826: Blocks (512 bytes): 11721045168 Dec 28 11:45:29 preclear_disk_ZAD201L30000C75286YQ_11826: Block size: 4096 Dec 28 11:45:30 preclear_disk_ZAD201L30000C75286YQ_11826: Start sector: 0 Dec 28 11:45:32 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-read: pre-read verification started 1 of 5 retries... Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 248.999 s, 8.4 kB/s Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.684 s, 8.4 kB/s Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:57 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.7 s, 8.4 kB/s Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.718 s, 8.4 kB/s Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.736 s, 8.4 kB/s Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.753 s, 8.4 kB/s Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.771 s, 8.4 kB/s Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: dd: error reading '/dev/sdc': Input/output error Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records in Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 1+0 records out Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-Read: dd output: 2097152 bytes (2.1 MB, 2.0 MiB) copied, 249.788 s, 8.4 kB/s Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: dd process hung at 2097152, killing ... Dec 28 11:49:58 preclear_disk_ZAD201L30000C75286YQ_11826: Pre-read: pre-read verification started 2 of 5 retries... I should have my 4pin molex to sata power cable to put in tomorrow to rule out my taping the 3.3v pin on the new drives. That's the only thing different about my new drive vs my older/other ones. They don't need the 3.3v disabled. Quote Link to comment
opiekeith Posted December 29, 2023 Share Posted December 29, 2023 (edited) New Molex to SATA cable installed and doing the same as before. Same errors on the initial read. This is my second drive and don't want to submit another return if the drive is not the issue. To recap, I have replaced the data cable, and the power cable for the drive, and I still get the same error. Edited December 29, 2023 by opiekeith Quote Link to comment
dlandon Posted December 29, 2023 Author Share Posted December 29, 2023 12 minutes ago, opiekeith said: New Molex to SATA cable installed and doing the same as before. Same errors on the initial read. This is my second drive and don't want to submit another return if the drive is not the issue. To recap, I have replaced the data cable, and the power cable for the drive, and I still get the same error. Looking back at the diagnosics you originally posted, I see disk errors: Dec 26 21:40:02 Hades kernel: mpt2sas_cm0: log_info(0x3112043b): originator(PL), code(0x12), sub_code(0x043b) Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2531 UNKNOWN(0x2003) Result: hostbyte=0x05 driverbyte=DRIVER_OK cmd_age=0s Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2531 Sense Key : 0x5 [current] Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2531 ASC=0x10 ASCQ=0x3 Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2531 CDB: opcode=0x7f, sa=0x9 Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2531 CDB[00]: 7f 00 00 00 00 00 00 18 00 09 20 00 00 00 00 00 Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2531 CDB[10]: 01 71 80 00 01 71 80 00 00 00 00 00 00 00 00 08 Dec 26 21:40:02 Hades kernel: protection error, dev sdc, sector 24215552 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 3 Dec 26 21:40:02 Hades kernel: mpt2sas_cm0: log_info(0x3112043b): originator(PL), code(0x12), sub_code(0x043b) Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2532 UNKNOWN(0x2003) Result: hostbyte=0x05 driverbyte=DRIVER_OK cmd_age=0s Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2532 Sense Key : 0x5 [current] Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2532 ASC=0x10 ASCQ=0x3 Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2532 CDB: opcode=0x7f, sa=0x9 Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2532 CDB[00]: 7f 00 00 00 00 00 00 18 00 09 20 00 00 00 00 00 Dec 26 21:40:02 Hades kernel: sd 7:0:1:0: [sdc] tag#2532 CDB[10]: 01 71 80 00 01 71 80 00 00 00 00 00 00 00 00 08 Dec 26 21:40:02 Hades kernel: protection error, dev sdc, sector 24215552 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 3 Dec 26 21:40:02 Hades kernel: Buffer I/O error on dev sdc, logical block 3026944, async page read This shows in the SMART report: Error counter log: Errors Corrected by Total Correction Gigabytes Total ECC rereads/ errors algorithm processed uncorrected fast | delayed rewrites corrected invocations [10^9 bytes] errors read: 3859853718 0 0 3859853718 0 393941.617 0 write: 0 0 0 0 0 70285.848 0 verify: 3160 0 0 3160 0 0.000 0 Non-medium error count: 98 Maybe one of the disk experts here can lend a hand on what might be the issue. Quote Link to comment
JorgeB Posted December 30, 2023 Share Posted December 30, 2023 15 hours ago, opiekeith said: Same errors on the initial read. Formatted with type 2 protection 8 bytes of protection information per logical block You need to remove this: https://forums.unraid.net/topic/93432-parity-disk-read-errors/?do=findComment&comment=864078 Quote Link to comment
opiekeith Posted December 30, 2023 Share Posted December 30, 2023 I ended up just adding it to the cluster. It rebuilt without errors. I know it was not ideal and I would have preferred to have done a pre clear but my unraid was down for longer than I wanted. If I run into more errors I will remove it from the cluster and run a pre clear with formatting with type 1 to see if that fixes the issue. I will let this soak for a few days before I start a motherboard/cpu upgrade on my unraid box. Quote Link to comment
JorgeB Posted December 31, 2023 Share Posted December 31, 2023 16 hours ago, opiekeith said: It rebuilt without errors. I believe a rebuild doesn't remove the protection, can you post new diags? Quote Link to comment
opiekeith Posted December 31, 2023 Share Posted December 31, 2023 5 hours ago, JorgeB said: I believe a rebuild doesn't remove the protection, can you post new diags? I did not think it did, but I went under the assumption based on my readings that newer version of unraid don't necessarily need to have drives precleared. It does some checks during the rebuild process. Thats why I just went ahead and did the rebuild. See attached for the latest diagnostics. It seems I am going to have to buy another drive to replace another failing drive so I will get a chance to run preclear on that one when I get it in and replaced. Then one day I will be able to replace my MB/CPU and add a new cache drive, one day. hades-diagnostics-20231231-0954.zip Quote Link to comment
JorgeB Posted January 1 Share Posted January 1 18 hours ago, opiekeith said: but I went under the assumption based on my readings that newer version of unraid don't necessarily need to have drives precleared. Preclear is always optional, but the disk still has the type2 format protection enabled, and it's still causing errors, you need to remove that, and I believe the disk need to be re-formatted to do that, so then you will need to repeat the rebuild after. Quote Link to comment
opiekeith Posted January 1 Share Posted January 1 (edited) ugh, started the format noticed it was going to take awhile, and when I came back to check on it, screen was gone and disk is now missing and is not showing up anymore. Restarted, and it does not even show up during the boot when the HBA starts scanning connected drives. Attached is the latest diag. The drive was /dev/sdc, and this was the command I ran to format it. sg_format --format --fmtpinfo=0 /dev/sdc --Keith hades-diagnostics-20240101-1718.zip Edited January 2 by opiekeith Quote Link to comment
JorgeB Posted January 2 Share Posted January 2 8 hours ago, opiekeith said: Restarted, and it does not even show up during the boot Try power cycling the server, instead of just rebooting. Quote Link to comment
opiekeith Posted January 2 Share Posted January 2 (edited) No go, I even tried switching the cables and still nothing. On another thread that I am asking about this, some one mentioned to "LSI HBA and retry the format on the /dev/sg[0-9] SCSI generic device" but not sure what the LSI HBA means. and lsblk, show s sdc as: sdc 8:32 0 0B 0 disk and it does show under SCSI Devices: SCSI Devices [5:0:1:0]disk SEAGATE ST6000NM0095 NE00 /dev/sdc - I wonder if the format screwed up and now I just need to reformat. I got the format regoing, this time in console instead of unraid terminal. Edited January 2 by opiekeith 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.