July 4, 20242 yr Last night i received two emails saying that my array has errors and that Disk 9 is in an error state. The main tab shows 807 errors. I ran a quick smart test on drive 9 and it came back healthy. The extended test is currently running (16tb drive) for the last 10 hours now, probably has a few hours to go. Once its done i'll attach the test results. Will the extended test give me a result like the quick test did, or will i need to comb through all the smart attributes? In the meantime, i've attached diagnostics. From reading around if the smart test comes back fine, i'll reboot once the server post another diagnostic and then start the rebuild process by: Stop array Unassign disk to be rebuilt Start array with disk unassigned Stop array Reassign disk to be rebuilt Start array to begin rebuild During this rebuild time, will my server be functional (plex)? I've got dual parity if that matters. If the smart test comes back bad (or something bad in the diagnostics), can i rebuild from a hot spare (14tb) thats part of the array but hasn't had anything written to it yet except 97.6gb which i'm assuming is the file system? Much appreciated. tower-diagnostics-20240704-1001.zip
July 4, 20242 yr Community Expert There were errors being logged for days for multiple disks before that one got disabled, so most likely a power/connection issue.
July 4, 20242 yr Author 26 minutes ago, JorgeB said: There were errors being logged for days for multiple disks before that one got disabled, so most likely a power/connection issue. Thanks! All my drives are on hba 9207-8i and connected through an intel RES2SV240 sas expander. I'll try reseating that plug on the sas -> sata breakout cable when i shut it down to reboot and re-run diagnostic.....once the extended test is complete.
July 5, 20242 yr Author 6 hours ago, JorgeB said: There were errors being logged for days for multiple disks before that one got disabled, so most likely a power/connection issue. Extended Smart report is attached. It completed without error. I shutdown, reseated cables, and ran another dignostic (attached). I'm going to start the rebuild process as i posted above with the same disk. tower-smart-20240704-1715.zip tower-diagnostics-20240704-1726.zip
July 18, 20241 yr Author On 7/5/2024 at 1:01 AM, JorgeB said: SMART looks OK and no errors in the syslog so far. We'll i just checked on the server and saw that disk 9 is in an error state AGAIN. I ran a short self test and it Completed without error. Attached is an updated diagnostic. I need to go back and look, but i think that this drive might still be under warranty....not sure if this is a connection issue, or the drive is failing. here's the smart info if that helps tower-diagnostics-20240718-0650.zip
July 18, 20241 yr Community Expert On 7/4/2024 at 6:27 PM, JorgeB said: There were errors being logged for days for multiple disks before that one got disabled, so most likely a power/connection issue. The above still applies to the latest diags, and don't see anything that suggests a disk problem
July 18, 20241 yr Author 12 minutes ago, JorgeB said: The above still applies to the latest diags, and don't see anything that suggests a disk problem Thanks for looking into it.....multiple disk errors you say? Something to be concerned about? I'll try using a different power and sata cable and see if it changes anything......now for another rebuild. Edited July 18, 20241 yr by danimal86
July 18, 20241 yr Community Expert 23 minutes ago, danimal86 said: I ran a short self test and it Completed without error. You really need to run the Extended test to get a good indication of the health of a disk.
July 18, 20241 yr Author 43 minutes ago, itimpi said: You really need to run the Extended test to get a good indication of the health of a disk. 43 minutes ago, itimpi said: You really need to run the Extended test to get a good indication of the health of a disk. Yeah thats a good call. I'll run the full test. It just takes forever for a 16tb drive and if its like last time (2 weeks ago) it will come back with no errors.
July 19, 20241 yr Author Extended smart test completed without any errors. Attached is the smart report. I'm going to go ahead and swap cables and start the rebuild process. tower-smart-20240719-0653.zip
July 30, 20241 yr Author Well, it happened for a 3rd time. At this point im at a loss. This will be the 3rd rebuild i've done. Last time i made sure to use different sata and power cable. I wonder if its something wrong with the hba or the sas expander card. I believe that the drive is still under warranty, but the health checks still come back without any errors, so i'm not sure if its an actual drive issue or what. Not sure if its worth spending another day running a extended health test, or if i should just start the rebuild process. tower-diagnostics-20240730-0655.zip
July 30, 20241 yr Community Expert If it's always the same drive failing with new cables with may be worth trying with a new one.
July 30, 20241 yr Author 31 minutes ago, JorgeB said: If it's always the same drive failing with new cables with may be worth trying with a new one. yeah, its the same drive that keeps failing. I didn't try new cables, just used power/sata cables that weren't previously used. I'll try reaching out to seagate and seeing what options i have. As far as the array goes, the failing drive (16tb) is the one that was currently being filled up, i hav two 14tb drives that have only had 97.6gb (file system?) written to them. Can i just do a rebuild onto one of the 14tb drives while i figure out whats going on with that 16tb drive?
July 31, 20241 yr Author 11 hours ago, JorgeB said: Nope, replacement disk cannot be smaller. any reason why I cant/shouldnt do the rebuild then shrink the array and remove that disk and rma it? I've got plenty of room
July 31, 20241 yr Community Expert That you can do, you can also move any data from that disk elsewhere and do a new config without that disk, parity will need to be re-synced.
July 31, 20241 yr Author 6 hours ago, JorgeB said: That you can do, you can also move any data from that disk elsewhere and do a new config without that disk, parity will need to be re-synced. I still need to do the rebuild first correct? Is unbalance plugin still the best way to move data off the drive? The extended test came back without error. Edited July 31, 20241 yr by danimal86
July 31, 20241 yr Community Expert You don't need to rebuild if you plan to remove that disk, just empty the emulated disk,
August 1, 20241 yr Author 11 hours ago, JorgeB said: You don't need to rebuild if you plan to remove that disk, just empty the emulated disk, Dang, well too late.....started the rebuild. Unbalance still the best tool for the job?
August 24, 20241 yr Author On 8/1/2024 at 12:24 AM, JorgeB said: It's an option. I've finally gotten a replacement disk precleared (and a backup) and i'm getting ready to RMA the disk that had the issues. It hasn't had the same issue as it did the last 3 times, but i dont trust it. I was wondering if you would be able to help me by letting me know what information i could provide to seagate regarding the error. I did a quick scan of the drive and it came back fine. I'm just trying to cover my bases before i submit the rma.
August 25, 20241 yr Author 9 hours ago, JorgeB said: Post a SMART report for that disk. seagate SMART.zip
August 25, 20241 yr Community Expert I don't see anything that suggests a disk problem with that SMART report, and Seagate may not replace it.
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.