neal_is_king Posted January 29 Share Posted January 29 (edited) Edit: After posting it occurs to me my post is quite long so here is a summary: Summary Array is 4 12TB Toshiba N300 drives, just under a year old Can't run short or extended SMART test on any of them - Completed: unknown failure 90% 3 of them logged a Error: ABRT trying to run the test and now my BIOS is telling me they're dead All plagued by a Seek_Error_Rate issue that is apparently quite common for these drives I think they're OK but maybe not Hi all, This likely is not an Unraid issue exactly, but the drives in question are the totality of my parity and data drives. I'm a little bit worried but also wonder if it's a big nothingburger. My current build has 4 Toshiba N300 drives, 2 data and 2 parity, and 2 970 EVO Plus cache drives in a RAID1 btrfs pool. I bought the mechanical drives in the same month about a year ago from 3 different stores, not sure how I did w/r/t different batches. The way my present issue came about is, last week I started playing with new DVR software and was also downloading a lot of Linux ISOs lets say and my cache pool was under a bit of stress. I think. I'm not sure exactly what happened, but I woke up the next day and my box was unreachable. Apparently it rebooted and was stuck in the BIOS which wouldn't boot until I acknowledged that one of the NVME drives was dying or dead. Turns out to be parity2, that's kind of a drag and I should probably RMA it because it's not that old but OK, order a new one, rebuild the pool, done. Somewhere throughout this process it occurs to me that I haven't run any SMART tests since I finished preclearing these 12TB drives 6808 LifeTime hours ago. I kick off a short test for each, but none of them finish! I just get Completed: unknown failure 90% in the self-test history, and now two of them have an Error: ABRT and one has Error: ICRC, ABRT in the error log as well. I powered off when I received a new heat sync for the replacement cache drive from above, and now the machine won't boot because the BIOS wants to tell me about these 3 SMART errors. After seeing very scary sounding warnings about replacing the drives immediately to prevent data loss, with some extra steps I am still able to boot. There is a BIOS option to not check SMART during post, which I have not changed yet. So back in Unraid, I try and try to get one of these drives to run a short SMART test, but it's the same thing every time, Completed: unknown failure 90%. I logged maybe 6 or 7 failures like that for each drive. I'm currently running an early parity check just to see if I get any errors or if the stress does anything to the SMART reporting. It's 70% done and so far so good. Also if I run the short SMART *during* the parity sync, it gets further along - Completed: unknown failure 10% - not sure what to make of that either. I tried an extended test one one of them too, also fails with 90% remaining. All of these drives are (quite annoyingly) plagued by the Seek_Error_Rate issue that seems to be just par for the course with them. My strategy there is just to ignore that metric completely. And all 4 of them report SMART overall-health self-assessment test result: PASSED. I don't know what SMART does outside of user initiated self tests, I assume it's reporting metrics without running a self test because I almost never do that. As for troubleshooting, I have tried re-seating the SATA cables, replacing the SATA cables, and moving all the drives from the motherboard to an HBA I have in a box from my previous build. Nothing has made a difference in getting these self tests to finish successfully. The only thing I can think of to try next is run SMART with the drives in an external enclosure plugged into another machine, but I don't think that would make a difference as the test is running "on" the drive itself? I also have no idea if the Error: ABRT would be fixed if I did get one to finish, which would let me keep that check on in the BIOS. I have a sense that these drives are fine to be honest. Reallocated_Sector_Ct, Current_Pending_Sector, Offline_Uncorrectable all fine. I'm also not really experienced enough in this to be confident in that assessment. I am curious what you guys would do in my situation, or if anyone has any advice or thoughts to offer. I am out of the return window at all 3 stores, and everything I've read about this Seek_Error_Rate issue says Toshiba just sends refurb drives that quickly have the same issue too. I haven't found anything about SMART tests failing to complete on these drives however. I attached diagnostics if anyone would like to see the SMART history. I have redundant, offsite backups of important data but there's plenty of non-critcal data I would strongly prefer to keep. Thanks! Edited January 30 by neal_is_king Quote Link to comment
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.