grither Posted February 8, 2016 Share Posted February 8, 2016 just my first unraid 6 email warning, saying the above in the subject line. I can post a smart report later tonight, but assuming the pending sector count is only '1', would the community recommend any steps be taken? or just continue to monitor? Quote Link to comment
tr0910 Posted February 8, 2016 Share Posted February 8, 2016 As long as you have a spare that is already pre cleared you don't need to panic. If not start looking. Best to be ready for a worst case failure. This is the first sign of a drive likely to have future problems. I have had only two serious drive fail in over 100. I've had numerous single sector issues like you mention. They aren't guaranteed to result in immediate drive failure, but they can. Often what I do is pull the offending drive at my leisure, and rebuild the data onto my spare. Then the offending drive gets pre-cleared once the array is fully rebuilt. Often the drive self heals itself during the pre-clear and I use it again later. I just had 2 WD EZRS 3tb drives do that last week. What is the make and model of the failing drive. Quote Link to comment
JorgeB Posted February 8, 2016 Share Posted February 8, 2016 Do an extended SMART test, if it fails replace it. Quote Link to comment
grither Posted February 9, 2016 Author Share Posted February 9, 2016 hi can someone tell me where the smart reports are kept? i ran the extended, but don't know where to see the outcome. there is a button called download, which i clicked but don't know where this saves to! Quote Link to comment
trurl Posted February 9, 2016 Share Posted February 9, 2016 hi can someone tell me where the smart reports are kept? i ran the extended, but don't know where to see the outcome. there is a button called download, which i clicked but don't know where this saves to! It's a download. It saves it on the computer your browser is running on, wherever your browser saves any other download. Quote Link to comment
grither Posted February 9, 2016 Author Share Posted February 9, 2016 ok oops, download did nothing, i think because the smart report failed. I am in the process of moving the data off this drive and onto other drives. unfortunately got cuaght without a spare, because i recently lauched a second server and wanted to start using my previous spare, for warranty reasons overnight, got two more messages, that pending had grown to 3, then to 5, then offline uncorrectable at 2. smart report attached if interested chimera-smart-20160209-0432.zip Quote Link to comment
JorgeB Posted February 9, 2016 Share Posted February 9, 2016 Single pending sector can be a false positive, but as it's increasing replace disk ASAP. Quote Link to comment
grither Posted February 9, 2016 Author Share Posted February 9, 2016 yes will do since i don't have a precleared disk ready to go, i am currently copying files off the disk, onto other disks within the array. I suppose that's okay? even if the disk totally fails, i could still rebuild once i have a precleared disk ready? Quote Link to comment
gundamguy Posted February 9, 2016 Share Posted February 9, 2016 yes will do since i don't have a precleared disk ready to go, i am currently copying files off the disk, onto other disks within the array. I suppose that's okay? even if the disk totally fails, i could still rebuild once i have a precleared disk ready? I think you've got two things to be concerned about. 1) If another disk fails before you can rebuild that would be majorly bad. (If you don't have a spare this is a risk you've got to take for a short amount of time...) 2) Is the data you are copying off of the "bad" disk free of errors. (I don't know the answer to this, but someone here should be able to give you advice.) Quote Link to comment
JorgeB Posted February 9, 2016 Share Posted February 9, 2016 I would leave it alone or off until you’ve got a spare, if the rebuild fails for some reason you can still recover data from the bad disk. Quote Link to comment
karateo Posted February 9, 2016 Share Posted February 9, 2016 I would leave it alone or off until you’ve got a spare, if the rebuild fails for some reason you can still recover data from the bad disk. +1 make use of unraid to build a good copy of your files by rebuilding into a new disk. keep the bad disk in case another one is failing during the rebuilding Quote Link to comment
Chris Pollard Posted February 9, 2016 Share Posted February 9, 2016 Perhaps he is copying data from the simulated disk.... That's what I have done in the past. Quote Link to comment
gundamguy Posted February 9, 2016 Share Posted February 9, 2016 Perhaps he is copying data from the simulated disk.... That's what I have done in the past. I don't think the disk is being simulated yet, since it hasn't actually failed yet. But this is a good point. Quote Link to comment
grither Posted February 10, 2016 Author Share Posted February 10, 2016 uh oh, i had begun a copy of the whole drive to other drives in the array, and then left for work. when i got home it was 99% complete, then finished. I set it to copy before i saw all the recommendations, oops! fortunately i think everything came off okay. i have now removed the drive from the array, and am rebuilidng parity. once parity complete i will run a preclear on the drive, and just see what comes of it. the drive jumped from pending sector, to 3, then to 5, all within about an hour. it appears to have stabilized... but we'll see the results of the preclear. drive is quite old, and out of warranty by 3 years. Quote Link to comment
itimpi Posted February 10, 2016 Share Posted February 10, 2016 uh oh, i had begun a copy of the whole drive to other drives in the array, and then left for work. when i got home it was 99% complete, then finished. I set it to copy before i saw all the recommendations, oops! fortunately i think everything came off okay. i have now removed the drive from the array, and am rebuilidng parity. once parity complete i will run a preclear on the drive, and just see what comes of it. the drive jumped from pending sector, to 3, then to 5, all within about an hour. it appears to have stabilized... but we'll see the results of the preclear. drive is quite old, and out of warranty by 3 years. The problem is that each pending sector means that there is a corresponding sector on the parity disk that is probably wrong (since pending sectors come from potential read failures at the hardware level) Quote Link to comment
grither Posted February 10, 2016 Author Share Posted February 10, 2016 Yes but now that I'm rebuilding parity without that disk won't it be okay? Quote Link to comment
itimpi Posted February 10, 2016 Share Posted February 10, 2016 Yes but now that I'm rebuilding parity without that disk won't it be okay? Yes. Missed the fact you were doing it without that disk Quote Link to comment
grither Posted February 10, 2016 Author Share Posted February 10, 2016 No worries, thanks for all your help! Quote Link to comment
grither Posted February 17, 2016 Author Share Posted February 17, 2016 so rebuilt parity just fine, excluding disk. Added disk back to array since it appeared like it was 100% fine everything has been fine for the past week, but just now i got an email indicating this disk now has pending sector count of 3. not sure what to do... i could just junk it, but this drama seems to demonstrate that pending sectors on this disk seem to drift up between zero and 5, but not more.... perhaps i leave it in and keep an eye on it? its also a few years old so i'm not too worried about junking it... i guess the best answer is that i'm not really going to trust it, so perhaps it should be out of the array Quote Link to comment
RobJ Posted February 17, 2016 Share Posted February 17, 2016 No, sorry, but bad sectors on one disk can cause rebuilds to fail, so this needs to be dealt with SOONER than later. The parity protection for your other drives is useless if you can't use it because of a drive that won't read correctly all the way through. A drive with an increasing pending sector count CANNOT be trusted. I'd replace it ASAP. Quote Link to comment
gnollo Posted June 14, 2017 Share Posted June 14, 2017 Are you guys saying that a single pending sector on a drive will invalidate parity? Shouldn't it be able to rebuild that drive fully (minus the pending sector)? I have a single drive with a single sector pending, it's a 2TB WDCG drive, has been like this forever... No other issues, and unraid reports Smart overall health: passed. Quote Link to comment
trurl Posted June 14, 2017 Share Posted June 14, 2017 41 minutes ago, gnollo said: Are you guys saying that a single pending sector on a drive will invalidate parity? Shouldn't it be able to rebuild that drive fully (minus the pending sector)? I have a single drive with a single sector pending, it's a 2TB WDCG drive, has been like this forever... No other issues, and unraid reports Smart overall health: passed. If all of your other drives are OK with no pending sectors, then you should rebuild the drive. This should cause the sector to get reallocated and rewritten. Then it will be a reallocated sector instead of a pending sector. The problem with a pending sector is you can't be sure it will be read, so it could compromise rebuilding a different drive since all drives are required to rebuild a missing drive. After you get the sector reallocated it should be fine. Drives have extra sectors just for this purpose. If you want further advice about your system, post your diagnostics. Quote Link to comment
JorgeB Posted June 14, 2017 Share Posted June 14, 2017 It's quite common for WD disks to get false positives, if the disk completes a parity check without read errors or it passes the extended SMART test it's OK. Quote Link to comment
gnollo Posted June 14, 2017 Share Posted June 14, 2017 (edited) Rebuilding parity check as we speak. Once completed I will do an extented SMART test. Attached the smart test I downloaded from the unraid drive screen. tower-smart-20170614-1322.zip Edited June 14, 2017 by gnollo attached file Quote Link to comment
gnollo Posted June 15, 2017 Share Posted June 15, 2017 (edited) Parity check completed on Thu 15 Jun 2017 07:58:29 AM BST (today), finding 0 errors. Duration: 19 hours, 24 minutes, 6 seconds. Average speed: 57.3 MB/sec. 1 Parity 4TB, 7 additional drives for storage available 20TB. Extended SMART test in process. Edited June 15, 2017 by gnollo 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.