Is this hard drive dying? SmartCtl Extended test also not progressing

[Dice]

AFAIK Seagate states the numbers output on drives cannot be read/compared to others manufacturers.
To what extent this applies the present example with infinite problems indicated - may just be due to Seagate's shenanigans.
More info:
http://knowledge.seagate.com/articles/en_US/FAQ/203971en

 

[Robert Trevellyan]

it has now realised those sectors aren't infact bad

All it knows is that it was able to write and read those sectors without error this time around. There's no telling what might happen in the future. I wouldn't trust that drive.

 

[ewhac]

All it knows is that it was able to write and read those sectors without error this time around. There's no telling what might happen in the future. I wouldn't trust that drive.

Seconded. If you care about your data, you don't need a drive that one moment says the sky is falling, and the next moment says Everything Is Fine.

 

[FreeNASftw]

FWIW (not a huge amount) I have a drive that is rock solid for many months at a time, but sometimes after a reboot it will fail - CAM status errors everywhere, chksum errors etc etc. I run test after test on it and it passes. Clear the errors, run more tests, reboot, all is fine for months... It could be a slightly dodgy connection on the SAS card, dodgy cable etc. I haven't bothered to do anything about it since spending several days on trying to figure out what was going on about a year ago.

 

[Crm]

FWIW (not a huge amount) I have a drive that is rock solid for many months at a time, but sometimes after a reboot it will fail - CAM status errors everywhere, chksum errors etc etc. I run test after test on it and it passes. Clear the errors, run more tests, reboot, all is fine for months... It could be a slightly dodgy connection on the SAS card, dodgy cable etc. I haven't bothered to do anything about it since spending several days on trying to figure out what was going on about a year ago.

This is true, I will replace the cable and try reseating others and see if that makes any difference although normally I would expect checksum errors rather than bad sectors in this scenario

 

[Robert Trevellyan]

I'm not aware of any external factors that can result in a drive reporting bad sectors.

 

[joeschmuck]

Well I couldn't help but chime in as well.

The drive has failed and did not automatically lock out the failing sectors. Running badblocks forces this action. So for now your hard drive "could" be good for a while however you need to understand the mechanics of the typical failure and then you would realize that the drive is not trustworthy. Basically the head floats on a cushion of air above the platter surface. This is one thin film of air (I'm using air as a general term, some drives use other gasses to allow the erase/write/read heads to fly closer to the media) and the heads are damn near touching the surface. Any defects in the surface or particles floating around could cause what's known as a head crash. Bits of the platter surface are damaged and flake off, now you don't have a perfectly smooth surface. So now you have some bad sectors, you run bad blocks to map the sectors out of existence but the heads still fly over that area all the time so more surface area flakes off thus creating more damage. This is why most drives cannot be saved once you start having sector errors. Let me be clear, this is an internal hard drive error, not external, meaning it's not a SATA cable nor SATA Controller.

I agree with my colleagues that the drive is bad. I would not use it at all, but if you must then don't use it for anything important.