I have Seagate ST3250620AS and it is disappearing in OS. It can happen just after one hour and sometimes once in a month. Seagate programs say that HDD is OK, but SMART read out gives same errors. When disk is online it works without problems. What could be a problem?

Here are a few suggestions, all IMHO:

a) Check the Windows system event log, to see if you get any clues about the type of problem there.

b) Make sure you have a valid, readable backup - then you don't need to care so much, if the disk does have a fatal problem.

c) Do you know whether those SMART attributes highlighted in yellow in your report, have been non-zero for a while?; or have they only changed after these problems started?; or do you not have any previous SMART data for comparison? It is usually possible to gather some more info about the timing of SMART recorded errors from internal drive logs, using smartctl (from smartmontools), if you don't know when those errors occurred, but the errors are only times in drive POH of course, not in "wallclock" date/time.

d) Watch for any changes in the SMART data immediately after any new problem.

e) I assume that by "disappearing", you mean the disk itself is no longer seen (e.g. not in Device Manager in Windows), as wwell as no longer seeing the relevant drive letter(s) in (My) Computer. That behaviour could be related to problems which are not caused by the disk (e.g. power, SATA cabling, motherboard SATA port etc.), but trying to diagnose that type of problem remotely via a web forum is inefficient. This is especially true if the frequency of the problem is very variable, as you are saying is true in this case.

f) IMHO there are several possible troubleshooting approaches, depending on how much time, effort & money you want to spend on this, as well as what other hardware & PCs you have access to.

g) You could run SeaTools to exercise/test the drive (after you have a good backup), but if there is any possibility of a problem outside the disk, as I mentioned above, then a SeaTools reported "failure" may not actually be a disk fault.

+1 to all above.
Full backup and zerofill drive. See if SMART warnings disappear. If doesn't, then need some working with firmware.
Interesting things, I never seen before: Parameter 190 value the "Worst" is below threshold. It looks like SMART has already tripped before, or that SMART is corrupted.

It does the same under Ubuntu, so i never checked windows log.


Already made one. But I would like to use this disk in old machine that i will move to my countryside house.


Will do that as i don't remember when errors have occurred.



Yellow ones hadn't changed but seek error rate, error rate indeed changed.



Yup, just disappears. No trace of it in device manager nor disk management. If program was started from that drive computer hangs. I've changed SATA cables, power cables, cleaned connectors on HDD, tried on other sata ports, switched ports with other 250gb disk which is boot disk on this machine and boot disk works fine, power supply is 650W seasonic two years old. Everything is pointing that drive is a culprit.



Have several PCs in my company, and have time as i want to find out what is happening with it. Never saw this kind of problem before and i'm very experienced user (build my own rigs, know how to repair motherboard, i did repaired HDD electronics with advice from users on this forum...) I mean i have money to replace this HDD but that's not a point.



SeaTools never reported any problem with this drive.



I'll zero fill it, but i think i already did that when problem first occurred. Will try it again certainly and report.