2nd harddrive suddenly slow (Seagate Barracuda 1TB)

Spaceblaster

Reputable
Oct 4, 2014
3
0
4,510
About 1½ months ago I built myself a high-end gaming desktop. Had no problems until 2 days ago, when my secondary harddrive, a Seagate Barracuda 1TB (model# ST100DM003-1ER162) became really slow.
ANY action performed on the disk will show 100% active time in Task Manager, and the read/write rates will be almost constantly locked at 0 KB/s. In short, any task performed on the disk takes a looooooooong time to complete, if the task doesn't crash along the way.
My main C: 250GB SSD disk however, runs just as smooth as ever. The problem seems to be isolated to only the secondary drive.

At one point I even noticed the Intel Rapid Storage Technology giving me an error saying it had removed the disk, afterwhich I could not find the disk in windows until I restarted the PC. (error link here: http://www.techpowerup.com/forums/threads/intel-rapid-storage-technology-enterprise-sata-available-disk-removed.175653/)

After lurking around on the internet I have tried the following:

Checked Task Manager & Resource Monitor, no suspect thread hogging resources found.
Scanned with Malwarebytes for any viruses, none found.
Disabled Superfetch, Prefetch & Windows Search, no effect.
Checked the drive for errors with Windows disk error checking, none found.
Reformatted the drive. Writing/Reading to the disk is still as slow as before.
Checked firmware version, I seem to have the latest (CC43)
Downloaded SeaTools and tried several things, first a SMART check, took a bit long but showed no problems. I also ran a short generic test, which took a really long time (1-2 hours) but showed no problems. I also tried the "Fix All" option, but no progress was ever displayed on the progress bar, so I assumed it froze.
I did a chkdsk /r which showed no problems, but when checking bad sectors froze at 8% completion.
Switched the SATA cable between mobo and the hdd, no effect.
Moved the SATA cable to a different port on the mobo. No effect.
Checked the power cable to the disk, its also connected to the SSD, which is working fine, so I assume the cable isn't broken.
Went to BIOS and switched sata connection from AHCI to RAID. No effect.

On the offchance it might afterall be a virus or software error, I tried doing a system restore back to last month, but after the pc restarted I get the error that the system restore failed, becuase the C: drive might be corrupt. It recommended me to do a chkdsk /r on C:, which I did. Still, trying to do a system restore always gave me the error to do a chkdsk /r.

All in all, I'm pretty sure there is somethinething wrong on the disk's end, but the system restore error's I'm getting I'm starting to suspect there might be a software error somewhere else on the PC.

Any suggestions on ways to solve this?
 

Spaceblaster

Reputable
Oct 4, 2014
3
0
4,510
¨


I didn't do a long test, only fast ones.
Did a Windows check on the C; disk, aswell as several tests with SeaTools. No problems at all, tests completed fast aswell.

I think I'm on to something though. Yesterday when running a chkdsk on the faulty disk F:, I somehow managed to get past the point where I kept getting stuck, namely stage 4 (checking for bad clusters). It said it found a huge amount of bad clusters, but also said it failed to complete stage 5 becuase of MHT or something, can't quite remember.

If I recall correctly, I managed to get past stage 4 of chkdsk when Intel Rapid Storage Technology decided to remove my F: disk (as in the link in the first post), which in turn was triggered by running a "Fix All > Fast" operation on F: from SeaTools... I think. I haven't been able to reproduce it yet. I'll update again if I ever do.

EDIT:
Managed to reproduce the "issue" making the chkdsk on F: able to proceed.
I had been running a chkdsk on the faulty F: disk for about 5 hours when I realized it had become stuck on 30% of stage 4. After doing the steps above, stage 4 completed but had 3 disk read erros becuase "the disk does not have enough space to replace bad clusters", even though only 5% or so of the 1TB is filled.

On stage 5: Looking for bad, free clusters, we get the message "232766616 free clusters processed."
And after that we get this:

Free space verification is complete.
Correcting erros in the Master File Table <MFT> mirror.
Insufficient disk space to repair master file table <MFT> mirror.
CHKDSK aborted.

So either almost all my diskspace vanished, or chkdsk lost access to it. I'm starting to realize the disk is busted. Going to contact the site where I bought it from and ask for a new one.
Incase anyone might be interested, all I can think of what might have caused this problem is either:
A: I have a habit of turning off the power to the PCU AFTER I have shut it down for the night. Though I have read from several sources that it won't cause harm to any hardware.
B: I have experienced a few times where my PC just froze whilst performing simple tasks like writing text or running Blender. Not quite sure why it froze, but when it did I had no choice but to hold down the power button (not the aforementioned button to cut power to the PCU) to shut down.
C: The disk was faulty when I recieved it but took a while to collapse into the state it is now.
 

Spaceblaster

Reputable
Oct 4, 2014
3
0
4,510




Thanks for the suggestion.

On CrystalDiskInfo I get a warning about the slow disk having a reallocated sectors count of 68. I take it that's proof of a hardware error I can't fix on my own?

About the system freezes, I think I isolated the cause to the program Blender. It seems to freeze whenever I'm using it for over 30 minutes or so.