Hi everyone,

I have a WD7500BPVT that I am/was using to store some Hyper-V Virtual Machines. At one point a machine stopped responding, and when I looked into it, the Host OS (Windows) was throwing many "IO operation was retried" errors for that drive. I opened the smart data, which shows 167 in Value for reallocated sectors. The value was static though. A quick SMART test didn't return anything. I found that if I stopped any activity on the drive, then ran a synthetic benchmark (constant writing) it would do absolutely fine and peak at over 100MB/s. I've since backed up the data on the drive, and I'm now running HDAT2 on it.

I'm currently running a VERIFY/WRITE/VERIFY test on it which is also moving along at a great clip. So far in the errors count, it shows 1/0/0 for blocks and 68/67/67 for sectors. I can see messages appearing that say "Selected device is hung - reset required". Can anyone tell me what all of this says about a drive? Is it actually near death, or is something else going on?

Thanks in advance!

I can't see anywhere to edit my post, but I wanted to add a couple of things:

I tried running chkdsk a few different ways. One was with virtually every option I could use (besides /F /R) to scan the disk, which would come back clean. Using /F or /R would make Stage 4 take an absurd amount of time but it didn't catch anything for the time that I did let it run.

I've read that device hung/operation retried errors could indicate a bad SATA cable or controller, so I've just replaced the cable now and I'm running the HDAT2 test again. So far so good. I just find it odd how a cable could suddenly go bad after consistently good use for months.

Would recommend checking SMART with other tools and confirm consistency.

_________________
Hard Disk Drive (HDD), Solid State Drive (SSD, SATA, NVMe, etc), USB Flash Drive and RAID Data Recovery Specialist in Massachusetts

Hi labtech,

Here's what happened since. After swapping the cable, I let the V/W/V test run until completion. And it did so with absolutely zero errors or warnings occurring. Perhaps the cable swap worked? I tried running a simple write-only blank operation afterwards, but that was running about 3-4 times as slowly for some reason so I cancelled it about 15% of the way through. Afterwards I took a snapshot of the SMART data, and Reallocated sectors count dropped to 145, while another re-allocated sector event value is showing 122 (IDs 5 and 196). Otherwise everything else appears to be normal.

I also rebooted to try some other diagnostics utilities, but I either couldn't get them to run or they didn't seem to have issues (HD Tune was still able to get consistent benchmark data). I have now rebooted back into the main OS, Windows Server 2012 R2 and I've initialized the disk to MBR and given it a single NTFS partition (quick format). I'm now trying to copy the data back, and I'm not certain if everything is totally resolved. The activity is slightly different than before - Response time for writing a single (large) file is hanging around 3,000ms and according to Resource Monitor, it's maintaining an average of about 10,000,000B/sec. However, and this is the most shocking, the Windows copy dialog is jumping up and down from a max of about 11MB/sec to something much lower. I'll attach a picture below, where you can see it saturating my Gb connection probably filling a cache up, but then when it comes to putting the data on the disk, it starts this up and down of writing the data to the disk. I'm going to let the file copy finish and boot up a VM to see how it behaves.

Am I just misinterpreting these graphs/data and everything is actually Kosher, or do you see anything out of the ordinary? I am dumbfounded how a SATA cable could be just bad enough that there would've been such inconsistent data errors as before.

Thank you for responding!

You appear to be looking at the normalised value of the Reallocated Sector Count, in which case the drive is failing rapidly (167 -> 145). I think you will find that 140 is the threshold for SMART failure. The raw value of this attribute is the actual number of reallocated sectors, probably now around 500.

In short, I wouldn't trust this drive with your data.

_________________
A backup a day keeps DR away.

Raw sector count is 466 and sector event is at 78 so it sounds like you're spot on. I guess this does make sense... those points of random total lock ups were probably failing sectors being found. I don't think I've put this much data on the drive in a long time so that may be why it's finding them now. Thank you for explaining this to me, it sounds like the insanity with the cable was uncalled for.