HDD failing, is it safe to backup?

coolrunner84

Limp Gawd
Joined
Sep 5, 2008
Messages
147
I am currently running WHS and have the following setup:

Intel ss4200 NAS
Rosewill external SATA enclosure (5 SATA)
5x2TB HDD

One of my drives has been giving me problems, and now it seems to be dying. I noticed it again tonight as I was trying to watch a movie from my server, when it began stuttering. I am running HD Tune quick test on it right now, and it is passing, but the test is moving very slowly.

My question is that I occasionally run backups of my server to another server, which I leave off when not in use. I slacked off, and have not made a backup for a couple of months. So is it safe to backup? I use robocopy to make a mirror of my primary server, and I don't want to damage any good files on my backup server.

P.S. It seems that the cause of the drive failure should be heat. It was situated right over the northbridge. If anyone has this NAS, they should know that it is slim and tall. The drives are placed right over the motherboard and PSU. I am really beginning to think that it would be better if I move to an ATX case instead to give the components more breathing room.
 
Last edited:
I moved the drive to my external enclosure, to rule out a faulty port or cable. It was still stuttering upon playback, and HD Tune quick scan was taking a long time.

I then tried to transfer a 2.7GB video file to my backup server as a test. It took 1:10:33 to complete, with a speed of ~0.7MB/s.The video played backed without any problem, and it appears to be fine.

Does anyone know what the problem can be? It seems to be that the data is intact, but the read speed is very slow.

I believe that I need to transfer ~300GB or more of data to my backup server. At the speed it was transferring at, it would take around 5 days to complete. Any suggestions are welcomed.


Edit: Is there a WHS utility I can use to determine which drive files are currently located on?
 
Last edited:
I have a 500GB Seagate that just started to do the same exact thing.

Seatools wouldn't let me run any tests except for the generic ones.

I have some WD drives as well, so I ran the WD tools and it said that everything was O.K.... passed S.M.A.R.T as well as the other tests.

I then did a google search and found "Active @ Hard Disk Monitor" It shows you all the logged S.M.A.R.T information. Low and behold, my Seagate drive was having a bunch of RAW Read errors as well as seek errors.

Luckily I was able to copy everything off the drive.
 
I have a 500GB Seagate that just started to do the same exact thing.

Seatools wouldn't let me run any tests except for the generic ones.

I have some WD drives as well, so I ran the WD tools and it said that everything was O.K.... passed S.M.A.R.T as well as the other tests.

I then did a google search and found "Active @ Hard Disk Monitor" It shows you all the logged S.M.A.R.T information. Low and behold, my Seagate drive was having a bunch of RAW Read errors as well as seek errors.

Luckily I was able to copy everything off the drive.

I downloaded that program and I get the following SMART information for the bad drive. This is with the drive plugged into another computer.

I am not sure how to read the information.

Active%20HDD.jpg
 
Well, you have RAW read errors AND you have "current pending sector count" - unstable sectors waiting to be remapped. The drive is definitely going bad.

The Description at the bottom is pretty helpful in telling you what is going on.

As for the spinup time, I wouldn't worry about that too much.

All 3 of the WD drives I currently have in my computer are 100% on everything except for the spinup time.
 
I've only ever used spinrite to be able to recover data from drives that would not read otherwise.

Spinrite doesn't actually repair the drive.. it just move/remaps the unreadable data to good sectors and marks the faulty sectors as bad.

It won't do anything for mechanical failure.

As for drives that are starting to fail.. it isn't worth it to do anything but recover the data if possible and then replace the drive.

Sure, the drive may work for a little while longer, but if it is developing bad sectors, etc. it will only get worse with time.
 
I had a drive with similar read errors once. I pulled all the data off, did 3 or 4 zero-write passes, and the problems automagically disappeared (or at least stopped getting reported). Still works to this day, 4+ years later.
 
Back
Top