"delayed write failure" cleared up after reboot - should I be concerned

[x509]

All of a sudden (of course), I started getting "delayed write failed" messages on the drive that I use for all my disk to disk backups. Some of the messages included "$MFT".

I couldn't chkdsk the drive. I could open the drive in a window and see the top-level directories, but each directory had 0 files or folders content. Which I knew was bad.

Finally I rebooted the system and ran chkdsk. That ran fine. Now the drive seems perfectly normal. So what happened? Should I be replacing this drive, or moving the contents to another drive and then doing a low-level format?

Thanks.

 

[YeuEmMaiMai]

Yes it is time for a new drive. you are not using the old defective sandy bridge chipset are you?

 

[BBA]

If you are using a portable drive, chances are you removed the drive or powered it down before cached data was written to it. If thats the case, I wouldn't worry as long as it doesn't happen again. You can also disable disk caching for the drive in device manager.

 

[x509]

Yes it is time for a new drive. you are not using the old defective sandy bridge chipset are you?

No, not the sandy bridge chipset. Something much older (never mind!) and due to be retired from service as soon as I can build up the replacement.

I'd like to add that I have had zero problems with this drive since that first reboot. That drive includes my Windows swap file, so any issues would become apparent right away.

 

[x509]

If you are using a portable drive, chances are you removed the drive or powered it down before cached data was written to it. If thats the case, I wouldn't worry as long as it doesn't happen again. You can also disable disk caching for the drive in device manager.

Thanks, but this drive is not a portable. As I noted in my previous reply, I have had zero problems since the original problem presented itself AND I did a reboot and a chkdsk.

 

[Mackintire]

First I would try to make an image of that drive, as an emergency backup. Sometimes the process of testing stresses a drive.

Then I would run crystaldisk SMART tools on it (this has almost zero chance of stressing the drive)

and (its vender specific) testing validation tools (some of these tools will stress the drive)

If all those show the drive is ok.... Then do some good ol fashion I/O Meter testing just to make sure. http://www.techrepublic.com/article/test-storage-system-performance-with-iometer/5735721 (This is about as brutal of a stress test as anything, if your drive performs fine here and all the previous tests the drive is probably fine.

 

[cyclone3d]

It could very well be a flakey data cable.

In the past I've had strange errors that looked like my drive was going bad.... lockups, failure to detect, slowness, etc.

After pulling my hair out trying to figure out what the problem was I finally decided to try replacing the SATA cable and the problems dissapeared.

I've had the same type of thing happen much more often in the past with IDE cables.

 

[x509]

It could very well be a flakey data cable.

In the past I've had strange errors that looked like my drive was going bad.... lockups, failure to detect, slowness, etc.

After pulling my hair out trying to figure out what the problem was I finally decided to try replacing the SATA cable and the problems dissapeared.

I've had the same type of thing happen much more often in the past with IDE cables.

cyclone3d, You GOT IT.

After I posted that all seemed to be OK again, I endured several days of flakey system performance. At times, the system would boot up but my drive H: (the problem drive) would simply not be online. Other times, my system would start the Windows bootup process but then I would get a black screen, forever it seemed.

So finally, I figured I would check the power and data cables. And, whaddya know, the SATA cable had a sharp bend right where the cable plugs into the drive.

How come? This system is housed in an old-but-good Antec P180. It cools well (enough) and looks great, but cable management simply ins't. And it uses a drive cage to mount four drives at the bottom front of the case. The cage slides into the body of the case, but there isn't much space between the back of the drives and the wall dividing the bottom part of the case. And I guess one of the SATA cables got bent over the last time I did some maintenance on my system. Why now and not a month ago when I did the maintenance? Who knows?

Anyway, I tossed the bad cable, put in a new one, and the system has been fine ever since. I've even down a few power cycles just to double-check.

Also, I have done a full backup of the drive contents. Nothing like having some spare drives just sitting around, just in case.

I'm hoping that this is my last post on this subject. Hoping.

 

[x509]

Anyway, I tossed the bad cable, put in a new one, and the system has been fine ever since. I've even down a few power cycles just to double-check.


I'm hoping that this is my last post on this subject. Hoping.

I wanted to update all you guys on this. It's been four days now since I put in the new data cable, and there have been zero, zilch, nada, problems with that drive. Cables are sure cheaper than new drives.

Lesson learned. Check your cables BEFORE breaking the glass and turning on the fire extinguisher.