Boy, when I asked for feedback on the HDD excessive usage article, you guys were not kidding! Check this out:

Al:
Links to some SMART monitoring programs, and a TechRepublic pdf download on HDDs.

PassMark DiskCheckup
A couple from PANTERAsoft
Detect hard disk failures before they happen (PDF)

Ted:
Short answer would be “no problem.” Does the person who wrote think about the server HDDs that are running 24/7?

Whatever anyone else says, there are only two types of HDD: those that have failed, and those that will, in time. Just have spare disk(s) on hand and a good backup for when it does fail.

As each client gets a new copy every day, if it fails, swap the disk, say, ten minutes once a year. Test 50+ clients every/day/week/month look at the logs, etc. How much longer is that over a year?

Don’t sweat the small stuff.

The servers? That is a different issue!

Phil:
One package that springs to mind is SpinRite 6 from Steve Gibson.

Don:
Hi, Matt , if your corrospondent was to go to Passmark, he might find a program to suit his needs. I use its diskcheckup program for this unit and it gives a lot of info that can be used to monitor the condition of the HDD. It has other programs as well that monitor other components of the computer system that he could look at.

Davis:
For checking drives periodically, HDTune will check the SMART attributes and offers a diagnostic read test as well as a benchmark. It is free for personal use; but you might want to check their site for an educational license.

HDDhealth and HDTemp are also useful and free for home use.

Restoring a drive image every night is going to reduce the life of the drives slightly and, to my mind, may be overkill. I might suggest that a little studying of ERUNT would allow for the automatic restoration of the system’s registry, thereby undoing any changes made by the students combined with a file cleanup utility to delete temporary files would accomplish virtually the same task.

Jeff:
If you want specific suggestions to pass to Dave, here are mine:

1. Take an inventory of ALL hard drives and when they were bought.

2. Look at how long that is, compared with the calendar age of the drive and how long a warranty each one had.
3. If the warranty is three years and the drive is four years old, start planning a replacement (with the right KIND of drive!) as soon as the budget will support it.
4. Continue your imaging every night, but consider buying some of the new network attached storage units from Maxtor or Seagate as targets for the drive images. Modern software will support such things - for instance, Symantec’s LiveState Recovery or Ghost both support this. Capacities are available up to 250 or 300 gigs per attached drive, so it should be plenty of storage space for whatever backups Dave wants to generate.
5. Each system should get a weekly master drive imaging to DVD-R, with the backups kept in a safe place.
6. He should buy at least one copy of Gibson’s SpinRite, in case a drive suffers a catastrophe too far from a backup.
7. Practice intelligent partitioning. Keeping the OS on one drive and creating others for programs and data is the only way to go.
8. Kneel down and pray every night that there isn’t a fire that destroys the whole lot.
9. Hire someone who knows what they’re doing to take care of numbers 1 through 7.

Steve:
Instead of re-imaging every day, he should use some lab management software that will accomplish the same thing. We use software called Deep Freeze, made by Faronics. It’s really pretty cool - this software does not restrict users in any way at the desktop. They can do whatever they normally do as they’re using the PC (install spyware, delete crucial system files, etc.), but on the next reboot, everything is reset to where it was, just as if you just re-imaged. If you’re worried about wear-and-tear on your hardware, this is a great solution.

Jeff A.
I work in Computer Services for a University and we have approximately 200 PCs and 40 Macs in our labs for student use. I have used Deep Freeze for the PCs and MacShield for the Macs for five years now and they don’t necessarily reimage the machine but they do revert it back to a base system state similar to GoBack. Our machines receive quite a bit of use but I haven’t noticed any more drive failures from using the software in the labs than on machines for the Faculty that doesn’t contain the software. I would highly recommend these products and suggest that David just keeps a few extra drives around for when one goes down, which seems to happen sometimes without warning. When this happens I replace the drive with a spare and use Norton Ghost for the PCs to reimage the machine and it will be up and running in about 20 minutes. For the Macs I also keep an image and use PSU Blast Image from Penn State University to reimage the drive. We usually have a three-year turnaround on our machines in the labs and I will need to replace on average three hard drives for about 240 machines in the three years. Personally I don’t find this excessive and it is quick and painless to do.

Randy
I use a small, memory resident, free, utility called HDDlife to monitor disk drive health. While I don’t think any utility has perfect predictive ability, I find this tool useful on my personal systems as well as client machines. HTH.

Don F.
A much better solution to doing the re-imaging would be to use a product such as one of these:

Horizon DataSys
Deep Freeze
Clean Slate

I have two programs that you can choose from. One is from HD Tune. It can do temperature, benchmarking, info, health, and error scanning of the drive. The other program is from HDD Life. This one can do temperature, health, status, and performance.

I have not used these for very long but both look like very good programs.

John:
I’ve been using the freeware version of HDDlife for about six months now and I think it might have saved my bacon on one occasion. HDDlife uses the drive’s SMART system to give you information about your drive’s health, performance and termperature (if the drive supports this feature) About two months ago I was doing a routine back up and Norton Ghost shut down with the message that the destination drive had experienced a read error. I fired the system up normally (I use a Ghost boot disk for system backups) and sure enough, HDDlife was reporting that the health of the drive in question had dropped to 0%. I immediately replaced that drive and did not experience any data loss. I think if I had been paying more attention to HDDlife (it resides in the system tray and is fairly eay to ignore) I might have caught the problem earlier. In any event, you can check out HDDlife here.