Click here to Skip to main content
15,890,438 members
Please Sign up or sign in to vote.
5.00/5 (1 vote)
See more: (untagged)
This is no joke!
For a second time in past two years I have this weird problem.
My internal hard drive (dynamic disk with mirrors) gradually disappeared from the BIOS.
By gradually I mean sometime it was there and another time it was not.
The Logical Disk Management (win2000) would report dynamic disk missing and reactivation would say the disk cannot be located.
During the windows startup I would get two consecutive short beeps at random until the windows starts – maybe two or five of them.
I had this happened before ( different box, same OS) and the temporary solution was to take the drive out of the machine (!) and place it vertically next to it! I am not kidding!
So, last night I took the drive out of the box (vertical mount) and turn the power on.
Did not touch any cabling!
Guess what   - it loaded and RAID started regenerating my mirrors!
I put it back into the box and it worked just peachy the whole day.
Today – zilch, same old problem - disk is just not there!

Of course the disk is out of warranty and the vendor could careless about my problem.

In reality I do not want to spend much time troubleshooting this, but this is a second time around and I feel the hard drive should last past the warranty period.

Before I toss this worthless drive – can anybody tell me what could be wrong and where are these beeps coming from.
My best guess is that the drive has some temperature related issues.
Or maybe “consumer grade “ run of the mill drives are not suitable for RAID – the forced or hidden regeneration may be too stressful.

Thanks for reading. Any constructive comments are appreciated, just please do not waste my time telling me that drives are cheap and to buy a new one.
Vaclav
Posted

1 solution

The beeps are most likely your RAID controller or BIOS reporting that it can't see the drive you've configured.

I'd check and replace the cables. They do fail if damaged, and often intermittent connections occur where it works in a certain range of temperature or certain positions, but not others.

Likewise, electronics can fail when subjected to shock or high temperatures. There can also be poor joints in the soldered board which eventually fail with vibration, shock or temperature.

If it's mirrored, and replacing the cables doesn't help, toss the drive. It just isn't worth investigating the problem. If it's bad electronics or weak joints it can't be fixed without replacing the drive's controller board anyway.

Consumer drives are pretty much identical to 'enterprise' ones, except that the enterprise drives generally have write-back buffers disabled (consumer drives will report that they have written data when they have written it to the on-drive cache, not actually to the disk itself, so you can lose data on losing power that the OS thought was written) and also have aggressive error-recovery disabled. Consumer drives will re-read sectors that have errors repeatedly, leading to long delays if there is a problem, but it might be able to recover the data. Enterprise drives are intended for use in mirrored or parity-protected arrays, the system can recover the data from other drives, so the recovery is much reduced so the OS or RAID controller can get on with doing that, then replace the bad sector.

Drives often do last quite a long time. We recently had to replace one that had been in service for six years in a RAID array in a server - this was a setup with Windows 2000 Server mirroring the data on consumer IDE drives. Unfortunately it's common for drives in the same batch to fail at more-or-less the same time, if they haven't been subjected to different stresses. Some administrators get drives from different batches or even different manufacturers for this reason.

Sometimes, though, drives fail much sooner. In a computer I bought for my parents in 2001, the disk failed in less than three months. (It was replaced under warranty by the manufacturer.)

 
Share this answer
 
Comments
Vaclav_ 25-Mar-11 10:21am    
I am a little late in posting my comments, but basically I had a failing hard drive which eventually crashed for good.
Lesson learned - if in doubt - toss it out!


CodeProject, 20 Bay Street, 11th Floor Toronto, Ontario, Canada M5J 2N8 +1 (416) 849-8900