My Win Server 2003 boot HD is reporting bad sectors  (revisited)

This is in reference to my question ?My Win Server 2003 boot HD is
reporting bad sectors?.

Here is a link: http://answers.google.com/answers/threadview?id=316171


Hi denco-ga and others?


I purchased 1 gig of a RAM from Crutial.com. Two 500mg chips.  I
replace all the Ram that was in there.

When I go to run the memory test again, I tells be the following:
Here is an approximation of the screen:

========================================================
Pass          Test        Cache
------------------------------
1	      MATS+	  On	          Failed (in red letters)
1	      INVC	  On	          Failed (in red letters)
1	      LRAND	  On	          Failed (in blue letters)? this one may
have succeeded? I can?t remember now)?
========================================================

and again, it froze up after that.


So what does this all mean?  What to do?  the things seems to run, but
trueImage Server wont let me image it, and the RAM test result is kind of scary.  

We relay on this server.  What should I do?


thanx in advnce,

---

Request for Question Clarification by denco-ga on 27 Apr 2004 20:06 PDT

Howdy sherpaj-ga,

I was really hoping new RAM would solve the problem, and it was certainly the
easiest, quickest and very probable thing to test.  As that failed, you might
be looking at a bad motherboard, but at this point it could be a bad hard drive
controller, or a great variety of other problems.  Not good, I am afraid.

If you have another system that you can move the hard drives to, so you can
eliminate the motherboard, etc. as the problem, that could move the diagnostic
process along.

Looking Forward, denco-ga - Google Answers Reseacher

Hi,

I've read though the other question and your current one. I'am going
to approach this in a different manner.

Disk imaging is mainly not for server reliabilty, it's better than
nothing, but requires some time to restore. Imaging is mainly used for
reproduction of workstations.

Not knowing your current setup and use of the server:

For reliabilty you want to use RAID system. In your case I suggest
RAID 1 (Disk Mirroring). This gives you a mirror of the boot drive.
(basically what your doing now). However this a live backup that is
updated near same time the orginal disk is changed. If (in your case
when) the orginal drive dies the other mirror drive will take over
seemlessly.

So you need to find out what brand and model number of the drive that
you curently have installed. Add this new drive and in software (Win
2k3 Srv) or VIA your hard drive controller you can start the mirror.

Why you might ask?

Well from the reports you posted, windows has already corrected some
PHYSICAL problems. So the drive wont get better, it will be down hill
from here. So you will be getting a new drive here soon anyhow. Sure
it may last a year or two or main fail 1 hour prior to your next
image. If you can, you can inable SMART drive and it will tell you
some info about the drive status.

As for the memory errors, it is hard to predict memory failer. But
usually strange things happens when they start to fail. Blue Screen Of
Death (BSOD) come to mind.

If your "Over Clocking" anything on your server, DON'T. If you don't
know what I'am talking about, then disreguard this statement.

Notes:

Also icremental/full backups VIA offline tape backup or other means
would be also be a good idea.

These are low cost and effective means to keep a server going with
minamal downtime.

Well there is google .50 cents worth. AKA my free advice.

X

Well, I have a program called powerquest V2i protector-server edition,
and booted off of the CD that came with it, and noticed it had a ver
of checkdisk on it.  It had lots more options.  I checked it to fix
thing, let it run for 2 hours, and it seemed to do the trick. After it
"fixed" the HD, I could use Trueimage sever CD to boot and image the
drive.

I got a image.  yeah!

Now the bad part.  I rebooted and started using the server normally. 
I went to install a virus program, no prob.  I rebooted, no prob.

Now when I do a virus scan, or when I go to install a program, I get
the blue screen of death.

The screen says: 

*** Hardware malfunction

Call you vendor for support

*** The system has halted ***


Now this is with the new RAM, so my question is:
Is this motherboard, or could it be more bad sectors or whatever on the drive?

also, should I call Dell, since it is a dell server (PowerEdge 600SC
server)?  Do they even sell them anymore?

I?ve ordered a new boot HD, and will use that image I made to load it
up, and then swap it for the bad one.

I still wonder if I?m wasting my time because the problem is a bad motherboard.

any ideas?

P.S.  Thanks to all from all the great advice....

Howdy sherpaj-ga,

I am really bothered by the RAM failure with the new RAM.  That is why I
suspect either the motherboard or controller.  The "Hardware malfunction"
error you receive would tend to suggest it is the motherboard, but hard
to tell.  Could be a virus as the current behavior is also symptomatic of
that.

Dell does have some refurbished PowerEdge 600SC servers for sale, ranging
from $625.00 to $3,483.00 with special discounts good through Friday.

Go to Dell's website and search (in the upper right hand corner) on:

PowerEdge 600SC

Click on "Systems" on the next page, then click on "Buy Direct from
Dell" under "Small Business" on that page.  Click on "View Inventory"
on the next page for what systems they have available.

Once you get the new boot drive, install it on your machine, and
bring it up by itself without doing the image install.  You can
then see if the motherboard, etc. acts up.  If it doesn't then I
would install a virus scanner, put the old drive in as a slave and
scan it for any virii it might have.

Barring any virus infestation or hardware inconsistencies, you can
then do the image copy to the new drive.

Looking Forward, denco-ga - Google Answers Researcher