Pre surgery advice

Hi everybody,

I am the systems administrator for a company with an OLD Dell Poweredge 4400 server with eight SCSI 36 gigabyte hard drives (PERC 3/DI controller) in it running Windows standard server 2003 that should have been retired a few years ago but this company got itself locked into two proprietary customer data base applications and they do NOT have the budget to port all there data over to more modern applications......so my only choice is to keep this old Dell server running.

If this server fails there company probably will also. So needless to say I am VERY nervous about fixing it and I wanted to get some advice before I start any repairs probably this next weekend.

Currently this server is running OK as the main domain controller for about 30 workstations.

But when the system starts up it says:

"Following containers have missing members and are degraded:
Container#0-RAID-5"

It then gives the options to enter the RAID configuration utility or just ignore and continue to boot the system. But it does also say that the RAID array is "critical".

Now.... It is my understanding that each of these SCSI drives are hot plug-able.

Each of these hard drives have three LEDS on the front of it from left to right:
A) Power
B) Access (read or write)
C) Failure

Currently ALL eight drives have power lights and NONE of them have lit "failure" LEDs.

But two of them (number 6 and 8) NEVER show active access (read/write) LEDs. The LEDs ARE working because they light up briefly during the POST.

I am guessing that those are the two drives that have failed but I am not positive.

I have 4 brand new 36 gigabyte replacement SCSI hard drives.

My question is;
Without having lit failure LEDs is there any other way to tell which of the 8 drives is not working correctly?
I looked in the RAID configuration utility and the only option that might help is: "Verify Disk Media"..... But I am not sure what that does and I do not want to make the situation worse.... So I thought I would ask for some input first.

Also.... (Once I verify which drives are bad) can I simply take ONE defective drive out at a time and install a new drive and let it rebuild? And should I do this when the server is off or when I am in the RAID configuration utility or when Windows is running.

Thanks in advance for your input (smile).