Hello Dave

What is happening is that the BIOS is going through the boot order and cannot find a bootable device. The NIC is the last boot device and that is why you are getting the PXE error. PXE is a network boot option.

Could you check to see if this server has a controller? If you have a RAID controller in the server then you should receive a CTL R or CTL C prompt to enter the controller BIOS. You could have a SAS 6 or PERC 6 controller in the server. Also, check your boot options in the system BIOS and see if there is an option for Hard Disk or a RAID controller in that list.

Thanks

Thank you for the replys.  I've made some progress but am now ung up again.  Please my actiosn below:

There are only 3 options listed at the BIOS screen 1)A drive (floppy) 2) Optical 3)NIC; no hard drives or RAID is listed.

There is a CTRL-C option when the RAID error is displayed.  I did that and the SAS configuration started.

Here what is listed:

Adapter: SAS1060E

PCI BUS: 08  PCI DEV: 00  PCI FNC:00  FW Rev: 0.20.48.00-IR  Status: Enabled  Boot Order 0

So I then dig into the configuration.  Boot Support is set to: Enabled BIOS and OS.

Under RAID properties I see both HD's but they are shown as INACTIVE.

I go to Manage Arrays and select Activate Arrays.  This is what is listed now for the drives(some info omited):

Slot Num: 0; RAID Disk = Yes; Hot Spr= No; Drive Status: FAILED; Pred Fail = No

Slot Num: 1; RAID Disk = Yes; Hot Spr= No; Drive Status: Primary; Pred Fail = No

This is where I'm at now.  Do I now need to Sync the drives or do I have a HD failure?

Thank you so much for your help.

-Dave

Dave

The SAS controller is capable of RAID 0 or 1. Could you check to see what the RAID level is on that array. If it is RAID 1 then one drive failed should not stop the array from functioning. I'm guessing that it is probably set up as a RAID 0 array.

Let me know what RAID level it is and I'll let you know what our options are.

Thanks

Ok, really dumb question....how do I figure out if it is a 0 or 1 array?

In the controller BIOS select RAID Properties>View Existing Array

It should have all of the array information listed there.

Thanks

Well.....when I go to the controller BIOS there is no option to look at the RAID anywhere so I return to the SAS configuration utility. Here is what is listed under RAID properties:

Array 1 of 1; ID=Dell Virtual Disk 1080l; Type=Integrated Mirror; Scan Order=0; Size=75340; Status=PermDegrd

Since I reactivated the array a few steps ago, I now have a E1810 HDD 0 fault scrolling on the LCD display.

Well, no more PXE error but instead I now get NTDETEXT Failed message.  I removed HDD 0, rebooted, then moved HDD1 to slot 0 and tried again, same thing.

Thanks,

Dave

Okay, it's a RAID 1 mirror. Disk 0 is in a failed state and disk 1 is online.

You should still be able to boot with one disk failed in a RAID 1. Disconnect the failed drive from the controller(HDD 0) and see if it will boot with just HDD 1 connected. If it still gives the PXE error then move HDD 1 over to slot 0 on the cable and see if it will detect there. Leave the failed drive out of the server or disconnected.

Also, once you remove the failed drive check to see if there is an option for Hard Disk or the SAS controller in the boot options menu of the system BIOS.

Thanks

Ok, I removed HDD 0, cheked the BIOS and see that C drive now shows up but at step 4.  I will move that to before the NIC and see if it will boot.  Standby.

Good

What we have done is verify the hardware. We know that the HDD in the server is functional, both cables are good, and the controller is functional. It appears that HDD 0 had failed to a point that it was causing the controller to not initialize properly, so that drive definitely needs to be replaced.

The current problem of "NTDETECT Failed" has to do with the boot manager of the operating system. It appears there was OS corruption that occurred due to the drive failure. You can attempt to repair this, but you will likely have another missing or corrupt file error after you fix NTDETECT. If there is not critical data on the drives, or if you have a backup then I recommend that you just reinstall the OS.

If you want to attempt the repairs then let me know what OS you are running and I will try to find some Microsoft KB articles on repairing NTDETECT.

Thanks

Ok Great.  So I should replace the defective drive, and then reimage the machine?  The only question I have is I have 2 80GB HDD and the replacement I have is 160GB.  That shouldn't cause any issues should it?  

If that is all I have to do then I will work on that first thing in the morning and post an update.

Thank you so much for your help, it is greatly appreciated!

-Dave

Yep, that is what you need to do. Yes, the 160GB HDD should work. As long as it is the same architecture(SAS/SATA) as the drives in the array and it is the same size or larger than the smallest drive in the array it will work.

Thanks

I inserted the new drive, and started the re-image.  I have the option of recovering the image to Disk 1 (new HDD) or disk 2 (existing HDD).  Should I recover it to Disk 2 and then it will get mirrord automatically to Disk 1?  Disk 1 is showing it as being uninitialized.  Is there any other steps I need to take.

Again, thanks for your help.

You need to get the array working again before you start reimaging. If the new drive does not automatically start rebuilding then you can set it as a hot spare to get it to pull back into the array. Alternatively, since you are wiping the data on the drives anyway you could just delete the current array and create a new mirror with the two drives. This would allow you to skip the rebuilding process.

Thanks