Alienware M11x won't POST after install

In the interest of misusing (read:REALLY using) my Alienware M11x I decided to blow away Windows 7 and install Debian. That was great until Iceweasel / I couldn't jive with their distro. So I downloaded FreeBSD 8.2 AMD64 and installed it. When I rebooted my system would not POST.

I was scouring the forums last night and LOST A LINK (doh) to a post describing a bug with modern BIOS systems that have an issue when (let's see if I get this right) the MBR does not sit squarely on a 'slice' of the hard disk.

In a frustrated effort I phoned Dell tech support. The tech had me disassemble my laptop down to the motherboard to get to the CMOS battery! Unable to solve my problem after I reassembled my laptop they issued an RMA.

Long story short is I remembered that post I read very late last night in the AM. I removed my HDD and wouldn't you know POST!

Can anyone shed more light on the issue or link me to this post? I've been searching but have been able to find any post other than one about LED lights.
 
aciddeath said:
I was scouring the forums last night and LOST A LINK (doh) to a post describing a bug with modern BIOS systems that have an issue when (let's see if I get this right) the MBR does not sit squarely on a 'slice' of the hard disk.
FWIW, I have several machines with misaligned partitions / slices - they all boot fine.
Here is an example:
Code:
root@kg-vm# uname -a
FreeBSD kg-vm.kg4.no 8.2-STABLE FreeBSD 8.2-STABLE #3: Fri Mar  4 21:16:39 CET 2011
     root@kg-vm.kg4.no:/usr/obj/usr/src/sys/GENERIC  amd64
root@kg-vm# grep GEOM /var/log/messages
Jun 16 19:23:16 kg-vm kernel: GEOM: da0: partition 1 does not start on a track boundary.
Jun 16 19:23:16 kg-vm kernel: GEOM: da0: partition 1 does not end on a track boundary.
Jun 16 19:30:38 kg-vm kernel: GEOM: da0: partition 1 does not start on a track boundary.
Jun 16 19:30:38 kg-vm kernel: GEOM: da0: partition 1 does not end on a track boundary.
Jun 16 19:43:32 kg-vm kernel: GEOM: da0: partition 1 does not start on a track boundary.
Jun 16 19:43:32 kg-vm kernel: GEOM: da0: partition 1 does not end on a track boundary.


aciddeath said:
Long story short is I remembered that post I read very late last night in the AM. I removed my HDD and wouldn't you know POST!
Oh, that is a different problem. Some BIOSes doesn't like it when the partitions (FreeBSD: slice) on the boot hard drive doesn't conform to what they want. Seen in two variants:

  1. a FreeBSD partition type (165) present on the boot hard drive is enough to make the machine refuse to boot at all. This is a very old bug, I am not sure if it still is present in any machine alive today. I first experienced this on a IBM ThinkPad (can't remember the model) many years ago, and it was fixed with a BIOS update.
  2. the BIOS require a certain partition type to be present on the hard drive, or the machine won't boot. I have not experienced this myself, just read about it (on this forum I believe).
Hope this helps.
 
Tingo, thank you. That information was very helpful. I began reading some more threads about people having difficulty booting a FreeBSD install with a SATA controller. So I went into my BIOS and changed 'SATA Operation' from 'AHCI mode' to 'ATA mode'.

Wouldn't you know -- welcome to FreeBSD!

Thanks all for the helpfulness and replies.
 
If you really want AHCI, you can experiment with various BIOS settings. For me, when I enable AHCI in BIOS, I won't be able to boot from CD.

Now I keep three profiles saved in BIOS:
  1. Boot from HDD
  2. Boot from USB
  3. Boot From CD

Perhaps your BIOS also has some "bug" that is triggered in various configurations.
 
aciddeath said:
Tingo, thank you. That information was very helpful. I began reading some more threads about people having difficulty booting a FreeBSD install with a SATA controller. So I went into my BIOS and changed 'SATA Operation' from 'AHCI mode' to 'ATA mode'.

Aha - I had completely forgotten about that one!
Well, glad you got it working. Enjoy your FreeBSD experience (and learning). :)
 
Back
Top