[H-GEN] System won't boot, motherboard at fault?

Tim Browne dugb at netspace.net.au
Mon Aug 9 02:25:22 EDT 2004


Jason Parker-Burlingham wrote:
> [ Humbug *General* list - semi-serious discussions about Humbug and     ]
> [ Unix-related topics. Posts from non-subscribed addresses will vanish. ]
> 
> Brief synopsis:
> 
> So I've removed one disk of a Linux software RAID-1 array from a
> system to do some maintenance (read:  move to another motherboard).
> But the system won't boot.  I've double and triple-checked all the
> jumpers and BIOS settings and everything looks fine.  Instead of
> booting correctly, GRUB runs, presents a menu and promptly does a soft
> reset of the system.  I'm looking for hints on what to check next.
> 
> Further details are beneath the form-feed.
> 
> The long version:
> 
> The OS installed is Red Hat 9.  Originally the drive I'm working from
> had no boot loader installed (I think this is a bug in Red Hat 9?) but
> I've gotten past that by running
> 
>    grub> root (hd0,0)
>    grub> setup (hd0)
>    grub> reboot
> 
> as per instructions.  The kernel is 2.4.20 or so (I'd have to switch
> keyboard and display again to check) and boots on the other
> motherboard without complaint.  The new board is an ABIT IS-20, the
> host drive is a simple 80GB Western Digital with three primary
> partitions of type 0xfd:  /boot, /, and /export.  If I add the
> "fstest" command to the GRUB command list, I see what looks like a
> list of blocks loading, and then "Loading Linux" before the system
> resets.
> 
> The system in question has no other peripherals attached:  the drive
> is set to a master, there's no slave drive present, no devices or
> cable on the second channel, or either of the SATA channels.  I've
> disabled various extra devices (sound and so forth) in the BIOS to
> attempt to rule out some kind of resource conflict to no avail.  I've
> removed and reseated the RAM and checked it's within spec (it is).
> 
> Otherwise, I'm out of options.  I can cannibalize one of my other
> systems for parts if necessary but wouldn't mind some inspiration on
> what to check next before I do.
> 
> jason

I don't know that there is an easy solution to this problem. It looks like your 
kernel is starting to boot then loading a driver that is incompatible causing a 
  machine reset. As the kernel never completely boots there is no way to look 
through a log to see what is going on. The only option may be to boot from a CD 
or floppy compile a new basic kernel and see if that will load.

Tim Browne.





More information about the General mailing list