[TriLUG] Kernel oops after upgrade in Debian

[T. Bryan]

Okay, I'm fairly new to Debian, and I'm not sure where to go with this one. 
One of my main machines was installed with an old netinstall image (made when 
sarge was testing, I think).  I've been running it off of testing, but the 
kernel is still pretty old: 
# uname -a 
Linux fezzik 2.4.27-1-386 #1 Fri Sep 3 06:24:46 UTC 2004 i686 GNU/Linux

So, I haven't been updating that machine recently, and I decided to do some 
work on it again.  I started aptitude, chose to upgrade my installed 
packages, selected a few new ones that I wanted to try, and upgraded. I 
didn't notice any errors when running the upgrade.  I rebooted, and now the X 
Server crashes as soon as kdm tries to come up.  The last things in the 
syslog before the X server crashes is 

Mar 19 14:58:58 localhost kernel: [drm] AGP 0.99 Aperture @ 0xd0000000 32MB
Mar 19 14:58:58 localhost kernel: [drm] Initialized mga 3.0.2 20010321 on 
minor 0
Mar 19 14:58:58 localhost kernel: Unable to handle kernel paging request at 
virtual address 0001f0a8
Mar 19 14:58:58 localhost kernel:  printing eip:
Mar 19 14:58:58 localhost kernel: f0a7cc8f
Mar 19 14:58:58 localhost kernel: *pde = 00000000
Mar 19 14:58:58 localhost kernel: Oops: 0000
Mar 19 14:58:58 localhost kernel: CPU:    0
Mar 19 14:58:58 localhost kernel: EIP:    0010:[<f0a7cc8f>]    Not tainted
Mar 19 14:58:58 localhost kernel: EFLAGS: 00013202
Mar 19 14:58:58 localhost kernel: eax: 0001f0a8   ebx: 00000000   ecx: 
00000010  edx: 00000010
Mar 19 14:58:58 localhost kernel: esi: 000000c0   edi: ec7946d0   ebp: 
ffffffed  esp: ecaabf30
Mar 19 14:58:58 localhost kernel: ds: 0018   es: 0018   ss: 0018
Mar 19 14:58:58 localhost kdm[1226]: X server for display :0 terminated 
unexpectedly
Mar 19 14:58:58 localhost kernel: Process Xorg (pid: 1244, stackpage=ecaab000)
Mar 19 14:58:58 localhost kernel: Stack: 00000000 ec7946d0 eecfedf0 efc2b310 
c0131d89 eecfedf0 ec7946d0 ec7946d0
Mar 19 14:58:58 localhost kdm[1226]: Unable to fire up local display :0; 
disabling.
Mar 19 14:58:58 localhost kernel:        eecfedf0 00000000 c0130e86 eecfedf0 
ec7946d0 00008002 ef31f000 4018eff4
Mar 19 14:58:58 localhost kernel:        bffff698 c0130da1 eecff4a0 efc2b310 
00008002 eecff4a0 efc2b310 bffff708
Mar 19 14:58:58 localhost kernel: Call Trace:    [chrdev_open+55/65] 
[dentry_open+221/368] [filp_open+65/73] [sys_open+48/121] [system_call+51/56]
Mar 19 14:58:58 localhost kernel:
Mar 19 14:58:58 localhost kernel: Code: 8b 08 85 c9 8b 5f 10 89 c2 74 16 51 e8 
f5 a6 69 cf 5a 31 d2
Mar 19 14:59:53 localhost kernel:  spurious 8259A interrupt: IRQ7.

Anyway, any suggestions?  I don't even know where to start.  I've searched a 
bit on google, but I don't see anything simple or consistent in the search 
results.  Should I expect problems this bad when I upgrade a lot of packages 
at once in Debian?  Or is this a package or package manager bug?  I can log 
into the box, and I'm currently ssh'd into it and running KMail forwarded 
across SSH to another Linux machine, so the broken machine can't be too 
broken.  

Any help or pointers would be appreciated.

Thanks,
---Tom