X_Entity
01-09-2001, 08:37 PM
Wow its been a busy evening. I got home from work early knowing full well that I'd missed the courier who was going to deliver the parts I've been waitng for since 05:01 24 Dec 2000. After a quick phone call to the courier and a 40 mile round trip later I arrived back home with all the goodies. I'd stripped the old machine the night before in preparation for this huge task. Installing the new bits was pretty rountine as its what I do as a sidline anyway the fun started when I came to start the machine up. I realised instantly that I'd probably lept in with both feet and that getting it all to work again was going to be a task.
After about half an hour of playing with the XF86Config file I managed to get X running (ver 3.3.6) so I started the kernel recompile which I've been waiting to do for the last few days. whilst that was going on I decided to install X4.0.2 again and make the most out of the new graphics card (Hercules 3D Prophet II 32Mb). That all went suprisingly smoothly at least as far as 2D goes (I haven't attempted 3D config yet) all was going well the kernel only took 6 attempts to compile correctly and I'd chosen a few options that were new to me as well i.e devfs and some lesser options. Not being one to read documentation I wasn't really concerned about how devfs did its thing as long as it worked.
After a restart things started going wrong...
the Kernel had picked up the SCSI info which is always good. but then I got an error message. Not the usual kernel panic I get but a major node problem that fsck can't fix. Not a problem I say I'll just boot up with the old kernel and recheck the options I chose and make sure everything is ok. So the machine reboots and lilo tells me I only have 1 kernel available. I realised that after editing lilo.conf I'd forgotten to run lilo. Ok it looked like I was going to have to figure this one out from single user mode and hope it would all work out.
After trying lots of different things it suddenly occured to me that I hadn't actually looked to see if the nodes existed.
cd /dev
ls
ahhh I say I seem to be missing about 500 devices or so. after following a string of directories (/dev/scsi/host0/bus0/target0/lun0/part1)
I find what I'm looking for what I used to call sda1 which is no longer called sda1. After much fiddling with fstab and a short break whilst I got some food I try again.
It booted all the way up to where wdm was called and then froze. Not a problem cos by now I've corected the lilo problem and boot back into the system and actually read the documentation on devfs to find out how to make it work.
a quick download later and here I am
Debian unstable/testing
kernel 2.4.0 WITH DEVFS!!!!!
X4.0.2
all running on my nice new
AMD T/Bird 1000Mhz
Gigabyte GA71XE4 Mobo
Hercules graphics card
256Mb RAM
18GB of SCSI drives
and all my devices working perfectly.
Happy happy joy joy
Any way sorry if bored you but no one made you read this.
I had to share it with someone and its 00:45 so I think I should be let off.
Oh one thing I did forget to mention is that with the new 2.4 kernel my sound card works properly for the first time.
Life is good (today).
Thanks for reading all the way to the bottom
After about half an hour of playing with the XF86Config file I managed to get X running (ver 3.3.6) so I started the kernel recompile which I've been waiting to do for the last few days. whilst that was going on I decided to install X4.0.2 again and make the most out of the new graphics card (Hercules 3D Prophet II 32Mb). That all went suprisingly smoothly at least as far as 2D goes (I haven't attempted 3D config yet) all was going well the kernel only took 6 attempts to compile correctly and I'd chosen a few options that were new to me as well i.e devfs and some lesser options. Not being one to read documentation I wasn't really concerned about how devfs did its thing as long as it worked.
After a restart things started going wrong...
the Kernel had picked up the SCSI info which is always good. but then I got an error message. Not the usual kernel panic I get but a major node problem that fsck can't fix. Not a problem I say I'll just boot up with the old kernel and recheck the options I chose and make sure everything is ok. So the machine reboots and lilo tells me I only have 1 kernel available. I realised that after editing lilo.conf I'd forgotten to run lilo. Ok it looked like I was going to have to figure this one out from single user mode and hope it would all work out.
After trying lots of different things it suddenly occured to me that I hadn't actually looked to see if the nodes existed.
cd /dev
ls
ahhh I say I seem to be missing about 500 devices or so. after following a string of directories (/dev/scsi/host0/bus0/target0/lun0/part1)
I find what I'm looking for what I used to call sda1 which is no longer called sda1. After much fiddling with fstab and a short break whilst I got some food I try again.
It booted all the way up to where wdm was called and then froze. Not a problem cos by now I've corected the lilo problem and boot back into the system and actually read the documentation on devfs to find out how to make it work.
a quick download later and here I am
Debian unstable/testing
kernel 2.4.0 WITH DEVFS!!!!!
X4.0.2
all running on my nice new
AMD T/Bird 1000Mhz
Gigabyte GA71XE4 Mobo
Hercules graphics card
256Mb RAM
18GB of SCSI drives
and all my devices working perfectly.
Happy happy joy joy
Any way sorry if bored you but no one made you read this.
I had to share it with someone and its 00:45 so I think I should be let off.
Oh one thing I did forget to mention is that with the new 2.4 kernel my sound card works properly for the first time.
Life is good (today).
Thanks for reading all the way to the bottom