I rather optimistically bought an Asus a7n8x Deluxe
motherboard (with the Nvidia nForce2 chipset), along with
an Athlon XP 2600+, and tried to install Severn2. Multiple
times. Each time it got partway into installing packages and
then a kernel panic or freeze.
I had hoped it would work, since I read (mis-read?) that
the a7n8x runs on RH 9 without patches. But that doesn't
seem to be the case either - each time I try to do a
gcc boostrap (on a different partition than the one I
tried to install Severn2 on), gcc fails in a different way.
My guess is some disk-related problem, such as DMA.
(I have moved a jumper to disable SATA, so the kernel would
not timeout for the non-existent drives.) Any suggestions?
[This is a re-port. The previous posting has spent almost
to 24 hours awaiting "moderator approval", so I decided I
needed to subscribe.]
I'm currently running redhat9, with various gnome 2.4 packages installed
from the nyquist repository (a 3rd party packager). Am I going to have
problems doing an upgrade install of fedora? How does the installer handle
this kind of situation? I saw in the release notes some mention of problems
with upgrading systems with ximian desktop installed, so I'm guessing there
might be similar problems with other non-redhat packages installed - am I
right, and if so, what would be the best way to proceed?
My Severn2 test system won't reboot. It seems to go so far and just stop. It won't even save my volume settings. In
order to get it to turn off I have to hit the power button, because if I wait and wait it still won't reboot. Please
help ASAP with this one.
Whoa... That is weird...
After installing 0.94, everything seemed peachy. I then installed
various packages via yum, and compiled additional src.rpms. Proceeded
to build vanilla 2.6.0-test6 and the nvidia module.
Everything still good to go.
Recompiled the kernel again as I forgot netatalk support and rebooted.
Went to do a yum update and yum bombs out with python errors and a
message saying that /var/cache/yum/<some-file> didn't exist.
Lo and behold, the /var/cache/yum directory didn't even exist.
Created the directory to see what that would get me, and then I notice
that yum reported 404 errors due to it believing $releasever is 9.0.93
Ran up2date and it also thought the system was 9.0.93.
cat /etc/redhat-release reports Fedora Core release 0.94 (Severn)
rpm -q redhat-release reported not-installed. So I did a rpm --rebuilddb
rpm now reports redhat-release-9.0.93-4. What?
It is reporting all kinds of old 9.0.93 packages, although the new
packages are installed, and during 0.94 install, I told it to format
everything but my separate /home partition.
Seeing as how it was far from a stock 0.94 install, I don't expect any
help, and certainly couldn't file a bug report, but I thought I would
send it to the list to see if anyone has experienced anything of the
How the heck did it pull 9.0.93 packages out of its butt when the
partitions were supposedly formatted? Hit a glitch in the space-time
I won't have time to reinstall until this weekend. :( Keeping my fingers
crossed that she holds together until then...
James Drabb wrote:
> On Mon, 2003-09-29 at 19:32, Deryk Lister wrote:
> > I posted the following message to the TransGaming messageboards, and
> > will also post it here in case anyone has any ideas! The problem, I
> > think, is with the kernel but I've yet to put my finger on it.
> > This thread seems to outline a similar problem, and the same method I
> > tried (grabbing a vanilla kernel and compiling with a Fedora .config
> > file) but whilst it worked for this person it didn't work for me:
> > http://www.redhat.com/archives/fedora-test-list/2003-September/msg00218.html
> > Any suggestions/workarounds appreciated!
> > (I'd quite like to see this fixed - without workarounds - before Fedora
> > is released, if anything to prevent newbs getting put off. Will help in
> > any way I can, but note I'm not a programmer)
> > Cheers
> It could be prelinking or nptl. You could try to set PRELINKING=no in
> You can also try to run winex with LD_ASSUME_KERNEL=2.4.1, e.x.
> LD_ASSUME_KERNEL=2.4.1 winex3 eqgame.exe
> Or just export LD_ASSUME_KERNEL=2.4.1 before you run winex3 from a
Thanks for the suggestions, unfortunately no joy on either of them :/
I was just reading about the fedora project's goals on the website, and
it says that it should be easier to get new packages included. What is
the process for this? Could snort be included?
RHCE+AS/IBM Certified Specialist
The Starken Group
I've been playing with Test 2 all weekend, looking at the emails on the
list and doing what all was suggested, so far things work, so nothing
extremely new to report.
My problem with the firewire drive still exists, Ben Hsu's,
rescan-scsi-bus.sh script solution works like a charm.
A new problem I've encountered now, is DHCP/Internet connectivity
problems. Prior to Test 2, I would get a DHCP connection from my cable
modem, which would assign the ip, and voila, i'd be surfing the net like
a happy camper. I could do this for 3-4 days on end, usually after which
i would turn the computer off, or reboot into windows for some games..
either way.. now, after a few hrs, I seem to lose my gw, or something,
as route takes forever to finish, and if I do a service network restart,
things start working again immediately.
I'm in San Antonio, I haven't tested this time out issue with windows, I
just left for work so I've left the computer on in Windows bootup,
letseee if it loses it's ip/gateway information there.. if so, then I
know it's a time warner problem.. otherwise it might be a Test 2
problem. If it is.. does anyone know what could be causing this ? Some
sort of power management ? turns network card off ?.
I'm using 3c509 card.
Every resource I access with samba requires a password multiple times.
For example if I copy a folder to my Ferdora box that has 13 files in
it, I get prompted for my password 13 times. Very annoying.
Any idea what I'm doing wrong?