Red Hat Bugzilla – Bug 36277
anaconda crash with the flag --noprobe in xconfig
Last modified: 2007-04-18 12:32:41 EDT
From Bugzilla Helper:
User-Agent: Mozilla/4.7 [en] (X11; I; Linux 2.2.14-15mdk i586)
In the kikstart file with the option --noprobe in xconfig
we crashed in xf86config.py in the line 174 :
with the error : Sequence index must be integer
Steps to Reproduce:
1.xconfig --noprobe --card "ATI Xpert 98" --server "Mach64"
Actual Results: crash in anaconda
With the --noprobe flag in the function configureX(self,server,card,..)
probe() is not call but setVidcar(self,card) is called.
In the function setVidcar()
vidCards is not initialized , probably a call to cards(card) is forgotten.
we have also the same problem if we don't set the flag --noprobe
with a video cart like "ATI Xpert 98" . With this ATi card kudzu.probe
return (none,unknow,ATI|unknow) so it's insufficient to find the server in
the file Cards.
We have seen this issue before with 6.2. I'm pretty sure that this is fixed in
Red Hat Linux 7.0 (and also 7.1, which was released yesterday).
Ok but is it possible to have an installer update for Red hat 6.2.
We generally do not release anaconda updates after a new version has been
released, but yes, an updates disk is technically possible. If you can attach
the complete traceback, I'll try to make an updates disk for you. It depends on
how complex the fix is. If it's a couple of lines, then I'll be able to help
you out, but I can't spend lots of time working on a release that is over a year
I understand your problem, but we have now to install about 50 PC with Red hat
6.2 for our customer. And we must also give to him a kickstart disk installer.
If you have only changes on the python files , i can include them into the disk
Thank you for your help.
Ok, I'll see what I can do. Like I said in my previous post, if you can attach
the complete traceback, that will help me in debugging.
Closing due to inactivity. Please reopen if you have any more info.