On Sat, 2007-10-06 at 14:47 +0200, Chitlesh GOORAH wrote:
On 9/27/07, Alexandre Magaz Graça wrote:
> Could someone give some hint about how to find out what is making it fail?
I encountered tracebacks only when creating live images on a F7 box?
Use a rawhide box instead with the git version of livecd-tools
Yes, my testing on Fedora 7 is rare. I run rawhide all the time on all
my boxes and have been doing so for many, many years now. I do quick
checks with Fedora 7 occasionally (and will do a larger set of them
before pushing an update to Fedora 7 within the next few weeks), but to
a large extent, if you're wanting to run on not the development tree and
are following livecd-tools.git, you may hit rough patches and need to
However my livecd images built on F8 automatically do a media check
Thus wasting about 15 minutes on a 2.7GHZ box.
Yeah, I accidentally (well, I knew it when I was writing the code, I
just forgot when I was committing and pushing) enabled verify as the
default when cleaning up some of the support for multiple kernels. As
dmc noticed, though, that should be fixed up in current .git.
Then there's kernel panics.
Need more details here to know what's going on. What exactly is the
panic you're seeing? The image I built this morning is fine.
The only way I could boot these livecd images is to edit the
arguments on grub:
* removing "check" ??? <<<<<<<<<<< why is this
by default ????????
* added enforcing=0
Hrmm -- on Thursday or Friday, there was something a little funky in the
policy package with fresh installs. If you were seeing problems that
were SELinux related, that could well be the cause. But that looks to
be fixed in today's rawhide