EFI install (F15 beta rc2)

[Josep Puigdemont]

El dl 18 de 04 de 2011 a les 09:12 -0400, en/na James Laska va escriure:
> On Sun, 2011-04-17 at 12:41 +0200, Josep Puigdemont wrote:
> > Hi,
> > 
> > I followed the instructions for installing FC15-Beta (RC2) on an EFI
> > system, and can confirm that the "Testcase Boot Methods efidisk.img"
> > passed without problems, at least for x86_64. Can I put a check mark on
> > that test [1], or should someone from the QA team do that?
> 
> Please feel free to add your result.  Thanks for testing!

Done!

> 
> > Then I also want to report a couple of issues:
> > 
> > The installation went smoothly, but when it finished and was about to
> > reboot, the kernel panicked (didn't copy the stack trace sorry). I don't
> > know if this is a known issue, I could file a bug report if you feel
> > it's necessary.
> 
> I'm not aware of any kernel panics on reboot.  Unless we can get the
> stack trace, I don't think it's worth filing a bug at this point.

I could also reproduce this issue with the final Beta released today,
filed it here:
https://bugzilla.redhat.com/show_bug.cgi?id=698004


> 
> > Then I run into a problem when booting Fedora, as it just brings me to
> > the grub prompt. I figured grub wasn't able to read the config file, but
> > that doesn't seem to be the case. The grub.conf file in the EFI
> > partition seems fine. Then I tried to manually type at the grub prompt
> > the relevant commands I found in grub.conf:
> > 
> > grub> device (hd0) HD(...)
> > grub> root (hd0,1)
> > Error 5: Partition table invalid or corrupt.
> > 
> > That error seems a bit odd, as the partition table appears to be fine
> > when I boot from a live CD disk (it's anaconda's default partition
> > table).
> > 
> > To test further, I did the EFI install of F14 and found the same issue.
> > Could it just be a problem with my hardware, perhaps?
> > 
> > Any pointers on how to debug the issue would be appreciated.
> 
> Booting the installer with 'rescue' would be good.  You can then have it
> locate your system partitions, and you can access installation log
> files.  Those files will help ensure that the installer did everything
> right (/root/*log and /var/log/anaconda/*).  Those files, along with
> 'parted' output, will help determine how your disk(s) are partitioned.

I have access to all this information from the drive. I didn't see
anything suspicious in the logs, but on the other hand I don't know what
I should be looking for :)

It's probably best that I file a bug. What component should that be,
anaconda or grub? I guess the latter, but it doesn't hurt to ask.
And what are the relevant log files I should attach, all of them?

> Do you recall where you directed the installer to write the bootloader?

I just let the installed decide, but since there is only one HD, I
suppose it used /dev/sda.

> If you have time, I might suggest doing a non-EFI install with the same
> partition selections, and compare results.  If the problem only happens
> with EFI installs, that'll help narrow down the issue.

Done. With a non-EFI install everything worked fine. I can boot without
problems, so it seems to be only an issue with the EFI install.

Thanks!

/Josep