[master/f20-branch][PATCH] More meaningful error for missing stage1 boot target

Martin Kolman mkolman at redhat.com
Mon Oct 21 17:46:03 UTC 2013


While I have not been able to actually reproduce the actual issue raised in 1012132,
I think the point that the current error message is confusing is valid.

As I was not able to find any robust method that would tell me what boot method
is currently being used, the first patch adds a new boot_method property
for the bootloader classes, that reports boot method type, if known.
Currently it either returns one of the MBR/EFI constants or None.
I've made the GRUB and GRUB2 classes report MBR and EFIGRUB reports EFI.
Are there any other I should mark as using MBR/EFI or any other distinct
boot method that would warrant its own constant ?

The second path is for blivet and just returns a more meaningful error message,
if the current boot method is known. The error messages are based on the ones
suggested in the bugreport, but I've dropped the boot partition size hints.


Martin Kolman (1):
  Add information about boot method for bootloader

 pyanaconda/bootloader.py | 19 ++++++++++++++++++-
 pyanaconda/constants.py  |  4 ++++
 2 files changed, 22 insertions(+), 1 deletion(-)

-- 
1.8.3.1



More information about the anaconda-patches mailing list