Re: Bug#49948: liloconfig makes bad lilo.conf when from scratch
>On Sun, 21 Nov 1999, Adam Di Carlo wrote:
>
>> Yes -- lilo should handle this gracefully.
>>
>> >>At the point where we build the base system, we do not know
>> >>what the machines disk layout will be. lilo must tolerate this or
>> >>else it can't be in base, which is also unacceptable.
>> >
>> >Pardon? If you don't know the disk layout you've got no business
>> >running lilo (which is what liloconfig does...) Do you mean that
>> >you need to be able to configure the package in some sort of
>> >no-operation way, or what?
>>
>> Yes -- it should exit out gracefully when fstab is missing
>> or empty.
>Oh?
>You want lilo to exit gracefully in case of /etc/fstab is missing?
>Ok, that would fix your problem for the boot floppies, but it would be
>broken for the "normal case", where lilo can't install and be configured
>without fstab file.
>Is there a way I can test whether liloconfig is executing in the initial
>base2_2 dir so I can make a special case? (like create a
>/tmp/no-fstab-ok-for-liloconfig empty file for example)
I dunno -- let me try to find some appropriate mechanism and let you
know.
Boot-floppies folks, any ideas?
.....Adam Di Carlo....adam@onShore.com.....<URL:http://www.onShore.com/>
Reply to: