Re: Bug#561099: initramfs-tools: determination of "newest kernel" in update-initramfs -u uses improper sorting

["Jeffrey B. Green" <jeff@kikisoso.org>]

maximilian attems wrote:
> On Fri, May 07, 2010 at 11:50:39AM -0400, Jeffrey B. Green wrote:
[...snip...]
>>
>> In the end, I agree with Thiemo in that the manpage and the program's
>> behaviour are at odds.
> 
> just nuke the trunk linux-image, it is severely outdated.
> 

That's sidestepping the issue. Solving the immediate problem of getting
the correct initrd to update is not really the problem at the moment.
(However, I'd need to find a backup kernel to put in its stead if I did
simply purge the trunk kernel, and then deal with that kernel later on
when obsolescence descends on it.)

The point that I believe Thiemo was raising and that I agree with is
that the manpage says the software does something that it actually
doesn't. Admittedly the phrase is in the context of an example, however
it does say that it is to "Update the initramfs of the newest kernel"
which it logically isn't configured to do. One could be humorous about
this example and say that it could be said to "Update the initramfs of
hopefully the newest kernel" which is in some way what is actually
happening. But a more useable statement might be "Update the initramfs
of the newest kernel as lexically determined with its name" or "Update
the initramfs of the newest kernel as determined from the name sort
order". People at that point might complain about the choice of a
lexically determined kernel but at least they'll know what is going on
with the update-initramfs tool.

regards,
-jeff