Bug#224366: Relative links vs. symlinks.
On Sun, Dec 21, 2003 at 02:09:42PM -0500, Jeff Bailey <firstname.lastname@example.org> wrote:
> There are arguments for both sides - Having absolute links makes it
Keep in mind that it's easy to detect and fix at installation time
(symlinks pointing to nowhere just aren't going to work...), so there
shouldn't be a problem supporting both sides here (in effect, making this
argument a non-issue).
> harder to mount remote systems (for use in cross-compilers)
True (but certainly a very rare occasion).
> and harder to work with chroots.
It's hard to see why this should be the case... chroot works perfectly
with absolute links, as the / moves, too.
> In any event, having /usr be a symlink is an unusual configuration (on
> Linux, anyway... *g*), so I'm going to set the severity to wishlist for
I strongly disagree. Just because the bug happens rarely (as you claim, it
happens on ~90% of the machines I know of) does not mean it's not a valid
bug affecting people. Setting it to wishlist to keep the record clean
undermines the purpose of a bug database.
Wishlist does _not_ mean "we don't bother to fix this bug". It means there
is no bug...
(After all, having /usr a symlink is fully supported, and there is no good
workaround short of adjusting the symlinks).
In short: don't feel compelled to fix it (it would be easy, as it is
easily detectable wether it's a problem on a given configuration), but
keep honest by admitting that this is a bug.
----==-- _ |
---==---(_)__ __ ____ __ Marc Lehmann +--
--==---/ / _ \/ // /\ \/ / email@example.com |e|
-=====/_/_//_/\_,_/ /_/\_\ XX11-RIPE --+
The choice of a GNU generation |