On Mon, Sep 10, 2001 at 05:16:27PM +0100, Julian Gilbey wrote: > On Sat, Sep 08, 2001 at 06:55:56PM -0400, Steve M. Robbins wrote:
> > postrm > > ------ > > > > An installed shared lib has been removed from the system just before > > "postrm remove" is run. This is the proper time to call "ldconfig" > > to notify the system of that fact. Current policy has this as a > > SHOULD requirement. I still haven't found out why it is not MUST. > > I wonder whether it's because the installed library won't be found if > the postinst ldconfig is not run, but an extra dangling symlink or > similar following a remove is not such a problem. Yes, it is less harmful to have a dangling symlink than to not have the symlink when required. My conjecture is that the former was only a SHOULD to avoid creating RC bugs. I was hoping that someone on -policy would remember whether that is the case or not. By the way, my investigation of 6507 packages (the i386 archive of a few days ago) show that perhaps 106 (or 1.6%) postrm scripts lack "ldconfig" when they should have it, so changing the requirement for postrm isn't really creating a huge whack of new bugs. For reference, 64 postinst scripts have the same problem (which *is* RC) -S -- by Rocket to the Moon, by Airplane to the Rocket, by Taxi to the Airport, by Frontdoor to the Taxi, by throwing back the blanket and laying down the legs ... - They Might Be Giants
