Re: Why udev does not use update-rc.d in its postinst
On Monday 11 September 2006 19:17, Frank Küster wrote:
> md@Linux.IT (Marco d'Itri) wrote:
> > On Sep 11, Frank Küster <email@example.com> wrote:
> >> That's the version in testing, but the source package in sid also has
> >> all that's needed to get it in again, unless there's a hard-to-see
> >> subtle error.
> > Like the update-rc.d bug discussed here in the last few days.
> Which wouldn't result in the udev binary package's postinst missing the
> update-rc.d call, as George asserted.
I was terribly wrong on that assertion, sorry for the confusion. I was looking
at mine own udev package (this is not like I don't trust the official one, of
course, just the opposide) which doesn't call update-rc.d intentionaly. The
package udev (0.100-1) currently in Sid just is fine.
> And this bug probably would
> disable more init scripts than this one (I didn't read it in detail,
> since by chance I didn't upgrade to any problematic version).
Right. This hosed my network and it took me quite awhile to get some clue
about what went wrong, i.e. to suspect the sysv-rc package. After having
inspected the changelog of sysv-rc and update-rc.d-recover.gz file I tried to
dpkg-reconfigure most important stuff: linux-image-* (to fix modules.dep),
udev, ifupdown, module-init-tools, and the list suggested by the recovery
script. Now after I'm networked again I can see the mail from Kevin B.
McCarty which is the case here. Yes, I was lucky I had sysv-rc 2.86.ds1-20
installed in the first place. Thanks for the help.
pub 4096R/0E4BD0AB 2003-03-18 <people.fccf.net/danchev/key pgp.mit.edu>
fingerprint 1AE7 7C66 0A26 5BFF DF22 5D55 1C57 0C89 0E4B D0AB