Re: postinstall of x11-common fails
On Mon, Jun 05, 2006 at 02:42:57PM -0400, hendrik@topoi.pooq.com wrote:
> On Sun, Jun 04, 2006 at 08:31:28AM -0400, hendrik@topoi.pooq.com wrote:
> > On Sun, Jun 04, 2006 at 12:40:21PM +0200, Toon Moene wrote:
> > > Like this:
>
> Deep in the middle of the xorg upgrade on my AMD-64 [running etch].
> Actually, I'm probably only barely started.
My son, lcpublic@gmail.com, is having a similar problem on a 32-bit
Athlon, so the problem appears not to be AMD64-specific.
I'm crossposting this to debian-user, and followups should be sent
there.
-- hendrik
>
> Following directions on the wiki page [wiki.debian.org/Xorg69to7],
> I first upgraded x11-common, and was told the upgrade failed;
> specifically because the symbolic link /usr/include/X11
> did not exist. This was while processing
> /var/cache/apt/archives/x11-common_1%3a7.0.20_amd64.deb
>
> Should I create this link manually? Or will that break the package
> (whatever it is) that is *supposed* to have installed it?
> If so, where should it point?
I got over it by hand-deleting almost every package that
depended on x11-common, directly or indirectly.
This took a long time.
Then I could delete x11-common and install it again.
And Lo, it turned out that /usr/include/X11 was not the link whose
absence was a cause for great distress, but an ordinary directory.
Incidentally, I had tried stepping outside aptitude to do a apt-get
remove and install with the -force-yes option, to, as far as I can tell,
no effect whatsoever. It gave me a few complaints and nothing else
changed. Is this a bug? Is this out-of-date documentation? Or is this
maybe the same problem I'm having trouble with in aptitude?
My son, Henk <lcpublic@gmail.com>, on a 32-bit Athlon runnin etch, can't
use my
method.
He has an order of magnitude more broken packages. The trouble is that,
every time he specifies another package for deletion, several more
break. My list of broken paclages peaked at about 80. His has
ballooned to 500 and still growing. So he backed out of the proposed
deletions with ^U and is looking for another method.
Help is much desired.
--hendrik
Reply to: