[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Desktop upgrade report / aptitude from sid



Hi Steve,

Am Montag, 26. März 2007 00:01 schrieb Steve Langasek:
> On Sun, Mar 25, 2007 at 04:52:24PM +0200, Rainer Dorsch wrote:
> > -> aptitude refused the upgrade: No soultion found for the conflicts
> >
> > open: 11182; closed: 4969; defer: 0; conflict: 3
> > No solution found within the allotted time.  Try harder? [Y/n]
> >
> > Not sure if aptitude from etch would have done better here. I went with
> > apt-get for the critical part.
>
> I don't suppose you have a /var/lib/aptitude/pkgstates from before the
> upgrade?  I don't really expect a particular aptitude upgrade failure to be
> reproducible using just a package list.

Julien mentioned that there might be a backup. I check that later....

> > -> postgres-client upgrade confused apt/dpkg completely. Installation
> >    of all following packages failed
> > Preparing to replace postgresql-client 7.4.7-6sarge4
> > (using .../postgresql-client_7.5.22_all.deb) ...
> > install: `/var/lib/postgres/dumpall/7.4': Not a directory
> > dpkg: warning - old pre-removal script returned error exit status 1
> > dpkg - trying script from the new package instead ...
> > dpkg: error
> > processing /var/cache/apt/archives/postgresql-client_7.5.22_all.deb
> > (--unpack):
> >  there is no script in the new version of the package - giving up
>
> Ok, so what is /var/lib/postgres/dumpall/7.4 on your system, and do you
> know how it got that way?  It is a directory on my systems, which is what
> it's supposed to be.

/var/lib/postgres was a file. I still have a copy of it. Can send it later.

> > Preparing to replace powernowd 0.90-3
> > (using .../powernowd_0.97-1_i386.deb) ...
> > Stopping powernowd: powernowd.
> > install: `/var/lib/postgres/dumpall/7.4': Not a directory
> >
> > Why is powernowd worried about postgres/dumpall/7.4 ???
>
> That would be a dpkg bug.  I've seen this once before, in a certain corner
> cases dpkg fails to clean up the maintainer scripts from the previous
> package it was working with when that other package failed, and it ends up
> invoking the broken script by mistake. :/
>
> I don't know if this has been filed as a bug against dpkg.

I will check and open one, if I don't find one. Unfortunately I cannot repro 
it any more.

> >  Hmm. The package shipped with a symbolic
> > link /lib/modules/2.6.18-4-686/source
> >  However, I can not read the target: No such file or directory
> >  Therefore, I am deleting /lib/modules/2.6.18-4-686/source
> >
> > Running depmod.
> > Finding valid ramdisk creators.
> > Using mkinitrd.yaird to build the ramdisk.
> > yaird error: command not found: /usr/bin/ldd (fatal)
> > mkinitrd.yaird failed to create initrd image.
> > Failed to create initrd image.
> > dpkg: error processing linux-image-2.6.18-4-686 (--configure):
> >  subprocess post-installation script returned error exit status 9
>
> Um... /usr/bin/ldd is part of the 'Essential' package set (provided by
> libc6).  What the heck happened here?!
>
> > Certainly there is an /usr/bin/ldd
> > silverboxy:/etc/cups# ls -l /usr/bin/ldd
> > -rwxr-xr-x 1 root root 5972 Feb 21 16:13 /usr/bin/ldd
> > silverboxy:/etc/cups#
>
> But presumably you're checking this now, after the fact, so that's not much
> of an indicator of what was wrong at the time this error happened, I'm
> afraid.

I checked that right after the fail. see the typescript log.

> > -> apt-proxy gets removed during aptitude upgrade. When apt-proxy is
> >    used for the upgrade, the upgrade is broken. So no upgrades with
> >    apt-proxy on localhost are possible with that method.
>
> Heh...
>
> Well, thanks for the report.  I'm afraid I don't know that we're going to
> do much with it for etch though, your chosen upgrade path doesn't really
> give us much feedback on the method we're currently recommending because
> it's not clear that your aptitude attempt was consistent with the first
> steps on <http://wiki.debian.org/Sarge2EtchUpgradeBlackboard>.

I was not aware of the wiki page. I think it should be formally announced that 
other developers/users are testing it at well. 

I wanted to test the proposed upgrade path, which I found on the mailing list. 
On the other hand, I wanted to have the system upgraded, and after aptitude 
dist-upgrade not working at all, in fact I was of that upgrade path early.

Nevertheless, I think the postgres issue would have shown up as well. Not sure 
about ldd...

Thanks,
Rainer

-- 
Rainer Dorsch
Lärchenstr. 6
D-72135 Dettenhausen
07032-359190
jabber: rdorsch@jabber.org
GPG Fingerprint: 5966 C54C 2B3C 42CC 1F4F  8F59 E3A8 C538 7519 141E
Full GPG key: http://pgp.mit.edu/



Reply to: