Hi Scott (2012.08.02_00:49:18_+0200)
> > I'd be interested to see the upgrade logs.
> Here's the apt-term.log. Let me know if you need more.
The python-support trigger is firing before the new python-support is
unpacked, even though it's going to be unpacked in the same dpkg run.
$ egrep '(python-support|configobj)' apt-term.log
Processing triggers for python-support ...
Preparing to replace python-configobj 4.7.1-1 (using .../python-configobj_4.7.2+ds-3build1_all.deb) ...
Unpacking replacement python-configobj ...
Processing triggers for python-support ...
WARNING: WARNING: /usr/share/pyshared/configobj.py is linked but does not belong to any package.
Preparing to replace python-support 1.0.4ubuntu1 (using .../python-support_1.0.14ubuntu2_all.deb) ...
Unpacking replacement python-support ...
Setting up python-configobj (4.7.2+ds-3build1) ...
Setting up python-support (1.0.14ubuntu2) ...
Processing triggers for python-support ...
Hi Scott (2012.08. 02_00:49: 18_+0200)
> > I'd be interested to see the upgrade logs.
> Here's the apt-term.log. Let me know if you need more.
The python-support trigger is firing before the new python-support is
unpacked, even though it's going to be unpacked in the same dpkg run.
$ egrep '(python- support| configobj) ' apt-term.log configobj_ 4.7.2+ds- 3build1_ all.deb) ... pyshared/ configobj. py is linked but does not belong to any package. support_ 1.0.14ubuntu2_ all.deb) ...
Processing triggers for python-support ...
Preparing to replace python-configobj 4.7.1-1 (using .../python-
Unpacking replacement python-configobj ...
Processing triggers for python-support ...
WARNING: WARNING: /usr/share/
Preparing to replace python-support 1.0.4ubuntu1 (using .../python-
Unpacking replacement python-support ...
Setting up python-configobj (4.7.2+ds-3build1) ...
Setting up python-support (1.0.14ubuntu2) ...
Processing triggers for python-support ...
This is vaguely similar to these dpkg bugs bugs.debian. org/671711 bugs.debian. org/484669
http://
http://
I don't know how best to handle this. An update-manager quirk?