Le jeudi 25 avril 2013 à 19:45 +0200, Julien Cristau a écrit :
I don’t think it is a problem for scripts in the postinst, nor in
triggered scripts. I thought prerm was supposed to be OK as long as
there are no loops in the upgrade (the policy is vague on the subject),
but this is not the
On Sun, 28 Apr 2013, Josselin Mouette wrote:
Le jeudi 25 avril 2013 à 19:45 +0200, Julien Cristau a écrit :
I don’t think it is a problem for scripts in the postinst, nor in
triggered scripts. I thought prerm was supposed to be OK as long as
there are no loops in the upgrade (the policy
Hi
Am 28.04.2013 13:54, schrieb Andreas Beckmann:
On 2013-04-28 11:58, Andreas Beckmann wrote:
attached are new versions of the patches that add empty dummy postrm
scripts, too.
And I can confirm that this really fixes the upgrade issue.
Thanks Andreas, for the patches. I've uploaded
On Sat, 2013-04-27 at 01:35 +0200, Andreas Beckmann wrote:
If skipping the old gconf-schemas --unregister call is not problematic
because an equivalent action will be done by trigger processing (and no
stale schemas will stay registered), adding the empty dummy prerm should
be a safe solution.
Followup-For: Bug #706110
Control: tag -1 patch
On 2013-04-27 10:35, Adam D. Barratt wrote:
If we could get it done over the weekend, that should be okay. I'd be
interested in seeing a binary debdiff of the rebuilt packages against
those in wheezy though.
Here are patch + binary debdiffs.
[re-added -release CC that went missing at some point]
On Sat, 2013-04-27 at 21:29 +0200, Josselin Mouette wrote:
Le samedi 27 avril 2013 à 16:34 +0100, Adam D. Barratt a écrit :
Some versions are bumped, but everything still satisfiable in wheezy.
There's also several new dependencies
Le jeudi 25 avril 2013 à 19:47 +0200, Julien Cristau a écrit :
On Thu, Apr 25, 2013 at 15:05:47 +0200, Michael Biebl wrote:
All packages shipping gconf schemas have a Depends on gconf2. It is
generated via dh_gconf which adds gconf2 (= 2.28.1-2) to misc:Depends.
This version is
Le vendredi 26 avril 2013 à 18:54 +0200, Josselin Mouette a écrit :
I think we only need to fix the packages:
* that still have a dh_gconf prerm snippet in squeeze
* AND that don’t have a prerm anymore in wheezy (or that haven’t been
uploaded since squeeze).
Here is the list. Forget those
Am 26.04.2013 19:17, schrieb Josselin Mouette:
Le vendredi 26 avril 2013 à 18:54 +0200, Josselin Mouette a écrit :
I think we only need to fix the packages:
* that still have a dh_gconf prerm snippet in squeeze
* AND that don’t have a prerm anymore in wheezy (or that haven’t been
uploaded
Am 27.04.2013 00:50, schrieb Michael Biebl:
Am 26.04.2013 19:17, schrieb Josselin Mouette:
Le vendredi 26 avril 2013 à 18:54 +0200, Josselin Mouette a écrit :
I think we only need to fix the packages:
* that still have a dh_gconf prerm snippet in squeeze
* AND that don’t have a prerm anymore
On 2013-04-27 00:56, Michael Biebl wrote:
Am 27.04.2013 00:50, schrieb Michael Biebl:
Am 26.04.2013 19:17, schrieb Josselin Mouette:
Le vendredi 26 avril 2013 à 18:54 +0200, Josselin Mouette a écrit :
I think we only need to fix the packages:
* that still have a dh_gconf prerm snippet in
Am 25.04.2013 01:11, schrieb Michael Biebl:
According to jwilk, this is a issue in the python package:
jwilk mbiebl: I'm pretty sure it's not /usr/sbin/gconf-schemas that is
missing, but /usr/bin/python (which is is -schema's shebang).
jwilk Or more precisely, /usr/bin/python is there, but
On Thu, Apr 25, 2013 at 14:09:24 +0200, Michael Biebl wrote:
Am 25.04.2013 01:11, schrieb Michael Biebl:
According to jwilk, this is a issue in the python package:
jwilk mbiebl: I'm pretty sure it's not /usr/sbin/gconf-schemas that is
missing, but /usr/bin/python (which is is -schema's
Am 25.04.2013 14:26, schrieb Julien Cristau:
Thoughts in random order...
1) this is terrible
A dangling /usr/bin/python symlink is indeed bad. At least that
convinced me to never use anything pythonish in the maintainer scripts
in the future.
2) in the absence of loops, dpkg ought to unpack
On 2013-04-25 15:05, Michael Biebl wrote:
Am 25.04.2013 14:26, schrieb Julien Cristau:
2) in the absence of loops, dpkg ought to unpack python2.7-minimal
before python-minimal
I'm not sure what's going on here, but in the log [1] from anbe
python2.7-minimal is not unpacked *at all*.
Le jeudi 25 avril 2013 à 15:05 +0200, Michael Biebl a écrit :
A dangling /usr/bin/python symlink is indeed bad. At least that
convinced me to never use anything pythonish in the maintainer scripts
in the future.
I don’t think it is a problem for scripts in the postinst, nor in
triggered
On Thu, Apr 25, 2013 at 16:26:23 +0200, Josselin Mouette wrote:
Le jeudi 25 avril 2013 à 15:05 +0200, Michael Biebl a écrit :
A dangling /usr/bin/python symlink is indeed bad. At least that
convinced me to never use anything pythonish in the maintainer scripts
in the future.
I don’t
On Thu, Apr 25, 2013 at 15:05:47 +0200, Michael Biebl wrote:
All packages shipping gconf schemas have a Depends on gconf2. It is
generated via dh_gconf which adds gconf2 (= 2.28.1-2) to misc:Depends.
This version is already satisfied by gconf2 from squeeze. We could bump
that and binNMU the
18 matches
Mail list logo
