Re: [Bug 1700468] Re: "krb5-auth-dialog --auto" segfaults on startup

2017-07-07 Thread Guido Günther
On Thu, Jul 06, 2017 at 08:33:06PM -, Daniel Richard G. wrote:
> Er...
> 
> $ klist -v
> klist: invalid option -- 'v'
> Usage: klist [-e] [-V] [[-c] [-l] [-A] [-d] [-f] [-s] [-a [-n]]] [-k [-t] 
> [-K]] [name]
> [...]
> 
> Remember, the segfault occurs with a user that is local-only. Kerberos
> infrastructure is installed on the system, but the user has no Kerberos
> tickets, no Kerberos envvars set, no cache file, nothing at all.
> 
> $ klist
> klist: Credentials cache file '/tmp/krb5cc_1000' not found

I can reproduce this now and this

https://git.gnome.org/browse/krb5-auth-
dialog/commit/?id=23ba826890d60c3c556a8d1a021e97b8d3fb416c

fixes it for me.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1700468

Title:
  "krb5-auth-dialog --auto" segfaults on startup

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 1700468] Re: "krb5-auth-dialog --auto" segfaults on startup

2017-07-06 Thread Guido Günther
Hi Daniel,
On Wed, Jul 05, 2017 at 10:38:35PM -, Daniel Richard G. wrote:
> Hi Guido, I think you mean "klist -V" (uppercase) :-)
> 
> On the system in question, that returns
> 
> $ klist -V
> Kerberos 5 version 1.13.2

Thath's useful too but I meant "klist -v" to get your ticket details.

> 
> -- 
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1700468
> 
> Title:
>   "krb5-auth-dialog --auto" segfaults on startup
> 
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions
>

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1700468

Title:
  "krb5-auth-dialog --auto" segfaults on startup

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 1700468] Re: "krb5-auth-dialog --auto" segfaults on startup

2017-07-03 Thread Guido Günther
Hi Daniel,
On Fri, Jun 30, 2017 at 04:39:15AM -, Daniel Richard G. wrote:
> Attached is a Valgrind log file produced from a debug build of k-a-d
> version 3.20.0.
> 
> All the errors appear to be accesses within freed memory...

Yeah that look suspicious. Can you either post (or send via private
mail) the output of "klist -v"?

> 
> ** Attachment added: "kad-valgrind-log.txt"
>
> https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+attachment/4906188/+files/kad-valgrind-log.txt
> 
> -- 
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1700468
> 
> Title:
>   "krb5-auth-dialog --auto" segfaults on startup
> 
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions
>

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1700468

Title:
  "krb5-auth-dialog --auto" segfaults on startup

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 1700468] Re: "krb5-auth-dialog --auto" segfaults on startup

2017-06-29 Thread Guido Günther
On Tue, Jun 27, 2017 at 07:49:07PM -, Daniel Richard G. wrote:
> Hunh. How odd... I can't imagine that there would be something
> particular to this system that is causing the crash. As you requested:
> 
> skunk@darkstar:/tmp/krb5-auth-dialog-3.20.0/_build/src$ G_MESSAGES_DEBUG=all 
> ./krb5-auth-dialog -a
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> principal: 
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> pk-userid: 
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> pk-anchors: 
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> pw-prompt-mins: 30
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> tgt-forwardable: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> tgt-proxiable: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> tgt-renewable: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> conf-tickets: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> principal: 
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> pk-userid: 
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> pk-anchors: 
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> pw-prompt-mins: 30
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> tgt-forwardable: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> tgt-proxiable: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> tgt-renewable: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_set_property: 
> conf-tickets: False
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_local_command_line: 
> Parsing local command line
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_startup: Primary 
> application
> (krb5-auth-dialog:16500): Gtk-DEBUG: Connecting to session manager
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_nm_client_state_changed_cb: 
> Network connected
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: monitor_ccache: Monitoring 
> /tmp/krb5cc_1000
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: ka_applet_command_line: 
> Evaluating command line
> (krb5-auth-dialog:16500): KrbAuthDialog-DEBUG: credentials_expiring: Checking 
> expiry <1800s

That looks sane.

> Segmentation fault (core dumped)

That still doesn't. Does running under valgrind's memcheck yield
anything?

valgrind /usr/bin/krb5-auth-dialog

It might make sense to either build with debugging symbpols or install
the distro's krb5-auth-dialog-dbsym package.
Cheers,
 -- Guido

> 
> -- 
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1700468
> 
> Title:
>   "krb5-auth-dialog --auto" segfaults on startup
> 
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions
>

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1700468

Title:
  "krb5-auth-dialog --auto" segfaults on startup

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 1700468] Re: "krb5-auth-dialog --auto" segfaults on startup

2017-06-27 Thread Guido Günther
I can't reproduce here neither when linked against MIT (as in Debian)
nor against heimdal (as in Xenial). Can you do a

G_MESSAGES_DEBUG=all ./src/krb5-auth-dialog -a

maybe this shows s.th. interesting before the crash.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1700468

Title:
  "krb5-auth-dialog --auto" segfaults on startup

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 1700468] Re: "krb5-auth-dialog --auto" segfaults on startup

2017-06-26 Thread Guido Günther
But it seems ka_krb5_context_init got called (you can check
kcontext_valid == TRUE) so in this case it got corrupted. Any chance you
can check 3.20?

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1700468

Title:
  "krb5-auth-dialog --auto" segfaults on startup

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1700468/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 1633207] Re: VM fails to start with dac security driver added

2016-10-28 Thread Guido Günther
dfbc9a83 was necessary since libvirt changed the paths of the monitor
socket in a89f05ba8df095875f5ec8a9065a585af63a010b. We had to switch
from VIR_DOMAIN_DEF_PARSE_INACTIVE to active since we need the domain id
(ctl->def->id) as it is part of the socket path now. It would probably
o.k. to skip validation but we need to parse the active domain config to
get the id.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1633207

Title:
  VM fails to start with dac security driver added

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1633207/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 1311487] Re: krb5-auth-dialog does not start in 14.04

2014-07-07 Thread Guido Günther
This is also fixed in 3.12.0 upstream as well as in 3.12.0-1 in Debian.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1311487

Title:
  krb5-auth-dialog does not start in 14.04

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/krb5-auth-dialog/+bug/1311487/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 668849] Re: Bug#668849: Re: gbp does not support 2nd orig tarball

2013-06-04 Thread Guido Günther
On Tue, Jun 04, 2013 at 12:53:14PM -, Éric Piel wrote:
 Can anyone suggest a manual workaround to use multiple orig tarballs?
 I've tried to follow the workflow descibed in the upstream bug [1]. It
 seems to do the trick for keeping the sources and tarballs, but building
 fails (dpkg-source  aborting due to unexpected upstream changes).
 
 So far is what I do:
 # import the main orig tarball
 git-import-dsc PKG.dsc --pristine-tar
 
 # import the next orig tarball in a separate branch into a subdirectory named 
 the same
 cd PKG
 
 git checkout --orphan upstream
 git rm -rf .
 git commit --allow-empty -m 'Initial upstream branch for PKG-ADD.'
 git checkout -f master
 
 tar xf PKG_VER-PKG-ADD.orig.tar.gz ../tmp
 git-import-orig --upstream-version=VER --upstream-branch=upstream-PKG-ADD 
 --upstream-tag=upstream-PKG-ADD/VER ../tmp
 
 # add the tarball
 pristine-tar checkout PKG_VER-PKG-ADD.orig.tar.gz
 
 However it fails to build:
 git-buildpackage --git-ignore-new -S
 :
 dpkg-source: info: local changes detected, the modified files are:
  comedilib/comedi-calibrate/configure
  comedilib/comedi-calibrate/man/comedi_calibrate.8
  comedilib/man/comedi_config.8
 dpkg-source: error: aborting due to unexpected upstream changes, see 
 /tmp/comedilib_0.10.0-3ubuntu1.diff.AtK0c1
 dpkg-source: info: you can integrate the local changes with dpkg-source 
 --commit
 dpkg-buildpackage: error: dpkg-source -i -I -b comedilib gave error exit 
 status 2

You should check what's actually in the diff. git-buildpackage shouldn't
try any tarball generation if it already finds one.
Cheers,
 -- Guido

 
 Is there some specific trick to do for building too? Could it be due to
 some weird trick in the package I'm trying to build? (It's
 comedilib-0.10.0 from Debian unstable)
 
 [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=561071#36
 
 -- 
 You received this bug notification because you are subscribed to the bug
 report.
 https://bugs.launchpad.net/bugs/668849
 
 Title:
   gbp does not support 2nd orig tarball
 
 To manage notifications about this bug go to:
 https://bugs.launchpad.net/ubuntu/+source/git-buildpackage/+bug/668849/+subscriptions

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/668849

Title:
  gbp does not support 2nd orig tarball

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/git-buildpackage/+bug/668849/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 1112831] Re: git-buildpackage incorrectly handles --notify command line parameter

2013-02-02 Thread Guido Günther
That's actually a bug in the manpage. It's missing the --git- prefix:

https://honk.sigxcpu.org/gitweb/?p=git-
buildpackage.git;a=commit;h=80b5c649a62a9fc1297e2345654f94ef762572b3

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1112831

Title:
  git-buildpackage incorrectly handles --notify command line parameter

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/git-buildpackage/+bug/1112831/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 668849] Re: gbp does not support 2nd orig tarball

2011-02-02 Thread Guido Günther
On Wed, Feb 02, 2011 at 07:45:01AM -, Rolf Leggewie wrote:
 isdnutils in Debian experimental has this 2nd tarball now if you need a
 place to have a look.  Git repo is at http://git.debian.org/?p=collab-
 maint/isdnutils.git
I'll try to have a look during fosdem. #561071 has some more discussion.
It'd be great if you could read through this and tell me if it makes
sense to you.
Cheers,
 -- Guido

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/668849

Title:
  gbp does not support 2nd orig tarball

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 668849] [NEW] gbp does not recreate 2nd orig tarball

2010-10-31 Thread Guido Günther
On Sat, Oct 30, 2010 at 07:00:33PM -, Rolf Leggewie wrote:
 Public bug reported:
 
 Binary package hint: git-buildpackage
 
 using a self-compiled, backported 0.5.4 package here (the latest I could
 backport to lucid)
 
 DebSrc3.0 allows for a second orig tarball, but gbp does not appear to
 support that yet.  Try building
 http://oss.leggewie.org/deb/isdnutils_3.9.20060704+dfsg.2-7.dsc which
 has an *orig-capi20.tar.bz2 tarball.  gitb-buildpackage --git-ignore-
 new won't recreate that tarball.

Gbp deosn't currently handle multiple tarballs at all. I do have some
patches for this though.
 -- Guido

-- 
gbp does not recreate 2nd orig tarball
https://bugs.launchpad.net/bugs/668849
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 637286] Re: git-import-dsc merges to master despite --debian-branch

2010-10-31 Thread Guido Günther
On Sat, Oct 30, 2010 at 09:19:23PM -, Rolf Leggewie wrote:
 Is this the same problem as
 
 cd /tmp  mkdir capifax  wget wget
 http://capifax.v3v.de/dl/capifax-0.7.3.tar.bz2  cd capifax  git init
  git-import-orig --pristine-tar --no-merge ../capifax-0.7.3.tar.bz2
 which still creates the master branch and merges to it despite the call
 not to do so?
Nope. This is unrelated. In fact the issue from above is not a bug since
the _inital_ import always goes to master (the upstream branch is then
created from there). Everything else would pretty much break what's
expected from a git repo.
Cheers,
 -- Guido

-- 
git-import-dsc merges to master despite --debian-branch
https://bugs.launchpad.net/bugs/637286
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 615212] Re: heuristic to determine compression-format from HEAD of upstream branch can lead to incorrect result when upstream is not merged to master

2010-10-07 Thread Guido Günther
On Tue, Sep 14, 2010 at 10:41:25AM +0200, Guido Günther wrote:
 Attached (very lightly tested patch) fixes the issue for me.
The fix for this is shipped in 0.5.9.
Cheers,
 -- Guido

-- 
heuristic to determine compression-format from HEAD of upstream branch can lead 
to incorrect result when upstream is not merged to master
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 615212] Re: heuristic to determine compression-format from HEAD of upstream branch can lead to incorrect result when upstream is not merged to master

2010-10-07 Thread Guido Günther
On Tue, Sep 14, 2010 at 10:41:25AM +0200, Guido Günther wrote:
 Attached (very lightly tested patch) fixes the issue for me.

The fix for this is shipped in 0.5.9.
Cheers,
 -- Guido

-- 
heuristic to determine compression-format from HEAD of upstream branch can lead 
to incorrect result when upstream is not merged to master
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 637286] Re: git-import-dsc merges to master despite --debian-branch

2010-10-07 Thread Guido Günther
This is fixed in 0.5.9

-- 
git-import-dsc merges to master despite --debian-branch
https://bugs.launchpad.net/bugs/637286
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 638371] Re: git-import-dsc failed to import a dsc

2010-09-16 Thread Guido Günther
I've pushed a fix for a missing debian/rules to gbp's git.

-- 
git-import-dsc failed to import a dsc
https://bugs.launchpad.net/bugs/638371
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 637286] Re: git-import-dsc merges to master despite --debian-branch

2010-09-14 Thread Guido Günther
On Tue, Sep 14, 2010 at 02:52:09AM -, Rolf Leggewie wrote:
 Are you planning to make another release to experimental anytime soon?
 I guess rather than backporting the fix to maverick I'd prefer to sync.
 Once Maverick has the fix, I'll seek an SRU exception for lucid.
No, not that soon. I have some other stuff pending which I need to
finish first and Squeeze eats up the little free time I have a at the
momene.

BTW. I've comitted a better fix for this issue which avoids the
unnecessary branch switch.
 -- Guido

-- 
git-import-dsc merges to master despite --debian-branch
https://bugs.launchpad.net/bugs/637286
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 637739] [NEW] improved heuristic for --upstream-branch

2010-09-14 Thread Guido Günther
On Tue, Sep 14, 2010 at 03:17:08AM -, Launchpad Bug Tracker wrote:
 You have been subscribed to a public bug by Rolf Leggewie (r0lf):
 
 Binary package hint: git-buildpackage
 
 the current default for gbp and friends is to use HEAD of the upstream
 branch for --upstream-branch.  Thinking about bug 615212, I felt that
 gbp could do much better than this.  Consider the very simple commit
 graph in http://launchpadlibrarian.net/55567353/graph.txt for example.
 Let's say that commit X were to correspond to the lenny package, for
 example, and you want to make a security release or backport an
 important bugfix to stable.  You branch off X and commit X'.  When you
 now build against X' the default is to use the HEAD of upstream which is
 wrong and gbp has the necessary information to determine it is wrong.
 
 I think gbp should the find the highest common ancestor on the
 upstream branch and use that for --upstream-branch.  I'm sure there is a
 ready-made, easy tool in git to do this.

Something like this is on the todo list since quiet some time, however
we can even do better and look at the specific tag of that upstream
version which so guessing is needed at all. This was delayed by #534494. 

The corresponding Debian Bug is #564791.
 -- Guido

-- 
improved heuristic for --upstream-branch
https://bugs.launchpad.net/bugs/637739
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 615212] Re: heuristic to determine compression-format from HEAD of upstream branch can lead to incorrect result when upstream is not merged to master

2010-09-14 Thread Guido Günther
I just noticed the issue after sending the last mail. The heuristic is
to look at pristine-tar not upstream (basically since pristine-tar only
supports a single branch) so we have to look at the commit messages.

-- 
heuristic to determine compression-format from HEAD of upstream branch can lead 
to incorrect result when upstream is not merged to master
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 615212] Re: heuristic to determine compression-format from HEAD of upstream branch can lead to incorrect result when upstream is not merged to master

2010-09-14 Thread Guido Günther
Attached (very lightly tested patch) fixes the issue for me.
 -- Guido


** Patch added: 0001-Improve-compression-format-detection.patch
   
https://bugs.launchpad.net/bugs/615212/+attachment/1582015/+files/0001-Improve-compression-format-detection.patch

-- 
heuristic to determine compression-format from HEAD of upstream branch can lead 
to incorrect result when upstream is not merged to master
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 636862] [NEW] git-import-dscs creates merges instead of simple commits

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 06:43:25AM -, Rolf Leggewie wrote:
 Public bug reported:
 
 Binary package hint: git-buildpackage
 
 I've rebuilt 0.5.4 with 88afa6117ea4dd6b5d8fd88936afffb11c25ec47
 reverted (Pass --multimaint-merge on to dch).  The devscripts package
 in lucid is not recent enough.  As of today, I'm (still) having the
 following issue with git-import-dsc(s).
 
 TEST CASE:
 $ cd /tmp
 $ for i in 1 2 3 4 ; do dget -ud 
 http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-$i.dsc;done
 $ git-import-dscs isdnutils_3.9.20060704+dfsg.1*dsc
 $ cd isdnutils
 $ git log -p
 
 If you look at the history with git log -p or qgit you will realize
 that git-import-dsc uses merges from the upstream branch and HEAD every
 (!) time instead of just a simple commit when possible.  This results in
 an overly complex history and breaks git log -p.
 
 ** Affects: git-buildpackage (Ubuntu)
  Importance: Undecided
  Status: New
 
 ** Description changed:
 
   Binary package hint: git-buildpackage
   
   I've rebuilt 0.5.4 with 88afa6117ea4dd6b5d8fd88936afffb11c25ec47
   reverted (Pass --multimaint-merge on to dch).  The devscripts package
   in lucid is not recent enough.  As of today, I'm (still) having the
   following issue with git-import-dsc(s).
   
   TEST CASE:
   $ cd /tmp
   $ for i in 1 2 3 4 ; do dget -ud 
 http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-$i.dsc;done
 - $ git-import-dscs 
   $ git-import-dscs isdnutils_3.9.20060704+dfsg.1*dsc
   $ cd isdnutils
   $ git log -p
   
   If you look at the history with git log -p or qgit you will realize
   that git-import-dsc uses merges from the upstream branch and HEAD every
   (!) time instead of just a simple commit when possible.  This results in
   an overly complex history and breaks git log -p.
This has in fact be fixed in 0.5.4. When not importing a new upstream
version only one parent is set.
 -- Guido

-- 
git-import-dscs creates merges instead of simple commits
https://bugs.launchpad.net/bugs/636862
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 615212] [NEW] gbp misses bz2 support

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 06:36:51AM -, Launchpad Bug Tracker wrote:
 You have been subscribed to a public bug by Rolf Leggewie (r0lf):
 
 Binary package hint: git-buildpackage
 
 gbp will create a new orig tar even when an orig.tar.bz2 is present
See the manpage:
--git-compression-level=auto
Cheers,
 -- Guido

 TEST CASE:
 $ cd /tmp
 $ dget -ud 
 http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-2.dsc
 $ git-import-dsc ./isdnutils_3.9.20060704+dfsg.1-2.dsc
 $ cd isdnutils
 $ nice git-buildpackage --git-ignore-new
 [...]
 isdnutils_3.9.20060704+dfsg.1.orig.tar.gz does not exist, creating from 
 'upstream/3.9.20060704+dfsg.1'
 [...]
 dpkg-source: error: several orig.tar files found 
 (./isdnutils_3.9.20060704+dfsg.1.orig.tar.bz2 and 
 ./isdnutils_3.9.20060704+dfsg.1.orig.tar.gz) but only one is allowed
 
 ** Affects: git-buildpackage (Ubuntu)
  Importance: Undecided
  Status: New
 
 -- 
 gbp misses bz2 support
 https://bugs.edge.launchpad.net/bugs/615212
 You received this bug notification because you are a direct subscriber of the 
 bug.


-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 636862] Re: git-import-dscs creates merges instead of simple commits

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 10:02:52AM -, Rolf Leggewie wrote:
 unfortunately, f63c4ed85ca16a6869cf35bc9070e3c28163594e is not a
 straight-forward cherry-pick against 0.4.65ubuntu1 and I did not feel
 confident enough to resolve the conflicts.
 
 Personally, I'll just use the backported 0.5.4 version but it would be
 nice to have a fixed version in lucid, I guess.
Thats mostly up to Ubuntu - not much I can do here.
 -- Guido

-- 
git-import-dscs creates merges instead of simple commits
https://bugs.launchpad.net/bugs/636862
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 615212] Re: gbp misses bz2 support

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 09:53:43AM -, Rolf Leggewie wrote:
 I guess a test-case would be to import with --pristine-tar the latest
 releases of isdnutils from Debian stable unstable and Ubuntu maverick.
 
 Debian stable: --upstream-branch=upstream debian-branch=master
 Debian unstable: --upstream-branch=dsfg (branched off from upstream) 
 --debian-branch=master
 Maverick: --upstream-branch=upstream debian-branch=ubuntu
 
 Then try to build the master branch against dfsg branch and you should
 run into this issue.
Could you build a simple testcase please? Tarball detection works fine
here and I assume your last commit on the pristine-tar branch simply
doesn't give the correct compression type.
 -- Guido

-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 615212] Re: gbp misses bz2 support

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 01:35:12PM -, Rolf Leggewie wrote:
 ** Description changed:
 
 ** Description changed:
 
   Binary package hint: git-buildpackage
   
 - gbp will create a new orig tar even when an orig.tar.bz2 is present
 - 
 - TEST CASE:
 - $ cd /tmp
 - $ dget -ud 
 http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-2.dsc
 - $ git-import-dsc ./isdnutils_3.9.20060704+dfsg.1-2.dsc
 - $ cd isdnutils
 - $ nice git-buildpackage --git-ignore-new
 - [...]
 - isdnutils_3.9.20060704+dfsg.1.orig.tar.gz does not exist, creating from 
 'upstream/3.9.20060704+dfsg.1'
 - [...]
 - dpkg-source: error: several orig.tar files found 
 (./isdnutils_3.9.20060704+dfsg.1.orig.tar.bz2 and 
 ./isdnutils_3.9.20060704+dfsg.1.orig.tar.gz) but only one is allowed
 + gbp will create a new orig tar even when an orig.tar.bz2 is present.
 + Please see attached file for test case.
This works here:

git-import-dsc --pristine-tar --download 
http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-2.dsc
rm *.bz2
cd isdnutils
git-buildpackage -d
...
dpkg-source: info: using source format `3.0 (quilt)'
dpkg-source: info: building isdnutils using existing 
./isdnutils_3.9.20060704+dfsg.1.orig.tar.bz2
...

The auto option was introduced in 0.5.0.
 -- Guido

-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 615212] Re: gbp misses bz2 support

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 04:14:08PM -, Rolf Leggewie wrote:
 Günther, thank you for trying to reproduce this issue.  But your test is
 way too simplistic ;-)  Try the script I have attached now (took a
 little longer due to bug 637286 throwing me off course for a while).
There no script at this mail. Please use the Debian BTS since.
 -- Guido

-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Re: [Bug 637286] [NEW] git-import-dsc merges to master despite --debian-branch

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 03:41:13PM -, Launchpad Bug Tracker wrote:
 You have been subscribed to a public bug by Rolf Leggewie (r0lf):
 
 Binary package hint: git-buildpackage
 
 git-import-dsc will merge to master even when --debian-branch is
 explicitly set.  Run the attached script as a test-case.

Attached patch should fix this.
 -- Guido


** Patch added: set-branch.diff
   
https://bugs.launchpad.net/bugs/637286/+attachment/1579039/+files/set-branch.diff

-- 
git-import-dsc merges to master despite --debian-branch
https://bugs.launchpad.net/bugs/637286
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


Re: [Bug 637286] [NEW] git-import-dsc merges to master despite --debian-branch

2010-09-13 Thread Guido Günther
On Mon, Sep 13, 2010 at 03:41:13PM -, Launchpad Bug Tracker wrote:
 You have been subscribed to a public bug by Rolf Leggewie (r0lf):
 
 Binary package hint: git-buildpackage
 
 git-import-dsc will merge to master even when --debian-branch is
 explicitly set.  Run the attached script as a test-case.
I've pushed the fix to git. Thanks for the testcase.
 -- Guido

-- 
git-import-dsc merges to master despite --debian-branch
https://bugs.launchpad.net/bugs/637286
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 615212] Re: gbp misses bz2 support

2010-09-13 Thread Guido Günther
Testcase fails with:

$ sh -e lp615212-test-case.sh 
`isdnutils_3.12.20071127-0ubuntu6.dsc' - 
`isdnutils.old/isdnutils_3.12.20071127-0ubuntu6.dsc'
`isdnutils_3.9.20060704+dfsg.1-2.dsc' - 
`isdnutils.old/isdnutils_3.9.20060704+dfsg.1-2.dsc'
`isdnutils_3.9.20060704+dfsg.1-3.dsc' - 
`isdnutils.old/isdnutils_3.9.20060704+dfsg.1-3.dsc'
`isdnutils_3.9.20060704-3.6.dsc' - 
`isdnutils.old/isdnutils_3.9.20060704-3.6.dsc'
dget: retrieving 
http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-1.dsc
  % Total% Received % Xferd  Average Speed   TimeTime Time  Current
 Dload  Upload   Total   SpentLeft  Speed
  0 00 00 0  0  0 --:--:-- --:--:-- --:--:-- 0
curl: (22) The requested URL returned error: 404
dget: curl isdnutils_3.9.20060704+dfsg.1-1.dsc 
http://ftp.de.debian.org/debian/pool/main/i/isdnutils/isdnutils_3.9.20060704+dfsg.1-1.dsc
 failed

-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 615212] Re: gbp misses bz2 support

2010-09-13 Thread Guido Günther
Ignoring the error the outcome is expected:

commit 40a16b8c1523282b8b53c034b5fa86b25b93473e
Author: Guido Günther a...@sigxcpu.org
Date:   Mon Sep 13 21:05:24 2010 +0200

pristine-tar data for isdnutils_3.9.20060704.orig.tar.gz

:00 100644 000... 8585744... A  isdnutils_3.9.20060704.orig.tar.gz.delta
:00 100644 000... 730e8af... A  isdnutils_3.9.20060704.orig.tar.gz.id

so the last version is a tar.gz and so we assume this is this is the
preferred format. So I'd say this is no bug.

-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 615212] Re: gbp misses bz2 support

2010-08-13 Thread Guido Günther
bz2 support support is there since quiet some time now. It detects the
compression type from the pristine-tar branch, it can also be given in
the config file or on the command line.

I'd be better if you could file the bugs in the Debian BTS since I
usually don't check launchpad.

-- 
gbp misses bz2 support
https://bugs.launchpad.net/bugs/615212
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 615211] Re: better support for dfsg packages

2010-08-13 Thread Guido Günther
That's what the upstream-branch option is for. You can put this in your
per-repo gbp.conf.

-- 
better support for dfsg packages
https://bugs.launchpad.net/bugs/615211
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 403299] Re: sequential import fails

2010-08-13 Thread Guido Günther
git-import-dscs (note the extra s) will handle this for you.

-- 
sequential import fails
https://bugs.launchpad.net/bugs/403299
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 508949] Re: can't determine version number

2010-05-17 Thread Guido Günther
This was fixed some time ago in Debian:

http://packages.debian.org/changelogs/pool/contrib/s/sapgui-package
/sapgui-package_0.0.9/changelog

-- 
can't determine version number
https://bugs.launchpad.net/bugs/508949
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 403299] Re: sequential import fails

2010-01-26 Thread Guido Günther
Works in Debian's 0.4.64.

-- 
sequential import fails
https://bugs.launchpad.net/bugs/403299
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 403299] Re: sequential import fails

2010-01-26 Thread Guido Günther
You need to change into the repo after the first import though (or use
git-import-dsc*s*)

-- 
sequential import fails
https://bugs.launchpad.net/bugs/403299
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 264974] Re: Window Picker Applet in Ubuntu Netbook Remix doesn't work with vertical panel (on side)

2009-07-17 Thread Guido Günther
We also have this issue reported in Debian:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=536868

** Bug watch added: Debian Bug tracker #536868
   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=536868

** Also affects: window-picker-applet (Debian) via
   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=536868
   Importance: Unknown
   Status: Unknown

-- 
Window Picker Applet in Ubuntu Netbook Remix doesn't work with vertical panel 
(on side)
https://bugs.launchpad.net/bugs/264974
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 47994] Re: PowerBook G4 12 - Suspend to disk doesn't work

2008-09-12 Thread Guido Günther
I'm using two tiny scripts for hibernate in Debian, pbbuttonsd isn't
needed, pm-utils can handle it:

http://honk.sigxcpu.org/con/Suspend_to_Disk_with_a_Apple_Powerbook_G4_12_.html

Maybe this helps. Gnome-power-manager forbidding hibernate altogether
seems to be a different issue though, maybe there's a Ubuntu specific
file  forbidding it in /usr/share/hal/fdi?

The system should indicate that it can hibernate via:
 power_management.can_hibernate = true
in the lshal output.

-- 
PowerBook G4 12 - Suspend to disk doesn't work
https://bugs.launchpad.net/bugs/47994
You received this bug notification because you are a member of Ubuntu
Bugs, which is a direct subscriber.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 73550] Re: Should dissappear when credentials are obtained through other mechanism

2008-06-09 Thread Guido Günther
This is fixed in Debian's krb5-auth-dialog among other things (currently
in incoming). I've based the package on my current krb5-auth-dialog
development branch at:

git-clone http://honk.sigxcpu.org/git/krb5-auth-dialog.git/
See also:
http://honk.sigxcpu.org/con/

This is from the Debian changelog:
* use GtkSecureEntry to fetch password
* adds trayicon
* fix localization
* close dialog when credentials were acquired by other means
* use libnotify and gconf
* session handling via dbus
* use g_timeout_add_seconds to save battery power
* set a proper icon

Cheers,
 -- Guido

-- 
Should dissappear when credentials are obtained through other mechanism
https://bugs.launchpad.net/bugs/73550
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs