Bug#1054386: bookworm-pu: package fssync/1.6-1.1+deb12u1

[Julien Muchembled]

Package: release.debian.org
Severity: normal
Tags: bookworm
User: release.debian@packages.debian.org
Usertags: pu
X-Debbugs-Cc: fss...@packages.debian.org, j...@jmuchemb.eu, sanv...@debian.org
Control: affects -1 + src:fssync

[ Reason ]
This release fixes Bug #858253: FTBFS randomly (AssertionError: Lists differ: 
[b'a', b'a/c'] != [b'a/c'])

[ Impact ]
If the update is not approved, anybody building the package from source will
experience that the package will FTBFS randomly (with high probability) for
no good reason.

[ Tests ]
I can reproduce the random FTBFS failures. The test suite is definitely not 
appropriate
for the current code. As there is a new upstream release in unstable and the 
version in
stable will not change, there is no point in keeping the flaky test suite in 
stable.

[ Risks ]
The program remains the same. Only the test suite is disabled.

[ Checklist ]
   [X] *all* changes are documented in the d/changelog
   [X] I reviewed all changes and I approve them
   [X] attach debdiff against the package in (old)stable
   [X] the issue is verified as fixed in unstable

[ Changes ]
The test suite is disabled, as it's extremely flaky.
diff -Nru fssync-1.6/debian/changelog fssync-1.6/debian/changelog
--- fssync-1.6/debian/changelog 2021-01-05 18:30:07.0 +0100
+++ fssync-1.6/debian/changelog 2023-10-22 12:25:00.0 +0200
@@ -1,3 +1,10 @@
+fssync (1.6-1.1+deb12u1) bookworm; urgency=medium
+
+  * Disable extremely flaky test suite, as it has been failing randomly
+for several years. Closes: #858253.
+
+ -- Julien Muchembled   Sun, 22 Oct 2023 12:25:00 +0200
+
 fssync (1.6-1.1) unstable; urgency=medium
 
   * Non maintainer upload by the Reproducible Builds team.
diff -Nru fssync-1.6/debian/rules fssync-1.6/debian/rules
--- fssync-1.6/debian/rules 2017-01-21 20:53:22.0 +0100
+++ fssync-1.6/debian/rules 2023-10-22 12:25:00.0 +0200
@@ -18,3 +18,5 @@
 
 override_dh_auto_install:
dh_auto_install -- PREFIX=/usr
+
+override_dh_auto_test:


OpenPGP_signature
Description: OpenPGP digital signature

Bug#1054329: RM: tuxonice-userui -- ROM; dead upstream; severely outdated

[Julien Muchembled]

Package: ftp.debian.org
Severity: normal
User: ftp.debian@packages.debian.org
Usertags: remove
X-Debbugs-Cc: tuxonice-use...@packages.debian.org
Control: affects -1 + src:tuxonice-userui

Last time TuxOnIce was usable on my laptop was with Linux 4.1:
at some point I had to give it up. This is a pity that it never
got merged upstream because vanilla hibernation is still so
limited (slow, memory constraints, no UI to cancel the process
or hide ugly consoles...). I guess it was too much work for the
author to follow changes in the kernel - no dev for 5 years.

And about the UI (i.e. the package I request RM), the upstream
repository has disappeared.

So I think it's time for tuxonice-userui to retire.

Julien

Bug#1037268: kwin-x11: resource leak: the number of threads increases over time, boundlessly

[Julien Muchembled]

Control: forwarded -1 https://bugs.kde.org/show_bug.cgi?id=470898

I added more information upstream.

Le 10/06/2023 à 20:22, Nicholas D Steeves a écrit :

I could find that such leak happens when viewing videos fullscreen e.g. with 
mpv. Can anyone reproduce it ? More precisely:
1. start mpv without --fs -> thread count does not change
2. switch to fullscreen -> usually -2 threads
3. leaves fullscreen -> usually +17 threads
4. exit mpv -> thread count does not change


I was able to reproduce (with mpv 0.35.1-4), but going fullscreen didn't
increase the kwin thread count on my system; however, exiting fullscreen
consistently increased kwin's thread count by exactly two threads each
time mpv leaves fullscreen.


That's actually what I wrote.


The good news is that one doesn't need to restart KDE.  Try this:

   kwin --replace&


Oh thanks!

But I finally decided to switch to Wayland, and now new issues :(


Version 4:5.27.2-2 was affected too. I previously had 4:5.24.4-1 and I'm almost 
sure it had no leak.


That's not a terrible range to bisect, if it comes to that.


Really? I first checked my setup by successfully building kwin 4:5.27.5-3 and 
then tried 4:5.26.5-1 from from snapshot.debian.org:
- without downgrading any build-depends (and without forcing -d), compilation 
error;
- the executable inside the binary package depends on libraries that aren't 
there anymore.
I gave up.


- I use mpv from deb-multimedia.


For reference, please note that bugs must be reproducible with Debian
alone, otherwise they may be marked invalid.


Sure.


- mpv conf: vo = gpu


Actually, there's also a leak with --vo=x11

Julien

Bug#1037268: kwin-x11: resource leak: the number of threads increases over time, boundlessly

[Julien Muchembled]

Package: kwin-x11
Version: 4:5.27.5-3
Severity: normal
Tags: upstream

After 17 days, the number of threads of /usr/bin/kwin_x11 process has exceeded 
2700 and it keeps increasing. At the beginning of a session, the process starts 
with 34 threads.

I could find that such leak happens when viewing videos fullscreen e.g. with 
mpv. Can anyone reproduce it ? More precisely:
1. start mpv without --fs -> thread count does not change
2. switch to fullscreen -> usually -2 threads
3. leaves fullscreen -> usually +17 threads
4. exit mpv -> thread count does not change

Such a leak is at least a severe issue for anyone like me who set a nproc limit 
to protect against fork-bombs, in particular because it causes mysterious 
process crashes. I only understood the problem when I wanted to build a 
software that spawned a lot of processes. It also makes nproc limiting quite 
unusable without restarting KDE from time to time.

Version 4:5.27.2-2 was affected too. I previously had 4:5.24.4-1 and I'm almost 
sure it had no leak.

A few notes that may be specific to my setup:
- I do have a 4k intel display. And regularly, for less than 1s, the display is 
corrupted. I haven't tried Wayland yet and I hope it will fix it.
- I use mpv from deb-multimedia.
- mpv conf: vo = gpu

I couldn't find anything on bug.kde.org: should I forward ?

Julien

-- System Information:
Debian Release: 12.0
  APT prefers testing
  APT policy: (900, 'testing'), (800, 'unstable'), (700, 'proposed-updates'), 
(700, 'stable'), (500, 'unstable-debug'), (500, 'testing-debug'), (500, 
'stable-security'), (500, 'stable-debug'), (500, 'oldstable-proposed-updates'), 
(500, 'oldstable'), (1, 'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 6.1.0-9-amd64 (SMP w/16 CPU threads; PREEMPT)
Kernel taint flags: TAINT_UNSIGNED_MODULE
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages kwin-x11 depends on:
ii  kwin-common4:5.27.5-3
ii  libc6  2.36-9
ii  libepoxy0  1.5.10-1
ii  libgcc-s1  12.2.0-14
ii  libkdecorations2-5v5   4:5.27.5-2
ii  libkf5configcore5  5.103.0-2
ii  libkf5configgui5   5.103.0-2
ii  libkf5configwidgets5   5.103.0-1
ii  libkf5coreaddons5  5.103.0-1
ii  libkf5crash5   5.103.0-1
ii  libkf5globalaccel-bin  5.103.0-1
ii  libkf5globalaccel5 5.103.0-1
ii  libkf5i18n55.103.0-1
ii  libkf5notifications5   5.103.0-1
ii  libkf5plasma5  5.103.0-1
ii  libkf5service-bin  5.103.0-1
ii  libkf5service5 5.103.0-1
ii  libkf5windowsystem55.103.0-1
ii  libkwineffects14   4:5.27.5-3
ii  libkwinglutils14   4:5.27.5-3
ii  libqaccessibilityclient-qt5-0  0.4.1-1+b1
ii  libqt5core5a   5.15.8+dfsg-11
ii  libqt5dbus55.15.8+dfsg-11
ii  libqt5gui5 5.15.8+dfsg-11
ii  libqt5qml5 5.15.8+dfsg-3
ii  libqt5quick5   5.15.8+dfsg-3
ii  libqt5widgets5 5.15.8+dfsg-11
ii  libqt5x11extras5   5.15.8-2
ii  libstdc++6 12.2.0-14
ii  libx11-6   2:1.8.4-2
ii  libxcb-composite0  1.15-1
ii  libxcb-keysyms10.4.0-1+b2
ii  libxcb-randr0  1.15-1
ii  libxcb-render0 1.15-1
ii  libxcb-shape0  1.15-1
ii  libxcb-xfixes0 1.15-1
ii  libxcb-xkb11.15-1
ii  libxcb11.15-1
ii  libxi6 2:1.8-1+b1
ii  libxkbcommon-x11-0 1.5.0-1
ii  libxkbcommon0  1.5.0-1

kwin-x11 recommends no packages.

kwin-x11 suggests no packages.

-- no debconf information

Bug#963181: python3-zstd: please package python3 zstandard library in debian

[Julien Muchembled]

I am the author of https://pypi.org/project/cython-zstd/ and I advise Debian to 
rather package it.

https://pypi.org/project/zstd/ issues:
- written in pure C: writing bindings without Cython is really looking for bugs
- it fails to compress b''

Other advantage of https://pypi.org/project/cython-zstd/:
- as an optimization, compress() resizes the output buffer rather than doing a 
copy or waste memory ('zstd' egg does the latter)

Unfortunately, for the only place where I planned to use zstd in a Python 
program, gzip remains better ( https://github.com/facebook/zstd/issues/1134 ). 
So I didn't follow recent changes in zstd API and I see that 
https://pypi.org/project/zstd/'s compress gained a 'threads' parameter (which 
https://pypi.org/project/cython-zstd/ does not have).

On https://lab.nexedi.com/nexedi/cython-zstd you'd find 2 branches:
- debian, where I started Debian packaging
- stream, for the stream API: I don't remember if it's working

Julien

Bug#938073: fssync marked for autoremoval due to python-pylibacl?

[Julien Muchembled]

With this bug fixed, fssync has just been marked for autoremoval because it 
would depend (transitively) on python-pylibacl. I don't understand why.

fssync only depends on python3-pylibacl.

Bug#951355: rspamd: Bayesian filters should not cause mails to be rejected

[Julien Muchembled]

Le 15/02/2020 à 06:14, Julien Muchembled a écrit :
> After a quick investigation, I understand that was because of bayesian
> filters.

I think I misused the term "Bayesian". I used it in opposition to filters that 
analyze the form of the message or how it is transported.

> If you think it's not possible with default configuration,
> I can provide more information and you can stop reading here.

I didn't see that the default logrotate settings are so aggressive and I don't 
have the logs anymore.

> In fact, I don't expect much from this bug report. I am already quite unhappy
> of rspamd for other reasons, mainly because it's too slow at learning ham/spam
> (I configured dovecot to trigger that when I move mails from/to the Spam
> folder).

Should I have feed rspamd with my spam & ham folders when I set it up? 
Bogofilter is clearer about the fact it has to be done.
I have "bayes_classify: skip classification as ham class has not enough learns: 
89, 200 required" logs and that would explain why I found it slow at learning 
ham/spam.

> With this incident, I am now decided to try bogofilter.

Done 2 weeks ago (along with postfix-policyd-spf-python) and now I purge rspamd 
from my server.

Julien

Bug#951355: rspamd: Bayesian filters should not cause mails to be rejected

[Julien Muchembled]

Package: rspamd
Version: 1.9.4-2~bpo10+1
Severity: important
Tags: upstream

Yesterday, I got a mail from the Listmaster Team because my server
rejected the following debian-user-french mail:

  
https://lists.debian.org/msgid-search/94de34b7-3d83-ba9d-49a4-254bcf3a4...@visionduweb.com

After a quick investigation, I understand that was because of bayesian
filters. If you think it's not possible with default configuration,
I can provide more information and you can stop reading here.

Else, I am horrified by such behaviour from rspamd. I consider that a mail
shall only be rejected if it is indubitably spam (e.g. SPF test failure).
But with Bayesian filters, it's impossible to be absolutely sure and the
only valid option is to accept it.

In fact, I don't expect much from this bug report. I am already quite unhappy
of rspamd for other reasons, mainly because it's too slow at learning ham/spam
(I configured dovecot to trigger that when I move mails from/to the Spam
folder). Configuring rspamd also looks horribly complicated and my only
settings are:

# head local.d/*
==> local.d/rbl.conf <==
enabled = false;

==> local.d/surbl.conf <==
enabled = false;

==> local.d/worker-normal.inc <==
enabled = false;

==> local.d/worker-proxy.inc <==
bind_socket = "localhost:11332";
upstream "local" {
  default = yes;
  self_scan = yes;
}

(no greylisting)

Whitelisting bendel.debian.org should be something doable easily
but I don't know how. 

With this incident, I am now decided to try bogofilter.

I tagged upstream because I doubt that Debian ships with different settings
than upstream about this issue.

-- System Information:
Debian Release: 10.0
  APT prefers proposed-updates
  APT policy: (900, 'proposed-updates'), (900, 'stable'), (500, 
'stable-debug'), (500, 'proposed-updates-debug')
Architecture: amd64 (x86_64)

Kernel: Linux 4.19.0-6-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_UNSIGNED_MODULE
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8), 
LANGUAGE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages rspamd depends on:
ii  adduser  3.118
ii  ca-certificates  20190110
ii  libc62.28-10
ii  libevent-2.1-6   2.1.8-stable-4
ii  libglib2.0-0 2.58.3-2+deb10u1
ii  libicu63 63.1-6
ii  libjs-bootstrap  3.4.1+dfsg-1
ii  libjs-jquery 3.3.1~dfsg-3
ii  libjs-requirejs  2.3.6-1
ii  libluajit-5.1-2  2.1.0~beta3+dfsg-5.1
ii  libmagic11:5.35-4
ii  libpcre2-8-0 10.32-5
ii  libsqlite3-0 3.27.2-3
ii  libssl1.11.1.1c-1
ii  libunwind8   1.2.1-9
ii  lsb-base 10.2019051400
ii  zlib1g   1:1.2.11.dfsg-1

rspamd recommends no packages.

rspamd suggests no packages.

-- no debconf information

Bug#908492: nat-rtsp-dkms: fails to build on new 4.18 kernels

[Julien Muchembled]

Someone wrote a fix at:

https://github.com/sforshee/rtsp-linux/commit/3a4a4866890e7daee96010291feb72396a99d9ed

I didn't try yet.

Julien

Bug#900781: git-buildpackage: import-orig does 'reset --hard' at then end, voiding the previously run 'pq import'

[Julien Muchembled]

Le 06/05/18 à 14:20, Guido Günther a écrit :
> Does this work as you'd expect it:
> 
> $ gbp clone g...@salsa.debian.org:python-team/modules/zc-lockfile.git
> $ gbp pq import
> $ git co debian/master
> $ gbp import-orig 
> https://files.pythonhosted.org/packages/f5/fe/efb94907d8b2b81c3beab1bd628ff67e310d82816b94aa00b52062727ea9/zc.lockfile-1.3.0.tar.gz
> 
> import-orig imports onto the branch you're currently on 
> causing the above commit to be lost. It should import on the
> debian-branch instead.

Yes, that works.

But import-orig is supposed to import onto the branch specified by 
--debian-branch (already 'debian/master' according to debian/gpb.conf), isn't 
it ?

Or if debian/master must really be checked out, then the problem is in the 
docs, and in the description of --debian-branch

Bug#900781: git-buildpackage: import-orig does 'reset --hard' at then end, voiding the previously run 'pq import'

[Julien Muchembled]

$ gbp clone g...@salsa.debian.org:python-team/modules/zc-lockfile.git
gbp:info: Cloning from 
'g...@salsa.debian.org:python-team/modules/zc-lockfile.git'
$ cd zc-lockfile
$ git branch
* debian/master
  pristine-tar
  upstream
$ gbp pq import
gbp:info: Trying to apply patches at '8c7a785b55b4b94093dd2b634ebfcb5d4f238371'
gbp:warning: Patch 'no_doc_txt.patch' has no authorship information, using 
'Debian/Ubuntu Zope Team '
gbp:info: 1 patches listed in 'debian/patches/series' imported on 
'patch-queue/debian/master'
$ git log -2 --oneline --graph debian/master @
* 8e236e0 (HEAD -> patch-queue/debian/master) no_doc_txt
*   8c7a785 (origin/debian/master, origin/HEAD, debian/master) Merge 
remote-tracking branch 'origin/svn'
|\
$ gbp import-orig 
https://files.pythonhosted.org/packages/f5/fe/efb94907d8b2b81c3beab1bd628ff67e310d82816b94aa00b52062727ea9/zc.lockfile-1.3.0.tar.gz
What is the upstream version? [1.3.0] 
gbp:info: Importing '../zc.lockfile-1.3.0.tar.gz' to branch 'upstream'...
gbp:info: Source package is zc.lockfile
gbp:info: Upstream version is 1.3.0
gbp:info: Replacing upstream source on 'debian/master'
gbp:info: Successfully imported version 1.3.0 of ../zc.lockfile-1.3.0.tar.gz
$ git log -2 --oneline
b465d76 (HEAD -> patch-queue/debian/master, debian/master) Update upstream 
source from tag 'upstream/1.3.0'
5706e9b (tag: upstream/1.3.0, upstream) New upstream version 1.3.0
$ 

Commit 8e236e0 is lost and can't be rebased. I expect patch-queue/debian/master 
to still point to it.

Bug#900781: git-buildpackage: import-orig does 'reset --hard' at then end, voiding the previously run 'pq import'

[Julien Muchembled]

Package: git-buildpackage
Version: 0.9.9
Severity: normal
Tags: upstream

Both the git-buildpackage documentation and
  https://wiki.debian.org/Python/GitPackagingPQ

invite the user to first run 'gbp pq import' before using 'import-orig'.

However, since the 'replace' merge mode, which is actually the default mode
for most packages, 'import-orig' resets the patch queue branch at the end.

Which means that all patches are lost and executing
'gbp pq rebase' + 'gbp pq export' do nothing.

-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (900, 'testing'), (600, 'unstable'), (500, 'unstable-debug'), 
(500, 'testing-debug'), (500, 'proposed-updates'), (500, 'stable'), (1, 
'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.1.51+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8), LANGUAGE= 
(charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages git-buildpackage depends on:
ii  devscripts 2.18.3
ii  git1:2.14.1-3
ii  man-db 2.7.6.1-2
pn  python3
ii  python3-dateutil   2.6.1-1
ii  python3-pkg-resources  36.6.0-1

Versions of packages git-buildpackage recommends:
ii  pbuilder  0.229.2
ii  pristine-tar  1.44
ii  python3-requests  2.18.1-1

Versions of packages git-buildpackage suggests:
pn  python3-notify2  
ii  sudo 1.8.21p2-1
ii  unzip6.0-21

-- no debconf information

Bug#880637: libcrypto++-dev: The amd64 package is huge because it contains debug_info

[Julien Muchembled]

Package: libcrypto++-dev
Version: 5.6.4-8
Severity: normal

'apt show -a libcrypto++-dev' on amd64:

VersionDownload-Size  Installed-Size
5.6.4-71 296 kB   13,0 MB
5.6.4-89 202 kB   81,6 MB

https://packages.debian.org/unstable/libcrypto++-dev to compare with other 
archs.

I inspected using 'ar -x /usr/lib/x86_64-linux-gnu/libcrypto++.a' and then, for 
example:

$ du -h */dll.o; file */dll.o
1,9M5.6.4-7/dll.o
11M 5.6.4-8/dll.o
5.6.4-7/dll.o: ELF 64-bit LSB relocatable, x86-64, version 1 (GNU/Linux), not 
stripped
5.6.4-8/dll.o: ELF 64-bit LSB relocatable, x86-64, version 1 (GNU/Linux), with 
debug_info, not stripped

Bug#740096: snapshot.debian.org: please enable HTTPS for https://snapshot.debian.org

[Julien Muchembled]

Control: severity -1 normal

Sorry for my stupid mistake with https://bugs.debian.org/860260
However, the arguments I gave there remain valid, so I allow myself to raise 
the severity of this bug. (I'd even add that I may not have done that mistake 
if it was HTTPS to beginning with.)

Bug#860260: snapshot.debian.org: corrupted binary package: binutils_2.22-8_amd64.deb

[Julien Muchembled]

Package: snapshot.debian.org
Severity: important

Hello,

When I download
  
http://snapshot.debian.org/archive/debian/20130223T095106Z/pool/main/b/binutils/binutils_2.22-8_amd64.deb

from
  http://snapshot.debian.org/package/binutils/2.22-8/#binutils_2.22-8

I get:
  $ ls -l binutils_2.22-8_amd64.deb 
  -rw-r--r-- 1 jm jm 4799776 Feb 23  2013 binutils_2.22-8_amd64.deb
  $ md5sum binutils_2.22-8_amd64.deb
  11ff1f1d331c608aebb6d2585d601522  binutils_2.22-8_amd64.deb

whereas both the snapshot.d.o page and https://tracker.debian.org/news/432162
shows that the md5sum must be

  3d1fb7c57aa32ef5a122cb832a9f83de7e3b2a71

The size of the file is correct.

BTW, the severity of #740096 ("please enable HTTPS") should be raised.
I also don't agree with the answer on #820423:

> snapshot.d.o provides read-only snapshots of the archive, it does not
> modify any files.

All this shows that some authentication mechanism is important, for 2 reasons:

1. unintentional data corruption, which is probably the case for the above file
   (bitflip by hardware ?)

2. MITM, and to protect against this when downloading binary package is to
   check the hashes on the related news on https://tracker.debian.org/,
   which I always do and it's very annoying.

Regards,
Julien

Bug#858253: fssync: FTBFS randomly (AssertionError: Lists differ: [b'a', b'a/c'] != [b'a/c'])

[Julien Muchembled]

Thanks for your report.

Le 03/20/17 à 13:32, Santiago Vila a écrit :
> ./test
> .missing b'b/c' on destination side
> missing b'b' on destination side
> ... create new inode for b'a'
> b'a/c': missing
> F
> ==
> FAIL: test2 (__main__.Test)
> --
> Traceback (most recent call last):
>   File "./test", line 371, in test2
> self.assertListEqual([b'a', b'a/c'], list(self.fssync.check(b'')))
> AssertionError: Lists differ: [b'a', b'a/c'] != [b'a/c']
> 
> First differing element 0:
> b'a'
> b'a/c'
> 
> First list contains 1 additional elements.
> First extra element 1:
> b'a/c'
> 
> - [b'a', b'a/c']
> + [b'a/c']
> 
> --
> Ran 2 tests in 0.382s

This error was recently reported by reproducible-builds (2017-02-04):
  https://tests.reproducible-builds.org/debian/history/fssync.html

At that time, I already tried to reproduce it on my machine, without success.
I suspected reprotest and tried it. Your report really confirms it's something 
else.

> The bug should be reproducible with sbuild on a single CPU virtual machine,
> provided you try enough times (as the failure happens randomly).

OK, so I've just set up a single-CPU VM. Still unable to reproduce the failure.

I don't think the number of cores matters here.

The issue seems to be at the line 368 of 'test':
os.rename(x + b'c', x + b'd')

The test assumes that this changes the modification time of the folder pointed 
to by x. On my side, it's increased by a few ms.

Maybe your /tmp folder use a FS with a very low time precision. I tried tmpfs 
and ext4.

Julien

Bug#854926: sagemath: useless (Build-)Depends: on python-zodb

[Julien Muchembled]

Source: sagemath
Severity: normal

$ grep -ri zodb
debian/control.runtime-depends: python-zodb,
debian/control: python-zodb,
debian/control: python-zodb,
sage/src/doc/en/reference/databases/index.rst:Sage includes the ZOPE object 
oriented database ZODB, which
sage/build/pkgs/conway_polynomials/SPKG.txt: * #12205: Rewrite database to not 
use ZODB

The last line says everything. (And the upstream documentation should stop 
referring to it.)

(I don't use sagemath and that's the only dep I have checked)

The long story is that I was surprised to see the popcon stat of python-zodb 
increasing so much recently. Which would be nice if it was for a good reason 
(after reading https://trac.sagemath.org/ticket/10353 I think that upstream 
underestimates ZODB but well, maybe there are other DB that fit better for 
their use). Graph of both sagemath and python-zodb:
  
https://qa.debian.org/popcon-graph.php?packages=sagemath+python-zodb_installed=on_legend=on_ticks=on_date=2016-11-01_date=_date=_fmt=%25Y-%25m=1

In any case, I'll have to care about reverse dependencies. There will be major 
updates of ZODB this year.


Julien

Bug#852148: RFS: zodb/1:3.10.7-2

[Julien Muchembled]

Package: sponsorship-requests
Severity: normal

 Dear mentors,

 I am looking for a sponsor for my package "zodb"

* Package name: zodb
  Version : 1:3.10.7-2
  Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
* URL : http://zodb.org/
* License : Zope-2.1
  Section : python

 It builds this binary package:

   python-zodb - Zope Object Database (ZODB)

 To access further information about this package, please visit the following 
URL:

   https://mentors.debian.net/package/zodb

 Alternatively, one can download the package with dget using this command:

   dget -x https://mentors.debian.net/debian/pool/main/z/zodb/zodb_3.10.7-2.dsc

 Changes since the last upload:

zodb (1:3.10.7-2) unstable; urgency=medium

  * Team upload.
  * New upstream release 3.10.7
- drop patches merged upstream:
  Fix-possible-data-corruption-after-FileStorage-is-tr.patch
  * Use pybuild instread of van.pydeb.
  * Use debhelper 10.
  * d/control:
- Drop 'Conflicts: zope3': this package was removed long time ago.
- Update Standards-Version to 3.9.8 (no change needed).
- Use HTTPS for Vcs-Browser.
- Set section back to 'python'. As clarified upstream:
  « ZODB doesn't depend on Zope in any way and is used in many projects
that have nothing to do with Zope. »
  This also fixes an override disparity.
  * d/copyright: update Copyright: year.
  * d/watch: fix by refreshing it from pypi.debian.net

 Regards,
  Julien Muchembled



signature.asc
Description: OpenPGP digital signature

Bug#852136: RFS: python-btrees/4.3.1-1 [ITP]

[Julien Muchembled]

Package: sponsorship-requests
Severity: normal

 Dear mentors,

 I am looking for a sponsor for my package "fssync"

* Package name: fssync
  Version : 1.6-1
  Upstream Author : Julien Muchembled <j...@jmuchemb.eu>
* URL : http://jmuchemb.eu/fssync.git
* License : GPL-3.0
  Section : utils

 It builds this binary package:

   fssync - File system synchronization tool (1-way, over SSH)

 To access further information about this package, please visit the following 
URL:

 https://mentors.debian.net/package/fssync

 Alternatively, one can download the package with dget using this command:

   dget -x https://mentors.debian.net/debian/pool/main/f/fssync/fssync_1.6-1.dsc

 Changes since the last upload:

fssync (1.6-1) unstable; urgency=medium

  * New upstream release.
  * Update Standards-Version to 3.9.8 (no change needed).
  * Use debhelper 10 (no change).
  * Unit tests are now run during build, so switch build-depends to python3-*
and add python3-pylibacl.
  * d/copyright: update Copyright: year

 Upstream changes since the last upload:

1.6 (2017-01-21)


- --check option crashed on destination when it has untracked inodes
  whose paths are greater (alphabetically) than any path on source side.

 Regards,
  Julien Muchembled



signature.asc
Description: OpenPGP digital signature

Bug#850564: override: python-zconfig: python/optional, python3-zconfig: python/optional

[Julien Muchembled]

Package: ftp.debian.org
Severity: normal

Hello,

I request 2 override changes for zconfig:

* python-zconfig: change priority from extra to optional

> zconfig (2.7.1-2) unstable; urgency=low
> 
>   [...]
>   * Set priority to optional as we are depended on by python-zodb which is
> priority optional.
>   [...]
>  -- Fabio Tranchitella   Sun, 08 Nov 2009 11:26:46 +0100

The disparity is currently reported at
  https://packages.qa.debian.org/z/zconfig.html

* python3-zconfig: change section from zope to python

As pointed by #849620, « The package ‘python3-zconfig’ installs primarily a 
library, of interest only to Python programmers or as dependencies to other 
packages. »

This also makes python3-zconfig consistent with python-zconfig, whic is already 
in 'python' section.

Then I'll change debian/control to 'python' section for both packages.

Regards,
Julien



signature.asc
Description: OpenPGP digital signature

Bug#783377: Packaging new version of ZODB (Zope Object Database)

[Julien Muchembled]

Updated status of ZODB packaging.

I didn't write it in my previous mail because there was still a lot time before 
deadlines but now, it's becoming short. I'd like to see ZODB updated for 
Stretch, at least that's what I had in mind when I started months ago. Having a 
very old version of ZODB until 2019, without even support for python 3, would 
be frustrating.

I am actually looking for sponsors who could upload them in time. They all 
should be in good state. Maybe Arnaud will be available again for the last 
uploads.

Here is the dependency graph of packages that need(ed) to be updated (I forgot 
to mention a few of them in my previous package):

zdaemon -> python-zc.customdoctests, zconfig
zodb -> python-persistent, transaction, zodbpickle, python-btrees, zconfig, 
zc.lockfile (bin-new)
zeo -> zodb, zdaemon
(+ zodb recommending zeo)

Packages would be uploaded in 3 rounds:
(1) python-zc.customdoctests (new), zconfig (bin-new)
(2) python-persistent, transaction, zodbpickle (new), python-btrees (new), 
zc.lockfile, zdaemon (bin-new)
(3) zodb (bin-new), zeo (new)

(1) is done already.

(2) 3 are on mentors.debian.net
  https://mentors.debian.net/package/python-btrees (RFS #847609)
  https://mentors.debian.net/package/python-persistent
  https://mentors.debian.net/package/zodbpickle (RFS #847608)
  
transaction, zc.lockfile and zdaemon are also finished on my side, but I guess 
the priority is the new packages.

They all should be in good state.

(3) So that zodb/zeo are uploaded and accepted before the 26th


About VCS, it's too late to think about it now. I'm interested in this topic so 
I'll be happy to work on this once ZODB is updated. For the moment:
- I followed blindly the policy for the new 
python-zc.customdoctests/zodbpickle/python-btrees packages
- I stay on svn://anonscm.debian.org/pkg-zope/ for the packages of the Zope 
team, and for the new 'zeo'.


Regards,
Julien

Bug#847609: RFS: python-btrees/4.3.1-1 [ITP]

[Julien Muchembled]

Package: sponsorship-requests
Severity: wishlist

 Dear mentors,

 I am looking for a sponsor for my package "python-btrees"

* Package name: python-btrees
  Version : 4.3.1-1
  Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
* URL : https://github.com/zopefoundation/BTrees
* License : Zope-2.1
  Section : python

 It builds those binary packages:

   python-btrees - scalable persistent object containers for Python
   python-btrees-doc - scalable persistent object containers for Python - 
documentation
   python3-btrees - scalable persistent object containers for Python 3

 To access further information about this package, please visit the following 
URL:

 https://mentors.debian.net/package/python-btrees


 Alternatively, one can download the package with dget using this command:

   dget -x 
https://mentors.debian.net/debian/pool/main/p/python-btrees/python-btrees_4.3.1-1.dsc

 Changes since the last upload:

python-btrees (4.3.1-1) unstable; urgency=low

  * Initial release (Closes: #842874).

 Regards,
  Julien Muchembled

Bug#847607: RFS: zconfig/3.1.0-1

[Julien Muchembled]

Package: sponsorship-requests
Severity: normal

  Dear mentors,

  I am looking for a sponsor for my package "zconfig".

 * Package name: zconfig
   Version : 3.1.0-1
   Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
 * URL : https://pypi.python.org/pypi/ZConfig
 * License : Zope-2.1
   Section : zope

  It builds those binary packages:

python-zconfig - Structured Configuration Library, for Python 2
python3-zconfig - Structured Configuration Library, for Python 3

  To access further information about this package, please visit the following 
URL:

  https://mentors.debian.net/package/zconfig


  Alternatively, one can download the package with dget using this command:

dget -x 
https://mentors.debian.net/debian/pool/main/z/zconfig/zconfig_3.1.0-1.dsc

  Changes since the last upload:

zconfig (3.1.0-1) unstable; urgency=low

  * Team upload.

  [ Gediminas Paulauskas ]
  * New upstream release 2.9.3
  * debian/control:
- Drop python-van.pydeb from Build-Depends
- Remove pydeb substvars.
- Bump Standards-Version to 3.9.4
- Enable autopkgtest. Closes: #692692.
- Change homepage to pypi.
  * debian/rules: remove the use of pydeb, build with python2 only.
  * debian/tests: switch to zope.testrunner.
  * debian/source/options: ignore .egg-info directory.

  [ Julien Muchembled ]
  * New upstream release 3.1.0
+ d/copyright: update Copyright: year
+ d/docs: remove NEWS.txt (it was renamed to CHANGES.txt upstream).
  * Add support for Python 3 (new python3-zconfig package).
  * d/watch: refresh from pypi.debian.net, mainly for HTTPS access.
  * d/control:
+ Remove useless X-Python-Version (>= 2.6 is implicit).
+ Update Standards-Version to 3.9.8 (no change needed).
+ Use canonical URI for Vcs-Svn and Vcs-Browser.
  * Use debhelper 10.

  Regards,
   Julien Muchembled

Bug#847608: RFS: zodbpickle/0.6.0-1 [ITP]

[Julien Muchembled]

Package: sponsorship-requests
Severity: wishlist

 Dear mentors,

 I am looking for a sponsor for my package "zodbpickle"

* Package name: zodbpickle
  Version : 0.6.0-1
  Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
* URL : https://github.com/zopefoundation/zodbpickle
* License : Python, Zope-2.1
  Section : python

 It builds those binary packages:

   pypy-zodbpickle - Fork of pickle module, for ZODB on PyPy (Python 2)
   python-zodbpickle - Fork of pickle module, for ZODB
   python3-zodbpickle - Fork of pickle module, for ZODB on Python 3

 To access further information about this package, please visit the following 
URL:

 https://mentors.debian.net/package/zodbpickle


 Alternatively, one can download the package with dget using this command:

   dget -x 
https://mentors.debian.net/debian/pool/main/z/zodbpickle/zodbpickle_0.6.0-1.dsc

 Changes since the last upload:

zodbpickle (0.6.0-1) unstable; urgency=low

  * Initial release (Closes: #842870).

 Regards,
  Julien Muchembled

Bug#783377: Packaging new version of ZODB (Zope Object Database)

[Julien Muchembled]

Le 11/01/16 à 20:10, Julien Muchembled a écrit :
> * python-btrees & python-zodbpickle (done)
> https://github.com/zopefoundation/BTrees
> https://github.com/zopefoundation/zodbpickle
> 
> - new packages
> - Debian Python Modules Team
> 
> [...]
> 
> * python-zeo (not started)
> https://github.com/zopefoundation/ZEO
> 
> [...]
> 
> * zc.customdoctests (done)
> https://github.com/zopefoundation/zc.customdoctests
> 
> - new package
> - Debian Python Modules Team
> 
> I haven't done any ITP yet.

Bug#842870: ITP: zodbpickle -- Fork of pickle module, for ZODB
Bug#842874: ITP: python-btrees -- scalable persistent object containers for 
Python
Bug#842876: ITP: zc.customdoctests -- Use Python doctest with other languages

I'll do ZEO later.

Julien

Bug#842876: ITP: zc.customdoctests -- Use Python doctest with other languages

[Julien Muchembled]

Package: wnpp
Severity: wishlist
Owner: Julien Muchembled <j...@jmuchemb.eu>

* Package name: zc.customdoctests
  Version : 1.0.1+1.gc142624-1
  Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
* URL : https://github.com/zopefoundation/zc.customdoctests
* License : Zope-2.1
  Programming Lang: Python
  Description : Use Python doctest with other languages

 doctest (and recently manuel) provide hooks for using custom doctest
 parsers. zc.customdoctests helps to leverage this to support other
 languages, such as JavaScript (with python-spidermonkey):
 .
 js> function double (x) {
 ... return x*2;
 ... }
 js> double(2)
 4
 .
 And with python-manuel, it facilitates doctests that mix multiple languages,
 such as Python, JavaScript, and sh.

This is a required dependency for the new version of ZODB, and I plan to
maintain it as part of the Debian Python Modules Team. See also

 
https://lists.debian.org/msgid-search/8e7791a6-97fd-6a68-8bc2-dee31266c...@jmuchemb.eu

Bug#842874: ITP: python-btrees -- scalable persistent object containers for Python

[Julien Muchembled]

Package: wnpp
Severity: wishlist
Owner: Julien Muchembled <j...@jmuchemb.eu>

* Package name: python-btrees
  Version : 4.3.1-1
  Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
* URL : https://github.com/zopefoundation/BTrees
* License : Zope-2.1
  Programming Lang: Python
  Description : scalable persistent object containers for Python

 This package contains a set of persistent object containers built around a
 modified BTree data structure. The trees are optimized for use inside ZODB's
 “optimistic concurrency” paradigm, and include explicit resolution of
 conflicts detected by that mechanism.

This is a required dependency for the new version of ZODB, and I plan to
maintain it as part of the Debian Python Modules Team. See also

 
https://lists.debian.org/msgid-search/8e7791a6-97fd-6a68-8bc2-dee31266c...@jmuchemb.eu

Bug#842870: ITP: zodbpickle -- Fork of pickle module, for ZODB

[Julien Muchembled]

Package: wnpp
Severity: wishlist
Owner: Julien Muchembled <j...@jmuchemb.eu>

* Package name: zodbpickle
  Version : 0.7.0~1.gc9e09e3-1
  Upstream Author : Zope Foundation and Contributors <zope-...@zope.org>
* URL : https://github.com/zopefoundation/zodbpickle
* License : Python, Zope-2.1
  Programming Lang: Python
  Description : Fork of pickle module, for ZODB

 This package presents a uniform pickling interface for ZODB:
 .
 Under Python 2, this package forks both Python 2.7’s pickle and cPickle
 modules, adding support for the protocol 3 opcodes. It also provides a new
 subclass of bytes, zodbpickle.binary, which Python2 applications can use to
 pickle binary values such that they will be unpickled as bytes under Python 3.
 .
 Under Python 3, this package forks the pickle module (and the supporting C
 extension) from both Python 3.2 and Python 3.3. The fork adds support for the
 noload operations used by ZODB.

This is a required dependency for the new version of ZODB, and I plan to
maintain it as part of the Debian Python Modules Team. See also

 
https://lists.debian.org/msgid-search/8e7791a6-97fd-6a68-8bc2-dee31266c...@jmuchemb.eu

Bug#783377: Packaging new version of ZODB (Zope Object Database)

[Julien Muchembled]

Hello,

I am co-maintaining the 'zodb' package and I am also an upstream developer.
  http://www.zodb.org/ - a native object database for Python

Debian currently packages branch 3.x of ZODB, which is getting quite old, with 
only support for Python 2. It's been a while that I am working locally in 
packaging a newer version, and I have almost finished.

With branch 4.x, ZODB has been splitted into 4 packages (-> means Depends):
 ZEO -> ZODB -> (Btrees, persistent)
The split of persistent was already backported.

And in addition to that:
- new dependencies that are not packaged yet
- some existing dependencies needs to be updated as well

I write to debian-python, because some of the involved packages are not 
specific to Zope. Actually, I even think that ZODB itself is not specific to 
Zope, but well, changing section of existing packages can be another topic. 
I've just requested to join the Python Modules Packaging Team.

So here is the status of my work:

* Generally, I add support for Python 3 everywhere it is missing. I only added 
PyPy support when it's trivial (e.g. zodbpickle), because that would need work 
on even more existing packages.

* python-zodb (almost finished)

- changelog entries to be written
- decide an upgrade plan for the split of ZEO (some users may have installed 
zodb for it): maybe a Recommends: ?
- to give a better overview, here's the new Depends: for Python2:
  python-btrees (>= 4.2.0), python-persistent (>= 4.2.0), python-six, 
python-transaction (>= 1.5.0), python-zc.lockfile, python-zconfig, 
python-zodbpickle (>= 0.6.0), python-zope.interface, python:any (<< 2.8), 
python:any (>= 2.7.5-5~)

* python-persistent (done)

- trivial update to 4.2.1
- I'd like to commit to python-persistent.git (python-modules), or someone does 
it for me

* python-btrees & python-zodbpickle (done)
https://github.com/zopefoundation/BTrees
https://github.com/zopefoundation/zodbpickle

- new packages
- Debian Python Modules Team

* python-transaction (done)

- update to version 1.6.1
- svn repos is late compared to the packaged version

* python-zeo (not started)
https://github.com/zopefoundation/ZEO

* python-zdaemon (done)

- needs to be updated for ZEO
- new deps: mock, zc.customdoctests

* zc.customdoctests (done)
https://github.com/zopefoundation/zc.customdoctests

- new package
- Debian Python Modules Team

I haven't done any ITP yet.

Julien

Bug#842201: hw-detect: Spurious "No network interfaces detected" under QEMU

[Julien Muchembled]

Package: hw-detect
Version: 1.108
Severity: normal
Tags: d-i

Using preseeding, I came into the problem that the installer occasionally fails
because it could not detect the network interface. There really is one, and if
I connect to the guest with VNC, I can go back, retry and it works.

Maybe things are going too fast at some point in a fully automated install.
I started to debug and I found that the issue is probably in the
/bin/hw-detect script, when it is called by /bin/ethdetect

I use the following command line to debug:

  $ qemu-system-x86_64 -k fr -enable-kvm -smp 1 -m 256 -net nic -net 
user,hostfwd=tcp:127.0.0.1:-:22 -cdrom debian-8.6.0-amd64-netinst.iso 
-kernel vmlinuz -initrd initrd.gz -vnc 127.0.0.1:1 -append 'priority=critical 
language=C country=FR keymap=fr(latin9) 
anna/choose_modules=network-console,openssh-client-udeb 
network-console/password=root network-console/password-again=root'

where vmlinuz & initrd.gz are extracted from the iso (from install.amd folder).

In order to easily reproduce the issue, I added the following line to the
/etc/inittab file in the initrd:

  tty5::once:tail -f -n 0 /var/log/syslog |(grep -m1 'rdnssd started'; reboot)

Now, the really strange thing is that the issue disappears if I add

  fb=false

to the kernel command line. At least, I have a workaround, but for how long?


Tested with qemu-system-x86 1:2.7+dfsg-1 on amd64.
Using '-net nic,model=virtio' does not change anything.

Bug#798096: usrmerge: Conversion fails if libautodie-perl package is installed

[Julien Muchembled]

Package: usrmerge
Version: 2
Severity: important

$ /usr/lib/convert-usrmerge
Undefined subroutine called at /usr/lib/convert-usrmerge line 248.

After uninstalling 'libautodie-perl' package:
$ /usr/lib/convert-usrmerge
The system has been successfully converted.

Even the last packaged version of libautodie-perl (2.29-1) breaks usrmerge,
so you should be able to reproduce it easily. I wonder why this package was
still there on my system since autodie is now provided by perl-modules.

And actually, it took me some time to understand the issue, but hopefully the
script is mostly idempotent. After several partial failures by trying to
modify the script, I had to manually delete /bin~delete~usrmerge~~ and
/sbin~delete~usrmerge~~ and the end. I don't know if it's related but libusb
couldn't be found after the conversion, breaking a few things: I had to run
ldconfig. debsums didn't report any error and my system now works fine.

Bug#778151: Processed: tuxonice-userui: diff for NMU version 1.1+dfsg1.gc3bdd83-3.1

[Julien Muchembled]

Le 07/19/15 19:56, gregor herrmann a écrit :
 On Sun, 19 Jul 2015 18:50:53 +0200, Julien Muchembled wrote:
 
 I am really sorry for wasting the time of everyone here.
 In fact, the job is already done but I have no upload right and my usual 
 sponsor lacks time currently (well, I preferred we finished something about 
 another RC-buggy package before asking him for this one).
 
 No worries.
 For the future: please mention this in the bug report (and tag the
 bug pending).
  
 So, there are other important bugfixes for tuxonice-userui, as you can see 
 on my own repository:
   http://jmuchemb.eu/tuxonice-userui.git [1]
 I only need to make this ready for release (write messages for the last 2 
 commits and update changelog).

 I'm available to finish if anyone agrees to sponsor all this.
 
 The other changes since -3 look good, so if you finish d/changelog
 I'm happy to sponsor the package.
 Just shout when you've pushed the final touches to your git repo.

http://jmuchemb.eu/tuxonice-userui.git updated
I'll push to collab-maint if everything's fine.



signature.asc
Description: OpenPGP digital signature

Bug#778151: Processed: tuxonice-userui: diff for NMU version 1.1+dfsg1.gc3bdd83-3.1

[Julien Muchembled]

Hello,

I am really sorry for wasting the time of everyone here.
In fact, the job is already done but I have no upload right and my usual 
sponsor lacks time currently (well, I preferred we finished something about 
another RC-buggy package before asking him for this one).

So, there are other important bugfixes for tuxonice-userui, as you can see on 
my own repository:
  http://jmuchemb.eu/tuxonice-userui.git [1]
I only need to make this ready for release (write messages for the last 2 
commits and update changelog).

I'm available to finish if anyone agrees to sponsor all this.

Regards,
Julien


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#762097: nat-rtsp-dkms: not work

[Julien Muchembled]

Le 12/26/14 09:22, Eduardo a écrit :
 El 25/12/14 a las 07:23, Eduardo escribió:
 ...

   I have written a tiny rtsp server and client. I created, with kvm, three 
 virtual machines, a firewall (router), a server and a client,
 ...
 
 Sorry, there are some errors in server and client, try this new version.
 

I've just tested your scripts and unfortunately, I couldn't reproduce the 
reported bug.
But at least, they require nf_nat_rtsp on the router so I'll make a unit test 
from them.

2 attached files:
- The 'test' script must be run as root in the same directory as your scripts. 
It creates 3 machines using network namespaces and runs the server  client at 
the end. The test fails if the module is not loaded before.
- 'rtsp_output' is what I get on my machine. Server  client outputs are mixed.
[*] father 21984, child 22006
[22006] describe rtsp://10.0.0.1:554/vod1_diascontadosV2PR.mpi RTSP/1.0
[22006] describe tx
RTSP/1.0 200 OK
CSeq: 2
Date: Fri, 05 Dec 2014 01:31:05 GMT
Content-Type: application/sdp
Content-Length: 309

v=0
o=- 1417743065482550 4201396289678 IN IP4 10.0.0.2
s=RTSP Session
t=0 0
i=
b=AS:2800
a=type:vod
a=range:npt=0-5457.661
c=IN IP4 0.0.0.0
a=control:rtsp://0.0.0.0:554/vod1_diascontadosV2PR.mpi
m=video 0 RTP/AVP 33
a=framerate:25.00
a=pmt:ArAdAALBAADh4fAAG+Hh8AAD4eLwBgoERVNQAJCKMR8=


[22006] setup rtsp://10.0.0.1:554/vod1_diascontadosV2PR.mpi RTSP/1.0
[22006] setup tx
RTSP/1.0 200 OK
CSeq: 3
Session: 3424382056990345778;timeout=180
Transport: MP2T/H2221/UDP;unicast;destination=0.0;client_port=27910-27911


[*] father 22005, child 22007
[*] listen udp 10.0.1.2:27910
[22006] play rtsp://10.0.0.1:554/vod1_diascontadosV2PR.mpi RTSP/1.0
RTSP/1.0 200 OK
CSeq: 4
Session: 3424382056990345778;timeout=180
Scale: 1.0
Range: npt=0.000-
x-playNow:
x-noFlush:


[*] father 22006, child 22011
Conecting rtp 10.0.0.2:27910
Hello world!
[22006] teardown rtsp://10.0.0.1:554/vod1_diascontadosV2PR.mpi RTSP/1.0
[22006] teardown tx
RTSP/1.0 200 OK
CSeq: 5
Session: 3424382056990345778;timeout=180
Scale: 1.0
Range: npt=0.000-
x-playNow:
x-noFlush:
#!/bin/bash
# XXX: There's no bashism but 'trap' is buggy in dash:
#https://bugs.debian.org/390433

set -e
ns=test-nat-rtsp-$$
server() { ip netns exec $ns.server $@; }
router() { ip netns exec $ns.router $@; }
client() { ip netns exec $ns.client $@; }

trap 'set +e; eval $at_exit' 0

for i in server router client; do
  at_exit=ip netns del $ns.$i; $at_exit
  ip netns add $ns.$i
  $i ip link set lo up
done

ip -b - EOF
link add name server type veth peer name router
link set server netns $ns.router
link set router netns $ns.server

link add name client type veth peer name router
link set client netns $ns.router
link set router netns $ns.client
EOF

server ip link set router up; server ip addr add 10.0.0.1/24 dev router
router ip link set server up; router ip addr add 10.0.0.2/24 dev server
router ip link set client up; router ip addr add 10.0.1.1/24 dev client
client ip link set router up; client ip addr add 10.0.1.2/24 dev router

router sysctl net.ipv4.ip_forward=1
router iptables -t nat -A POSTROUTING -o server -j MASQUERADE
client ip route add default via 10.0.1.1

client ping -q -c 1 10.0.0.1 /dev/null
echo Network ready.

#bash -i
#exit

at_exit=pkill -f ./rtsp_server.pl; $at_exit
server ./rtsp_server.pl 
sleep 1 # XXX
client ./rtsp_client.pl 10.0.0.1

exit
# With VLC as RTSP client  server, it works without nf_nat_rtsp module.

vlm_conf=`mktemp`
at_exit=rm $vlm_conf; $at_exit
cat EOF $vlm_conf
new tst vod enabled
setup tst input /dev/urandom option demux=rawvid option rawvid-fps=10 option 
rawvid-width=64 option rawvid-height=64 option rawvid-chroma=YV12 output 
#transcode{vcodec=mp4v}
EOF
chmod +r $vlm_conf

at_exit=pkill -f $vlm_conf; $at_exit
set su -s /bin/sh nobody -c
server $@ exec vlc -d -I dummy -A adummy -V vdummy --rtsp-port 8080 
--vlm-conf $vlm_conf

bash -i

#client $@ exec vlc --play-and-exit -I dummy -A adummy -V vdummy 
rtsp://10.0.0.1:8080/tst

Bug#774393: fssync: possible data corruption on destination side

[Julien Muchembled]

Package: fssync
Severity: grave
Tags: fixed-upstream
Justification: causes non-serious data loss

I fixed 2 serious bugs upstream. In some cases, the DB was wrongly updated,
which would later cause useless resync or corruption.

See 
http://jmuchemb.eu/fssync.git/commit/45a3425f48a42d415887333f67cb317b01df24da?js=1
and 
http://jmuchemb.eu/fssync.git/commit/5f98a3720f39d75d993051be80d928aedbeb305e?js=1
for more information.

I also fixed another (non-RC) bug which required a small change in the protocol:
  
http://jmuchemb.eu/fssync.git/commit/2e05ddbb78ba5f9a4d272a1fa5a2f05a3312c01e?js=1
If an upload is done only with the first 2 patches, the stable version would
immediately get non-interoperable with the testing/sid when Jessie is released.

I'd like to tag the current master branch as 1.5 and that this new version
is part of Jessie. Otherwise, I prefer not to waste time and that fssync gets
removed from testing.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#773727: mariadb-server-10.0: Mroonga TokuDB plugins missing

[Julien Muchembled]

Package: mariadb-server-10.0
Version: 10.0.15-2
Severity: normal
Tags: patch

Hello,

The attached patch fixes it. Could you please apply it?

(I don't know why MariaDB decided to bundle Mroonga. For me, it looks saner
to build a source package, like mysql-source-5.5, and have a separate 'mroonga'
source package from upstream.)

Julien
From: Julien Muchembled j...@jmuchemb.eu
Date: Mon, 22 Dec 2014 17:50:14 +0100
Subject: [PATCH] Fix inclusion of Mroonga  TokuDB plugins in
 mariadb-server-10.0

---
 debian/rules | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/debian/rules b/debian/rules
index 3c5c441..a38bee2 100755
--- a/debian/rules
+++ b/debian/rules
@@ -137,10 +137,10 @@ override_dh_auto_install:
 	dh_testroot
 	
 	# If TokuDB plugin was built add it to the server install list.
-	[ ! -f $(BUILDDIR)/usr/lib/mysql/plugin/ha_tokudb.so ] || echo 'usr/lib/mysql/plugin/ha_tokudb.so\netc/mysql/conf.d/tokudb.cnf\nusr/bin/tokuftdump\nusr/share/doc/mariadb-server-10.0/README-TOKUDB\nusr/share/doc/mariadb-server-10.0/README.md'  debian/mariadb-server-10.0.install
+	[ ! -f $(BUILDDIR)/storage/tokudb/ha_tokudb.so ] || echo 'usr/lib/mysql/plugin/ha_tokudb.so\netc/mysql/conf.d/tokudb.cnf\nusr/bin/tokuftdump\nusr/share/doc/mariadb-server-10.0/README-TOKUDB\nusr/share/doc/mariadb-server-10.0/README.md'  debian/mariadb-server-10.0.install
 	
 	# If Mroonga plugin was built add it to the server install list.
-	[ ! -f $(BUILDDIR)/usr/lib/mysql/plugin/ha_mroonga.so ] || echo 'usr/lib/mysql/plugin/ha_mroonga.so'  debian/mariadb-server-10.0.install
+	[ ! -f $(BUILDDIR)/storage/mroonga/ha_mroonga.so ] || echo 'usr/lib/mysql/plugin/ha_mroonga.so'  debian/mariadb-server-10.0.install
 	
 	# If libthrift-dev was available (manually installed, as it is not in Debian)
 	# and ha_cassandra.so was thus built add it to the server install list.
-- 
2.1.3

Bug#770958: src:linux: Please enable CONFIG_ZRAM_LZ4_COMPRESS

[Julien Muchembled]

Package: src:linux
Version: 3.16.7-2
Severity: wishlist

Dear Maintainer,

For compressed RAM block devices, 3.15 added LZ4 compression in addition to LZO.
LZO remains the default selected algorithm so it's a safe change.
Could you please enable CONFIG_ZRAM_LZ4_COMPRESS ?

I hope it's not too late for Jessie.

Thanks.

-- System Information:
Debian Release: jessie/sid
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (300, 'experimental'), (200, 
'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.16.7+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#762097: nat-rtsp-dkms: not work

[Julien Muchembled]

Thank you for the detail. And just an email to write what I've done so far.

If I'm going to contribute upstream, I prefer to first have a way to easily 
test things, so I started to write a unit test which sets up 3 nodes (server, 
router  client) using network namespaces.

However, I'm stuck because it works without nf_nat_rtsp loaded. I tried VLC  
GStreamer as server. I discovered that UDP streams are initiated by the client 
and that both servers reply to correct source port so simple NAT is of course 
enough.

Just wondering why there are rtsp servers in the wild that don't simply work 
like that. I guess you don't know any server I could use to finish this unit 
test.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#762097: nat-rtsp-dkms: not work

[Julien Muchembled]

Control: tags -1 upstream

Hello,

Le 09/14/14 23:37, Eduardo a écrit :
 get_skb_tcpdata does not return a pointer to the actual  protocol string. 
 http://mike.it-loops.com/item/29
 The following patch fixed it for me, http://pastebin.com/JGw6QKVN

This is an upstream issue and I don't know this software well enough to help 
without more information.
The fact is nat-rtsp-dkms works for my use, so can you detail what does not 
work ? Step to reproduce, actual and expected results, from user point of view 
? I read http://mike.it-loops.com/item/29 and I still don't understand the 
issue.

And let's find a better title for this bug report.

Julien


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#760355: tuxonice-userui: FTBFS - missing include of stdio.h

[Julien Muchembled]

Control: tags -1 upstream
Control: tags 759856 upstream

This was already reported by https://bugs.debian.org/759856 and reassigned to 
libmng.
With the 'affects' BTS command, I expected that #759856 would still be visible 
on the tuxonice-userui PTS page.

Anyway, unless stdio.h is included by upstream libmng or libjpeg, upstream 
tuxonice-userui will want to include it.



signature.asc
Description: OpenPGP digital signature

Bug#759856: tuxonice-userui: FTBFS: jpeglib.h:954:30: error: unknown type name 'FILE'

[Julien Muchembled]

Control: retitle -1 stdio.h not included anymore before jpeglib.h (causes 
tuxonice-userui to FTBFS)
Control: reassign -1 libmng-dev
Control: affects -1 src:tuxonice-userui
Control: found -1 1.0.10+dfsg-3.1
Control: notfound -1 1.0.10-3

Well, I prefer to first reassign to libmng for the moment because libjpeg 
explicitly documents that stdio.h must be included before jpeglib.h, and it's 
like this for years.

But feel free to forward to libjpeg. I'm curious to know why libjpeg could not 
even do something like:
#ifndef FILE
#include stdio.h
#endif

For libmng, this is a recent regression. The include of stdio.h was removed 
by the backport of lcms2 support.

For more information, see also https://bugs.launchpad.net/bugs/1342459 which is 
now the same bug.



signature.asc
Description: OpenPGP digital signature

Bug#759856: tuxonice-userui: FTBFS: jpeglib.h:954:30: error: unknown type name 'FILE'

[Julien Muchembled]

Hello,

Le 08/30/14 21:05, Lucas Nussbaum a écrit :
 Relevant part (hopefully):
 cc -g -O2 -fstack-protector-strong -Wformat -Werror=format-security -Wall 
 -O3 -DUSE_FBSPLASH -Wall -O3 -g -I/usr/include/freetype2/ -I. -c 
 mng_callbacks.c -o mng_callbacks.o
 In file included from /usr/include/libmng_types.h:210:0,
  from /usr/include/libmng.h:386,
  from mng_callbacks.c:12:
 /usr/include/jpeglib.h:954:30: error: unknown type name 'FILE'
  EXTERN(void) jpeg_stdio_dest JPP((j_compress_ptr cinfo, FILE * outfile));
   ^
 /usr/include/jpeglib.h:955:29: error: unknown type name 'FILE'
  EXTERN(void) jpeg_stdio_src JPP((j_decompress_ptr cinfo, FILE * infile));
  ^
 make[3]: *** [mng_callbacks.o] Error 1

I confirm the issue after upgrading libmng-dev from 1.0.10-3 to 1.0.10+dfsg-3.1

I'm ready to add a patch so that stdio.h is included at the beginning 
mng_callbacks.c

But isn't it rather a bug in /usr/include/jpeglib.h if it does not include 
stdio.h before using FILE ?
I am tempted to reassign to both libjpeg-turbo  libjpeg8

Ah, just found https://bugs.debian.org/461602 and 
https://bugs.launchpad.net/bugs/1342459‎
So maybe reassign to libmng instead.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#746463: systemd: Add support for TuxOnIce hibernation

[Julien Muchembled]

Package: systemd
Version: 208-1
Severity: normal
Tags: patch upstream
Control: affects -1 tuxonice-userui

systemd 208+ tries to detect whether there's enough swap space to hibernate and
this is irrelevant when TuxOnIce is used. I've just upgraded systemd from
version 204 and hibernate key stopped working because of that. In other words,
this breaks package tuxonice-userui (which I maintain).

Could you please add the attached patch to experimental branch ?

I tried to push it upstream but there seems to be a hard policy to reject
anything that is not in vanilla Linux:
 http://lists.freedesktop.org/archives/systemd-devel/2014-April/018963.html
From fe5cf5ff434d1b7f636086c83cf3174f72459432 Mon Sep 17 00:00:00 2001
From: Julien Muchembled j...@jmuchemb.eu
Date: Tue, 29 Apr 2014 11:40:50 +0200
Subject: [PATCH] Add support for TuxOnIce hibernation

systemd does not support non-mainline kernel features so upstream rejected this
patch.
It is however required for systemd integration by tuxonice-userui package.

Forwarded: http://lists.freedesktop.org/archives/systemd-devel/2014-April/018960.html
---
 src/shared/sleep-config.c | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/src/shared/sleep-config.c b/src/shared/sleep-config.c
index d068bfc..ee74095 100644
--- a/src/shared/sleep-config.c
+++ b/src/shared/sleep-config.c
@@ -226,6 +226,12 @@ static bool enough_memory_for_hibernation(void) {
 size_t size, used;
 int r;
 
+/* TuxOnIce is an alternate implementation for hibernation.
+ * It can be configured to compress the image to a file or an inactive
+ * swap partition, so there's nothing more we can do here. */
+if (access(/sys/power/tuxonice, F_OK) == 0)
+return true;
+
 r = hibernation_partition_size(size, used);
 if (r  0)
 return false;
-- 
1.8.5.2.988.g9b015e5.dirty

Bug#746463: systemd: Add support for TuxOnIce hibernation

[Julien Muchembled]

Le 04/30/14 13:07, Michael Biebl a écrit :
 Am 30.04.2014 11:40, schrieb Julien Muchembled:
 
 systemd 208+ tries to detect whether there's enough swap space to hibernate 
 and
 this is irrelevant when TuxOnIce is used. I've just upgraded systemd from
 version 204 and hibernate key stopped working because of that. In other 
 words,
 this breaks package tuxonice-userui (which I maintain).

 Could you please add the attached patch to experimental branch ?

 I tried to push it upstream but there seems to be a hard policy to reject
 anything that is not in vanilla Linux:
  http://lists.freedesktop.org/archives/systemd-devel/2014-April/018963.html
 
 Personally I'm not too thrilled having a patch which is not going
 upstream anytime soon or even ever. We already have too many of them.
 That said, even if the popcon numbers of tuxonice are rather low, the
 patch looks simple enough.

This is kind of chicken-and-egg issue because the absence of such patch is 
another barrier for people who would like to use TuxOnIce.
TOI deserves better than this but there seems to be little interest by kernel 
developers for hibernation.
There was a serious attempt to merge it in 2009[1][2]
TOI is still maintained and I still hope its features will get merged in the 
future.

 If you are willing to keep the patch up-to-date for new upstream
 releases and handle bug reports related to this patch, I guess I'd be
 fine with merging it.

As a maintainer of tuxonice-userui, yes, of course.
But also because Linux and systemd are 2 components I always upgrade as soon as 
possible and I couldn't use my laptop without TOI.


[1] https://lkml.org/lkml/2009/5/6/249
[2] http://lists.tuxonice.net/pipermail/tuxonice-devel/2009-August/005537.html



signature.asc
Description: OpenPGP digital signature

Bug#746463: systemd: Add support for TuxOnIce hibernation

[Julien Muchembled]

Le 04/30/14 11:40, Julien Muchembled a écrit :
 systemd 208+ tries to detect whether there's enough swap space to hibernate 
 and
 this is irrelevant when TuxOnIce is used. I've just upgraded systemd from
 version 204 and hibernate key stopped working because of that. In other words,
 this breaks package tuxonice-userui (which I maintain).
 
 Could you please add the attached patch to experimental branch ?

Ah, I've just noticed that the systemd commits that check swap were backported 
yesterday to master (204-x).
So this patch for TOI should also go there.



signature.asc
Description: OpenPGP digital signature

Bug#746276: python-nemu: Fails to parse output from `tc qdisc show'

[Julien Muchembled]

Le 04/29/14 14:13, Martín Ferrari a écrit :
 On 28/04/14 18:43, Julien Muchembled wrote:
 

 After upgrading Linux from 3.13 to 3.14, the output of `tc qdisc show` for
 my wlan interface (driver iwlwifi) has changed on my laptop:
 
 regexp in nemu.iproute.get_tc_tree() function fails to parse these 4 new 
 lines,
 making the package unusable for me.
 
 Thanks for reporting this!
 
 I will install the new kernel and try to come up with a patch. Do you
 have a script to reproduce this, or did it just happen with any
 configuration?

python -c 'from nemu.iproute import get_tc_tree; get_tc_tree()'
is enough to trigger the error on my laptop.

It may be reproducible with specific hardware and I don't use Debian kernels.
I sent the output in my previous mail because I didn't expect you would try to 
reproduce.
Here, Nemu fails for an Intel Wireless-N 1000 and driver iwlwifi.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#746276: python-nemu: Fails to parse output from `tc qdisc show'

[Julien Muchembled]

Package: python-nemu
Version: 0.2-1
Severity: important
Tags: upstream

After upgrading Linux from 3.13 to 3.14, the output of `tc qdisc show` for
my wlan interface (driver iwlwifi) has changed on my laptop:

@@ -1 +1,5 @@
 qdisc mq 0: dev wifi root 
+qdisc pfifo_fast 0: dev wifi parent :1 bands 3 priomap  1 2 2 2 1 2 0 0 1 1 1 
1 1 1 1 1
+qdisc pfifo_fast 0: dev wifi parent :2 bands 3 priomap  1 2 2 2 1 2 0 0 1 1 1 
1 1 1 1 1
+qdisc pfifo_fast 0: dev wifi parent :3 bands 3 priomap  1 2 2 2 1 2 0 0 1 1 1 
1 1 1 1 1
+qdisc pfifo_fast 0: dev wifi parent :4 bands 3 priomap  1 2 2 2 1 2 0 0 1 1 1 
1 1 1 1 1

regexp in nemu.iproute.get_tc_tree() function fails to parse these 4 new lines,
making the package unusable for me.

-- System Information:
Debian Release: jessie/sid
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (300, 'experimental'), (200, 
'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.14.2+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages python-nemu depends on:
ii  bridge-utils1.5-7
ii  iproute 1:3.14.0-1
ii  procps  1:3.3.9-4
ii  python  2.7.5-5
ii  python-passfd   0.2-1
ii  python-unshare  0.1-2

python-nemu recommends no packages.

Versions of packages python-nemu suggests:
ii  xauth  1:1.0.7-1

-- no debconf information


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#739717: python-cliff: Please version dependencies

[Julien Muchembled]

Package: python-cliff
Version: 1.4.5-1
Severity: normal

Hello,

setup.py defines the following requirements:
  PrettyTable=0.6,0.8
  pyparsing=2.0.1
  cmd2=0.6.7

and I think binary packages should also have versionned dependencies.

python-cliff and its dependencies can be installed on Wheezy without any
rebuild provided cmd2 is updated to version =0.6.7. Otherwise:

  $ python -c 'import pkg_resources; pkg_resources.load_entry_point(cliff, 
cliff.formatter.list, table)'
  [...]
  pkg_resources.VersionConflict: (cmd2 0.6.4 
(/usr/lib/python2.7/dist-packages), Requirement.parse('cmd2=0.6.7'))

This is also useful on testing/sid machines that have an outdated version of
python-cmd2. I use sid on my laptop and I'd get mad if I always had to do a
complete upgrade before installing any package.

I searched how to do and it seems easiest to add the following files:

  $ cat debian/pydist-overrides 
  cmd2python-cmd2; PEP386
  PrettyTable python-prettytable; PEP386
  pyparsing   python-pyparsing; PEP386
  $ cat debian/py3dist-overrides 
  cmd2python3-cmd2; PEP386
  PrettyTable python3-prettytable; PEP386
  pyparsing   python3-pyparsing; PEP386

Like this, no need to edit debian/* if setup.py updates versions of these
dependencies.

I checked on pypi and all these packages seem to respect PEP386.

For python-cliff, this produces:
  python-prettytable (= 0.6), python-pyparsing (= 2.0.1), python-cmd2 (= 
0.6.7)

No idea why python-prettytable ( 0.8) is missing
but it's probably not worth working around this.

Regards,
Julien

-- System Information:
Debian Release: jessie/sid
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (300, 'experimental'), (200, 
'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.12.11+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#719529: O: tuxonice-userui -- user-space interfaces for TuxOnIce

[Julien Muchembled]

Le 12/22/13 22:50, Julien Muchembled a écrit :
 Le 08/12/13 22:14, Andrey Rahmatullin a écrit :
 There is one minor
 commit in its collab-maint repo which was not a part of an upload. 1.1-2~exp1
 should be reuploaded to unstable, as it was uploaded to experimental only
 because of the wheezy freeze.
 
 Yes. Then, I would certainly work on systemd integration.

I have finished to code what I wanted, and published it there:
 http://jmuchemb.eu/tuxonice-userui.git

Last changes are in a single catch-all commit (wip). I still have to split 
it, update changelog and add an entry in NEWS.Debian

Julien


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#719529: O: tuxonice-userui -- user-space interfaces for TuxOnIce

[Julien Muchembled]

Hello,

Le 08/12/13 22:14, Andrey Rahmatullin a écrit :
 For last year or two I was not able to use ToI (or often other hibernation
 implementations) on my main machine for various reasons (aka Linux 
 hibernation
 is still crap) so I am forced to orphan this package.

It's true that tuxonice was in quite a bad state more than 1 year, especially 
with XFS, but it's been several months now that I haven't got any crash.
Did you retry recently ?

In any case, I'd be happy to maintain this package. For me, STD is a very 
important feature.
I would be sponsored by previous maintainer Arnaud Fontaine.

 There is one minor
 commit in its collab-maint repo which was not a part of an upload. 1.1-2~exp1
 should be reuploaded to unstable, as it was uploaded to experimental only
 because of the wheezy freeze.

Yes. Then, I would certainly work on systemd integration.

Regards,
Julien


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#732948: TuxOnIce: fbsplash userui does not work and X server must be restart after resume

[Julien Muchembled]

Package: pm-utils
Version: 1.4.1-13
Severity: normal
Tags: patch upstream

Attached patch fixes detection of userui executable when fbsplash is chosen,
to make sure VT is changed. Otherwise, when current VT is manager by X, it
hibernates without any userui and after resume, VT are in such a bad state
that X server must be restarted.

-- System Information:
Debian Release: 7.0
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (300, 'experimental'), (200, 
'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.12.6+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages pm-utils depends on:
ii  powermgmt-base  1.31

Versions of packages pm-utils recommends:
ii  console-tools  1:0.2.3dbs-70
ii  ethtool1:3.4.2-1
ii  hdparm 9.43-1
ii  procps 1:3.3.4-2
pn  vbetoolnone

Versions of packages pm-utils suggests:
pn  cpufrequtilsnone
pn  radeontool  none
ii  wireless-tools  30~pre9-8

-- no debconf information
From: Julien Muchembled j...@jmuchemb.eu
Date: Sun, 22 Dec 2013 23:22:28 +0100
Subject: TuxOnIce: fix support for fbsplash userui

userui now ships a single executable that takes a argument to choose between
the default text interface and fbsplash. Both of them requires chvt to be done.

$ cat /etc/pm/config.d/tuxonice-userui
TUXONICE_USERUI_PROGRAM=/usr/lib/tuxonice-userui/tuxoniceui -f
---
 pm/module.d/tuxonice | 8 
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/pm/module.d/tuxonice b/pm/module.d/tuxonice
index 05f208e..f5bfb3b 100755
--- a/pm/module.d/tuxonice
+++ b/pm/module.d/tuxonice
@@ -13,10 +13,10 @@ done
 
 if [ -n $TUXONICE_LOC ]; then
 toi_maybe_chvt() {
-	local toi_ui=$(cat $TUXONICE_LOC/user_interface/program)
-	local toi_ui_en=$(cat $TUXONICE_LOC/user_interface/enabled)
-	if [ -x $toi_ui ]  [ $toi_ui_en = 1 ]  \
-	! state_exists console; then
+	local x y
+	read x y   $TUXONICE_LOC/user_interface/program
+	read y  $TUXONICE_LOC/user_interface/enabled
+	if [ -x $x -a $y = 1 ]  ! state_exists console; then
 	fgconsole |savestate console
 	chvt 63
 	fi

Bug#732533: btrfs-tools: New upstream release v3.12

[Julien Muchembled]

Due to upstream changes in Makefile, I couldn't build it Wheezy without adding 
the following lines to debian/rules:

# distclean target does not exist and make would waste time generating
# dependency files and version.h before failing. debhelper  9.20130624
# would even think distclean exists because something is printed.
override_dh_auto_clean:
make clean

Even if not required for testing/unstable, this even speeds up build slightly.
diff --git a/debian/changelog b/debian/changelog
index 8c6a5d6..a5c17e5 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,14 @@
+btrfs-tools (3.12-1) unstable; urgency=low
+
+  * New upstream release. (Closes: #732075)
+  * Drop patches merged upstream:
+- 07-manpage2.patch
+- 10-soname.patch
+- Btrfs-progs-fix-wrong-arg-sb_bytenr-for-btrfs_scan_fs_devices.patch
+  * Refresh patches. (Closes: #719072)
+
+ -- Julien Muchembled j...@jmuchemb.eu  Wed, 18 Dec 2013 11:51:13 +0100
+
 btrfs-tools (0.19+20130705-3) unstable; urgency=low
 
   * Import patch by Shilong Wang to resolve btrfs-convert (Closes:
diff --git a/debian/patches/02-ftbfs.patch b/debian/patches/02-ftbfs.patch
index 3b188b8..e52052d 100644
--- a/debian/patches/02-ftbfs.patch
+++ b/debian/patches/02-ftbfs.patch
@@ -8,7 +8,7 @@ Description:
 
 --- a/btrfs-convert.c
 +++ b/btrfs-convert.c
-@@ -2520,7 +2520,7 @@
+@@ -2444,7 +2444,7 @@ static int do_rollback(const char *devname, int force)
  	ext2_root = btrfs_read_fs_root(root-fs_info, key);
  	if (!ext2_root || IS_ERR(ext2_root)) {
  		fprintf(stderr, unable to open subvol %llu\n,
diff --git a/debian/patches/03-manpage.patch b/debian/patches/03-manpage.patch
index d8c1b25..e0983e3 100644
--- a/debian/patches/03-manpage.patch
+++ b/debian/patches/03-manpage.patch
@@ -5,7 +5,7 @@ Description:
 
 --- a/man/btrfs.8.in
 +++ b/man/btrfs.8.in
-@@ -497,7 +497,7 @@
+@@ -752,7 +752,7 @@ case of failure.
  
  .SH AVAILABILITY
  .B btrfs
@@ -14,20 +14,9 @@ Description:
  and not suitable for any uses other than benchmarking and review.
  Please refer to the btrfs wiki http://btrfs.wiki.kernel.org for
  further details.
 a/man/btrfsck.8.in
-+++ b/man/btrfsck.8.in
-@@ -8,7 +8,7 @@
- \fIdevice\fP is the device file where the filesystem is stored.
- .SH AVAILABILITY
- .B btrfsck
--is part of btrfs-progs. Btrfs is currently under heavy development,
-+is part of btrfs-tools. Btrfs is currently under heavy development,
- and not suitable for any uses other than benchmarking and review.
- Please refer to the btrfs wiki http://btrfs.wiki.kernel.org for
- further details.
 --- a/man/btrfs-image.8.in
 +++ b/man/btrfs-image.8.in
-@@ -44,7 +44,7 @@
+@@ -44,7 +44,7 @@ option if your extent tree is corrupted to make sure that all of the metadata is
  captured.
  .SH AVAILABILITY
  .B btrfs-image
@@ -38,7 +27,7 @@ Description:
  http://btrfs.wiki.kernel.org for further details.
 --- a/man/mkfs.btrfs.8.in
 +++ b/man/mkfs.btrfs.8.in
-@@ -84,7 +84,7 @@
+@@ -97,7 +97,7 @@ As default the unit is the byte, however it is possible to append a suffix
  to the arguments like \fBk\fP for KBytes, \fBm\fP for MBytes...
  .SH AVAILABILITY
  .B mkfs.btrfs
diff --git a/debian/patches/04-linker.patch b/debian/patches/04-linker.patch
index b3dc028..479402f 100644
--- a/debian/patches/04-linker.patch
+++ b/debian/patches/04-linker.patch
@@ -3,7 +3,7 @@ Description: Fixes FTBFS with --no-add-needed (Closes: #554059).
 
 --- a/Makefile
 +++ b/Makefile
-@@ -23,7 +23,7 @@
+@@ -24,7 +24,7 @@ TESTS = fsck-tests.sh
  INSTALL = install
  prefix ?= /usr/local
  bindir = $(prefix)/bin
diff --git a/debian/patches/07-manpage2.patch b/debian/patches/07-manpage2.patch
deleted file mode 100644
index 5c876bf..000
--- a/debian/patches/07-manpage2.patch
+++ /dev/null
@@ -1,34 +0,0 @@
-Author: Daniel Baumann daniel.baum...@progress-technologies.net
-Description: Update manpage to match /sbin/btrfs (Closes: #638778, #642302).
-
 a/man/btrfs.8.in
-+++ b/man/btrfs.8.in
-@@ -25,6 +25,8 @@
- [-s \fIstart\fR] [-t \fIsize\fR] -[vf] \fIfile\fR|\fIdir\fR \
- [\fIfile\fR|\fIdir\fR...]
- .PP
-+\fBbtrfs\fP \fBfilesystem df\fP\fI path \fP
-+.PP
- \fBbtrfs\fP \fBfilesystem sync\fP\fI path \fP
- .PP
- \fBbtrfs\fP \fBfilesystem resize\fP\fI [devid:][+/\-]size[gkm]|[devid:]max filesystem\fP
-@@ -33,7 +35,7 @@
- .PP
- \fBbtrfs\fP \fBfilesystem balance\fP\fI path \fP
- .PP
--\fBbtrfs\fP \fBdevice scan\fP\fI [--all-devices|device [device...]]\fP
-+\fBbtrfs\fP \fBfilesystem scan\fP\fI [--all-devices|device [device...]]\fP
- .PP
- \fBbtrfs\fP \fBdevice stats\fP [-z] {\fIpath\fP|\fIdevice\fP}
- .PP
-@@ -199,6 +201,10 @@
- Show information of a given subvolume in the \fIpath\fR.
- .TP
- 
-+\fBfilesystem df\fR\fI path\fR
-+Resize a filesystem identified by \fIpath\fR.
-+.TP
-+
- \fBfilesystem defragment\fP -c[zlib|lzo] [-l \fIlen\fR] [-s \fIstart\fR] \
- [-t \fIsize\fR] -[vf] \fIfile\fR|\fIdir\fR [\fIfile\fR|\fIdir\fR...]
- 
diff --git a/debian/patches/09-unaligned-memaccess.patch b/debian/patches/09

Bug#732533: btrfs-tools: New upstream release v3.12

[Julien Muchembled]

Package: btrfs-tools
Version: 0.19+20130705-3
Severity: wishlist
Tags: patch

Hello,

See announcement at http://thread.gmane.org/gmane.comp.file-systems.btrfs/30403

The attached patch updates Debian package to this new version 3.12
I think it is complete and it closes 2 bugs: #719072  #732075

Regards,
Julien

-- System Information:
Debian Release: 7.0
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (300, 'experimental'), (200, 
'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.12.5+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages btrfs-tools depends on:
ii  e2fslibs1.42.8-1
ii  libblkid1   2.20.1-5.3
ii  libc6   2.17-97
ii  libcomerr2  1.42.8-1
ii  liblzo2-2   2.06-1
ii  libuuid12.20.1-5.3
ii  zlib1g  1:1.2.8.dfsg-1

btrfs-tools recommends no packages.

btrfs-tools suggests no packages.

-- no debconf information
diff --git a/debian/changelog b/debian/changelog
index 8c6a5d6..a5c17e5 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,14 @@
+btrfs-tools (3.12-1) unstable; urgency=low
+
+  * New upstream release. (Closes: #732075)
+  * Drop patches merged upstream:
+- 07-manpage2.patch
+- 10-soname.patch
+- Btrfs-progs-fix-wrong-arg-sb_bytenr-for-btrfs_scan_fs_devices.patch
+  * Refresh patches. (Closes: #719072)
+
+ -- Julien Muchembled j...@jmuchemb.eu  Wed, 18 Dec 2013 11:51:13 +0100
+
 btrfs-tools (0.19+20130705-3) unstable; urgency=low
 
   * Import patch by Shilong Wang to resolve btrfs-convert (Closes:
diff --git a/debian/patches/02-ftbfs.patch b/debian/patches/02-ftbfs.patch
index 3b188b8..e52052d 100644
--- a/debian/patches/02-ftbfs.patch
+++ b/debian/patches/02-ftbfs.patch
@@ -8,7 +8,7 @@ Description:
 
 --- a/btrfs-convert.c
 +++ b/btrfs-convert.c
-@@ -2520,7 +2520,7 @@
+@@ -2444,7 +2444,7 @@ static int do_rollback(const char *devname, int force)
  	ext2_root = btrfs_read_fs_root(root-fs_info, key);
  	if (!ext2_root || IS_ERR(ext2_root)) {
  		fprintf(stderr, unable to open subvol %llu\n,
diff --git a/debian/patches/03-manpage.patch b/debian/patches/03-manpage.patch
index d8c1b25..e0983e3 100644
--- a/debian/patches/03-manpage.patch
+++ b/debian/patches/03-manpage.patch
@@ -5,7 +5,7 @@ Description:
 
 --- a/man/btrfs.8.in
 +++ b/man/btrfs.8.in
-@@ -497,7 +497,7 @@
+@@ -752,7 +752,7 @@ case of failure.
  
  .SH AVAILABILITY
  .B btrfs
@@ -14,20 +14,9 @@ Description:
  and not suitable for any uses other than benchmarking and review.
  Please refer to the btrfs wiki http://btrfs.wiki.kernel.org for
  further details.
 a/man/btrfsck.8.in
-+++ b/man/btrfsck.8.in
-@@ -8,7 +8,7 @@
- \fIdevice\fP is the device file where the filesystem is stored.
- .SH AVAILABILITY
- .B btrfsck
--is part of btrfs-progs. Btrfs is currently under heavy development,
-+is part of btrfs-tools. Btrfs is currently under heavy development,
- and not suitable for any uses other than benchmarking and review.
- Please refer to the btrfs wiki http://btrfs.wiki.kernel.org for
- further details.
 --- a/man/btrfs-image.8.in
 +++ b/man/btrfs-image.8.in
-@@ -44,7 +44,7 @@
+@@ -44,7 +44,7 @@ option if your extent tree is corrupted to make sure that all of the metadata is
  captured.
  .SH AVAILABILITY
  .B btrfs-image
@@ -38,7 +27,7 @@ Description:
  http://btrfs.wiki.kernel.org for further details.
 --- a/man/mkfs.btrfs.8.in
 +++ b/man/mkfs.btrfs.8.in
-@@ -84,7 +84,7 @@
+@@ -97,7 +97,7 @@ As default the unit is the byte, however it is possible to append a suffix
  to the arguments like \fBk\fP for KBytes, \fBm\fP for MBytes...
  .SH AVAILABILITY
  .B mkfs.btrfs
diff --git a/debian/patches/04-linker.patch b/debian/patches/04-linker.patch
index b3dc028..479402f 100644
--- a/debian/patches/04-linker.patch
+++ b/debian/patches/04-linker.patch
@@ -3,7 +3,7 @@ Description: Fixes FTBFS with --no-add-needed (Closes: #554059).
 
 --- a/Makefile
 +++ b/Makefile
-@@ -23,7 +23,7 @@
+@@ -24,7 +24,7 @@ TESTS = fsck-tests.sh
  INSTALL = install
  prefix ?= /usr/local
  bindir = $(prefix)/bin
diff --git a/debian/patches/07-manpage2.patch b/debian/patches/07-manpage2.patch
deleted file mode 100644
index 5c876bf..000
--- a/debian/patches/07-manpage2.patch
+++ /dev/null
@@ -1,34 +0,0 @@
-Author: Daniel Baumann daniel.baum...@progress-technologies.net
-Description: Update manpage to match /sbin/btrfs (Closes: #638778, #642302).
-
 a/man/btrfs.8.in
-+++ b/man/btrfs.8.in
-@@ -25,6 +25,8 @@
- [-s \fIstart\fR] [-t \fIsize\fR] -[vf] \fIfile\fR|\fIdir\fR \
- [\fIfile\fR|\fIdir\fR...]
- .PP
-+\fBbtrfs\fP \fBfilesystem df\fP\fI path \fP
-+.PP
- \fBbtrfs\fP \fBfilesystem sync\fP\fI path \fP
- .PP
- \fBbtrfs\fP \fBfilesystem resize\fP\fI [devid:][+/\-]size[gkm]|[devid:]max filesystem\fP
-@@ -33,7 +35,7 @@
- .PP
- \fBbtrfs\fP \fBfilesystem balance\fP\fI path \fP
- .PP
--\fBbtrfs\fP \fBdevice scan\fP\fI [--all-devices|device [device...]]\fP
-+\fBbtrfs\fP

Bug#732332: ITP: fssync -- File system synchronization tool (1-way, over SSH)

[Julien Muchembled]

Package: wnpp
Severity: wishlist
Owner: Julien Muchembled j...@jmuchemb.eu

* Package name: fssync
  Version : 1.0
  Upstream Author : Julien Muchembled j...@jmuchemb.eu
* URL : http://jmuchemb.eu/fssync.git
* License : GPL 3
  Programming Lang: Python
  Description : File system synchronization tool (1-way, over SSH)

fssync is a 1-way file-synchronization tool that tracks inodes and maintains a
local database of files that are on the remote side, making it able to:
- handle efficiently a huge number of dirs/files
- detect renames/moves and hard-links

It aims at minimizing network traffic and synchronizing every detail of a file
system:
- all types of inode: file, dir, block/character/fifo, socket, symlink
- preserve hard links
- modification time, ownership/permission/ACL, extended attributes
- sparse files

Other features:
- it can be configured to exclude files from synchronization
- fssync can be interrupted and resumed at any time, making it tolerant to
  random failures (e.g. network error)
- algorithm to synchronize file content is designed to handle big files
  like VM images efficiently, by updating fixed-size modified blocks in-place

Main usage of fssync is to prevent data loss in case of hardware failure,
where RAID1 is not possible (e.g. in laptops).

On Btrfs file systems, fssync is an useful alternative to `btrfs send` (and
`receive`) commands, thanks to filtering capabilities. This can be combined
with Btrfs snapshotting at destination side for a full backup solution.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#732026: ITP: nat-rtsp -- Connection tracking and NAT support for RTSP (DKMS)

[Julien Muchembled]

Package: wnpp
Severity: wishlist
Owner: Julien Muchembled j...@jmuchemb.eu

* Package name: nat-rtsp
  Version : 0.7+1.g2ea3cb6
  Upstream Author : Michael Guntsche m...@it-loops.com
* URL : https://github.com/maru-sama/rtsp-linux
* License : GPL-2.0+
  Programming Lang: C
  Description : Connection tracking and NAT support for RTSP

This package extends Linux Netfilter to provide connection tracking and NAT
support for RTSP. Module is built with DKMS.

Real Time Streaming Protocol is a network control protocol designed for use in
entertainment and communications systems to control streaming media servers.

For example in France, the modem provided by Free ISP (Freebox) uses this
protocol to broadcast TV/radio channels on the LAN. This module is required
for any router between the modem and end-users.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#725344: python-nemu: incompatible with recent /bin/ip

[Julien Muchembled]

Package: python-nemu
Version: 0.1-1
Severity: grave
Tags: patch upstream
Justification: renders package unusable

Nemu depends on /bin/ip to get addresses of interfaces:
  ip -o addr list

But recent iproute/iproute2 does not show link information anymore,
making Nemu fail with following traceback:

Traceback (most recent call last):
  File /usr/lib/python2.7/dist-packages/nemu/protocol.py, line 272, in run
cmd[0](cmd[1], *cmd[2])
  File /usr/lib/python2.7/dist-packages/nemu/protocol.py, line 480, in 
do_ADDR_ADD
nemu.iproute.add_addr(ifnr, a)
  File /usr/lib/python2.7/dist-packages/nemu/iproute.py, line 461, in add_addr
addresses = get_addr_data()[1][ifname]
  File /usr/lib/python2.7/dist-packages/nemu/iproute.py, line 440, in 
get_addr_data
bynam[name].append(_parse_ip_addr(match.group(4)))
KeyError: 'lo'


iproute 20120521 is the latest version that lists addresses in expected format. 
So this bug does not affect Wheezy.

Attached patch fixes parsing of /bin/ip output.

-- System Information:
Debian Release: 7.0
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (300, 'experimental'), (200, 
'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.10.10+ (SMP w/4 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages python-nemu depends on:
ii  bridge-utils1.5-6
ii  iproute 1:3.11.0-1
ii  procps  1:3.3.4-2
ii  python  2.7.3-5
ii  python-passfd   0.2-1
ii  python-unshare  0.1-2

python-nemu recommends no packages.

Versions of packages python-nemu suggests:
ii  xauth  1:1.0.7-1

-- no debconf information
--- src/nemu/iproute.py	2012-04-02 05:00:00.0 +0200
+++ src/nemu/iproute.py	2013-10-04 13:46:16.318263897 +0200
@@ -434,9 +434,10 @@
 raise RuntimeError(Invalid `ip' command output)
 idx = int(match.group(1))
 name = match.group(2)
-if match.group(3):
+if name not in bynam:
 bynam[name] = byidx[idx] = []
-continue # link info
+if match.group(3): # BBB: old iproute also shows link info
+continue
 bynam[name].append(_parse_ip_addr(match.group(4)))
 return byidx, bynam
 

Bug#556366: Can't install python-zc.lockfile and python-zc.buildout simultaneously

[Julien Muchembled]

Package: python-zc.lockfile
Version: 1.0.0-3
Severity: important

Hello,


Is there any reason why python-zc.lockfile conflicts with python-zc.buildout ?

I'd like to install python-zc.buildout and python-zc.lockfile, but they both
provide python-zc and python-zc.lockfile conflicts with python-zc.


Regards,
Julien

-- System Information:
Debian Release: squeeze/sid
  APT prefers unstable
  APT policy: (900, 'unstable'), (400, 'testing'), (200, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.31.5-1 (SMP w/2 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#484025: a Debian patch against python-reportlab may prevent to use python-xml and reportlab simultaneously

[Julien Muchembled]

Package: python-reportlab
Version: 2.1dfsg-2
Severity: important
Tags: patch

python-reportlab_2.1dfsg-2.diff.gz currently contains the following patch:
--- python-reportlab-2.1dfsg.orig/reportlab/lib/xmllib.py
+++ python-reportlab-2.1dfsg/reportlab/lib/xmllib.py
@@ -7,7 +7,8 @@
 import string
 
 try:
-import sgmlop   # this works for both builtin on the path or relative
+from _xmlplus.parsers import sgmlop
+#import sgmlop   # this works for both builtin on the path or relative
 except ImportError:
 sgmlop = None
 

This is a bad way to import sgmlop, as you can see:
 import xml.parsers
 from _xmlplus.parsers import sgmlop
 import xml.sax.writer
Traceback (most recent call last):
  File stdin, line 1, in ?
  File /usr/lib/python2.4/site-packages/_xmlplus/sax/writer.py, line 146, in ?
class _XMLDTDLoader(xml.parsers.xmlproc.xmlapp.DTDConsumer):
AttributeError: 'module' object has no attribute 'xmlproc'

For example, it breaks Zope completely if CMFReportTool product is installed.

sgmlop must be imported as follows:
diff -u python-reportlab-2.1dfsg/reportlab/lib/xmllib.py 
python-reportlab-2.1dfsg/reportlab/lib/xmllib.py
--- python-reportlab-2.1dfsg/reportlab/lib/xmllib.py
+++ python-reportlab-2.1dfsg/reportlab/lib/xmllib.py
@@ -7,7 +7,7 @@
 import string
 
 try:
-from _xmlplus.parsers import sgmlop
+from xml.parsers import sgmlop
 #import sgmlop   # this works for both builtin on the path or relative
 except ImportError:
 sgmlop = None


BTW, I don't understand how #468614 has been fixed, since both ways to import 
sgmlop fails without python-xml:

 from _xmlplus.parsers import sgmlop
Traceback (most recent call last):
  File stdin, line 1, in ?
ImportError: No module named _xmlplus.parsers

 from xml.parsers import sgmlop
Traceback (most recent call last):
  File stdin, line 1, in ?
ImportError: cannot import name sgmlop


Anyway, why not keeping sgmlop.so in python-reportlab-accel?
This would be the more simple solution:
 * drop python-xml
 * drop the patch to python-reportlab
 * drop the deletion of sgml.so in reportlab-accel

-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (990, 'testing'), (400, 'unstable'), (200, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.24.4-1 (SMP w/2 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages python-reportlab depends on:
ii  python2.5.2-1An interactive high-level object-o
ii  python-central0.6.1  register and build utility for Pyt

Versions of packages python-reportlab recommends:
ii  python-imaging1.1.6-1Python Imaging Library
pn  python-renderpm   none (no description available)
ii  ttf-bitstream-vera1.10-7 The Bitstream Vera family of free 
pn  ttf-dustinnone (no description available)

-- no debconf information



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#421772: xserver-xorg: EDID takes precedance over DisplaySize

[Julien Muchembled]

tag 421772 + patch
forwarded 421772 https://bugs.freedesktop.org/show_bug.cgi?id=9758
found 421772 2:1.4.1~git20080131-1
stop

The bug is still there.
I've just posted a patch upstream.
--- xorg-server-1.4.1~git20080131/hw/xfree86/modes/xf86EdidModes.c	2008-02-01 04:40:14.0 +0100
+++ xorg-server-1.4.1~git20080131/hw/xfree86/modes/xf86EdidModes.c	2008-02-28 12:16:00.725935688 +0100
@@ -407,8 +407,15 @@
 
 Monitor-DDC = DDC;
 
-Monitor-widthmm = 10 * DDC-features.hsize;
-Monitor-heightmm = 10 * DDC-features.vsize;
+/*
+ * Skip widthmm  heightmm because xf86SetDpi assumes they contain
+ * values specified in the config file. Setting them now would break the
+ * DisplaySize option, and xf86SetDpi would never write to the log that
+ * DPI settings have been probed.
+ *
+ * Monitor-widthmm = 10 * DDC-features.hsize;
+ * Monitor-heightmm = 10 * DDC-features.vsize;
+ */
 
 /* If this is a digital display, then we can use reduced blanking */
 if (DDC-features.input_type)

Bug#466186: FTBFS on amd64: missing library libldap_r-2.4.so.2 (for wldap32.dll.so)

[Julien Muchembled]

Ove Kaaven a écrit :
 but I still want to see your config.log to figure out
 what's wrong with the freetype in ia32-libs.
oh of course

build32/config.log attached


wine-0.9.54-cfg2.bz2
Description: Binary data

Bug#466186: FTBFS on amd64: missing library libldap_r-2.4.so.2 (for wldap32.dll.so)

[Julien Muchembled]

 Hmm, looks like that one could be a missing build dependency, which
 means there's at least *one* thing I might be able to do something
 about... So, try installing lib32z1-dev, what happens then?

(45min to build wine without -j2)

It's better:
 # distribute files into debian/packagename
 # according to the packagename.install files
 dh_install -s
 # patch marlett.ttf due to fontforge bug #458234
 mensis -script debian/marlett.mensis 
 debian/libwine/usr/share/wine/fonts/marlett.ttf
 ...

Next errors:
 Forcing extra dependencies...
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libcapi20.so when searching for 
 -lcapi20
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libcapi20.a when searching for 
 -lcapi20
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libcapi20.so when 
 searching for -lcapi20
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libcapi20.a when searching 
 for -lcapi20
 /usr/bin/ld: skipping incompatible /usr/lib/libcapi20.so when searching for 
 -lcapi20
 /usr/bin/ld: skipping incompatible /usr/lib/libcapi20.a when searching for 
 -lcapi20
 /usr/bin/ld: cannot find -lcapi20
 collect2: ld returned 1 exit status
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libpng.so when searching for 
 -lpng
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libpng.a when searching for -lpng
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libpng.so when searching 
 for -lpng
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libpng.a when searching 
 for -lpng
 /usr/bin/ld: skipping incompatible /usr/lib/libpng.so when searching for -lpng
 /usr/bin/ld: skipping incompatible /usr/lib/libpng.a when searching for -lpng
 /usr/bin/ld: cannot find -lpng
 collect2: ld returned 1 exit status
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libjack.so when searching for 
 -ljack
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libjack.a when searching for 
 -ljack
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libjack.so when searching 
 for -ljack
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libjack.a when searching 
 for -ljack
 /usr/bin/ld: skipping incompatible /usr/lib/libjack.so when searching for 
 -ljack
 /usr/bin/ld: skipping incompatible /usr/lib/libjack.a when searching for 
 -ljack
 /usr/bin/ld: cannot find -ljack
 collect2: ld returned 1 exit status
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libcups.so when searching for 
 -lcups
 /usr/bin/ld: skipping incompatible 
 /usr/lib/gcc/x86_64-linux-gnu/4.2.3/../../../libcups.a when searching for 
 -lcups
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libcups.so when searching 
 for -lcups
 /usr/bin/ld: skipping incompatible /usr/bin/../lib/libcups.a when searching 
 for -lcups
 /usr/bin/ld: skipping incompatible /usr/lib/libcups.so when searching for 
 -lcups
 /usr/bin/ld: skipping incompatible /usr/lib/libcups.a when searching for 
 -lcups
 /usr/bin/ld: cannot find -lcups
 collect2: ld returned 1 exit status

And a fatal one:
 dpkg-shlibdeps: failure: no dependency information found for 
 /emul/ia32-linux/usr/lib/libsane.so.1 (used by 
 debian/libwine-sane/extradep32).



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#466186: FTBFS on amd64: missing library libldap_r-2.4.so.2 (for wldap32.dll.so)

[Julien Muchembled]

 So, when you fixed the shlibs issue in ia32-libs, did you also fix it
 for libsane?
No. Only libxml2, as described by #456914.

 dpkg-shlibdeps: failure: no dependency information found for 
 /usr/lib32/libxml2.so.2 (used by debian/libwine/usr/lib/wine/msxml3.dll.so).
 dpkg-shlibdeps: failure: no dependency information found for 
 /emul/ia32-linux/usr/lib/libsane.so.1 (used by 
 debian/libwine-sane/extradep32).
Hmmm... I didn't notice that's the same kind of error.
However, I wonder why the second one doesn't happen if I build from
amd64.tar.lzma.uu

I guess I should add :
 libsane 1 ia32-libs (= 1.6)
to ia32-libs.shlibs.

rebuilding...




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#466186: FTBFS on amd64: missing library libldap_r-2.4.so.2 (for wldap32.dll.so)

[Julien Muchembled]

Package: libwine-ldap
Version: 0.9.54-1
Severity: serious
Justification: no longer builds from source

I have libldap-2.4-2 (2.4.7-5) installed but I still get the following error:
dpkg-shlibdeps: failure: couldn't find library libldap_r-2.4.so.2 needed by 
debian/libwine-ldap/usr/lib/wine/wldap32.dll.so (its RPATH is '').

I suppose the problem is in ia32-libs.

-- System Information:
Debian Release: lenny/sid
  APT prefers unstable
  APT policy: (990, 'unstable'), (200, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.24.2-1 (SMP w/2 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#466186: FTBFS on amd64: missing library libldap_r-2.4.so.2 (for wldap32.dll.so)

[Julien Muchembled]

 If it said No shlibs information, then *that* would be a known bug in
 ia32-libs, and known to be the current reason for Wine to FTBFS on the
 amd64 build daemons.
(I'm not sure I understand what you say.)
There isn't any amd64 build daemon for Wine at this time: Dep-Wait:
ia32-libs ( 2.2) because of libxml2 (#458013  #456914).
On my PC, I edited /var/lib/dpkg/info/ia32-libs.shlibs manually in order
to build wine myself.

 But your error looks more like a problem that dpkg-shlibdeps used to
 only treat as a warning, not an error. Does the build actually stop
 there, and not because of some other error?
Yes it stops there.

Since I don't need libwine-ldap :
1. I added a 'bash -i' line before and after:
   dh_shlibdeps -s -Llibwine -ldebian/libwine/usr/lib'
2. before dh_shlibdeps, I did:
   $ mv debian/libwine-ldap/usr/lib/wine/wldap32.dll.so ..
3. after:
   $ mv ../wldap32.dll.so debian/libwine-ldap/usr/lib/wine/

(Finally, I got a libwine-ldap_0.9.54-1_amd64.deb package without any
dependency.)

 If so, what if you delete debian/amd64.tar.lzma.uu from the unpacked
 sources before building? That would build a Wine that only uses the
 32-bit libraries that are actually available in ia32-libs (but might
 still fail due to No shlibs information...)
Oops. Not tried, and I should redownload the source package to test.
I've already upgraded Wine to 0.9.54-1.
Tell me if this test would be useful to you.



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#457557: wine: FTBFS on amd64

[Julien Muchembled]

block 457557 by 456914
found 457557 0.9.51-1
stop

Hello,

The missing dependency to gcc-multilib is not the only problem.

dh_shlibdeps fails for 2 reasons :

The 1st one might be related to #453885 : I could work around the issue
by adding /emul/ia32-linux/lib and /emul/ia32-linux/usr/lib to the -l
flag. It produced many warnings.

The 2nd one is already reported at #456914 :
ia32-libs: Missing shlibs entry for libxml2

JM




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#421407: checkinstall: Can't install python programs if filesystem translation is enabled.

[Julien Muchembled]

Package: checkinstall
Version: 1.6.1-2
Severity: normal

This is due to the os.listdir function. Installation scripts usually use it
whereas this function doesn't work if the --fstrans=no isn't passed.
An OSError exception is raised, reporting there is no such file or directory:

$ checkinstall -d2 --nodoc --maintainer jm -si --install=no -y \
 python -c import os; print os.listdir('/')

checkinstall 1.6.1, Copyright 2002 Felipe Eduardo Sanchez Diaz Duran
   This software is released under the GNU GPL.

debug: TAR=tar
debug: VISUAL=

debug: Setting umask = 0022
debug: The temporary directory is: [ /var/tmp/paQmaJaHTjPkIapATYDaT ]


*
 Debian package creation selected ***
*

This package will be built according to these values:

0 -  Maintainer: [ jm ]
1 -  Summary: [ Package created with checkinstall 1.6.1 ]
2 -  Name:[ checkinstall ]
3 -  Version: [ 20070428 ]
4 -  Release: [ 1 ]
5 -  License: [ GPL ]
6 -  Group:   [ checkinstall ]
7 -  Architecture: [ amd64 ]
8 -  Source location: [ checkinstall ]
9 -  Alternate source location: [  ]
10 - Requires: [  ]

Enter a number to change any of them or press ENTER to continue:
debug: CK_INCLUDE_FILE =

Installing with python -c import os; print os.listdir('/')...

= Installation results 
===
debug: INSTW_EXCLUD before sort 
=/tmp/jm/checkinstall,/dev,/proc,/tmp,/var/tmp
debug: INSTW_EXCLUDE=/dev,/proc,/tmp,/tmp/jm/checkinstall,/var/tmp,
debug: INSTW_ROOTPATH=/var/tmp/paQmaJaHTjPkIapATYDaT
debug: INSTW_LOGFILE=/var/tmp/paQmaJaHTjPkIapATYDaT/newfiles.tmp
debug: INSTW_DBGFILE=/var/tmp/paQmaJaHTjPkIapATYDaT/dbgfile
debug: INSTW_DBGLVL=2
Traceback (most recent call last):
  File string, line 1, in ?
OSError: [Errno 2] No such file or directory: '/'

  Installation failed. Aborting package creation.

Cleaning up...OK

Bye.

-- System Information:
Debian Release: lenny/sid
  APT prefers unstable
  APT policy: (990, 'unstable'), (200, 'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.20.4-1 (SMP w/2 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages checkinstall depends on:
ii  file  4.20-4 Determines file type using magic
ii  findutils 4.2.30-1   utilities for finding files--find,
ii  libc6 2.5-4  GNU C Library: Shared libraries

checkinstall recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]