Bug#338938: kaffe: LANG environement variable doesn't set default charset

2005-11-13 Thread Jean-Michel Vourgère
Package: kaffe Version: 2:1.1.6-3 Severity: normal Tags: patch I have LANG=en_US.UTF-8 and java does print: GetEnv.getEnv(LANG) = en_US.UTF-8 however default charset is incorrect new java.io.OutputStreamWriter(System.out)).getEncoding() = ISO8859_1 Workaround: Add -Dfile.encoding=UTF-8 in JVM

Bug#631865: alsa-utils: underrun!!! (at least 881183351.502 ms long)

2011-06-27 Thread Jean-Michel Vourgère
Subject: alsa-utils: underrun!!! (at least 881183351.502 ms long) Package: alsa-utils Version: 1.0.23-3 Severity: minor I'm using arecord|aplay to play audio from an usb device with that command: arecord --buffer-size=256 -r 44000 -D iec958:CARD=default,DEV=0 -f S16_LE | aplay --buffer-size=256

Bug#632735: dh_installcron man page should underline words that should be replaced by a value

2011-07-05 Thread Jean-Michel Vourgère
Subject: debhelper: dh_installcron man page should underline words that should be replaced by a value Package: debhelper Version: 8.0.0 Severity: minor Tags: patch *** Please type your report below this line *** I think it is a convention for man that words that are to be replaced by a value

Bug#644095: jwchat sends invalid body xmlns='' with iceweasel 6

2011-10-02 Thread Jean-Michel Vourgère
Subject: jwchat sends invalid body xmlns='' with iceweasel 6 Package: jwchat Version: 1.0+dfsg-1 Severity: normal *** Please type your report below this line *** Using iceweasel 6, messages cannot be received by a pidgin user. Messages appear in pidgin XMPP console, but tyhey are discarded as

Bug#644097: 404 not found error when DEBUG=true

2011-10-02 Thread Jean-Michel Vourgère
Subject: jwchat: 404 not found error when DEBUG=true Package: jwchat Version: 1.0+dfsg-1 Severity: minor Tag: patch *** Please type your report below this line *** If you set DEBUG=true in /etc/jwchat/config.js, you get a 404 in debugger window. You need to add these lines to debian/links:

Bug#644133: lintian: dpkg 1.16.1 transition: warns about existing CFLAG in debian/rules

2011-10-03 Thread Jean-Michel Vourgère
Package: lintian Version: 2.5.3 Severity: wishlist Some packages used to start debian/rules with things like: CFLAGS=-Wall -Werror=format-security However recent dpkg updates[1] not only make these pratices useless, but it also removes usefull flags such as -fstack-protector in that example.

Bug#675543: Unicode support for mdbtools

2012-06-29 Thread Jean-Michel Vourgère
severity 675543 normal thanks Thank you for your good patch. However, on Debian, UTF-8 is already working well, and UCS is usually not needed. So I am decreasing the severity to normal. Your patch will be merged in next upstream release. signature.asc Description: This is a digitally signed

Bug#680021: Please add support for Access 2010 files

2012-07-02 Thread Jean-Michel Vourgère
Standard ACE DB rather than Standard JET DB at byte 4. Author: Jean-Michel Vourgère jmv_...@nirgal.com Bug-Debian: http://bugs.debian.org/bugnumber Forwarded: no Reviewed-By: name and email of someone who approved the patch Last-Update: 2012-06-03 --- file-5.11.orig/magic/Magdir/database +++ file

Bug#680021: Please add support for Access 2010 files

2012-07-03 Thread Jean-Michel Vourgère
Judit Foglszinger wrote: Could you provide a sample .accdb file? Unfortunaly, I don't have that version of access. But you can download these kind of files from many places. I tested one at http://www.rogersaccesslibrary.com/forum/uploads/5/RollingAverage07.zip signature.asc Description:

Bug#680852: bash-completion: dh_bash-completion processes only one package if proper snippet

2012-07-08 Thread Jean-Michel Vourgère
Package: bash-completion Version: 1:2.0-1 Severity: normal My package mdbtools provides 2 bash-completion snippet. Only one get installed. I suppose this might be caused by a intall+exit(0) on line 83 after detection that there is not a list of file in the package.bash-completion file. But there

Bug#680879: ITP: eyefiserver -- Daemon to receive images from EyeFi SD card

2012-07-08 Thread Jean-Michel Vourgère
Package: wnpp Severity: wishlist Owner: Jean-Michel Vourgère jmv_...@nirgal.com * Package name: eyefiserver Version : 0.1 Upstream Author : Jeff Tchang someb...@example.org * URL : https://github.com/tachang/EyeFiServer/ * License : GPL3+ Programming Lang

Bug#680852: Patch

2012-07-09 Thread Jean-Michel Vourgère
tags 680852 +patch thanks Attached is a patch. I'm not familiar with perl: This works for me, but please review carefully. Description: Don't stop processing packages after proper snippet Author: Jean-Michel Vourgère jmv_...@nirgal.com Forwarded: not-needed Last-Update: 2012-01-04 --- debian

Bug#681565: clive: Please show result filename

2012-07-14 Thread Jean-Michel Vourgère
Package: clive Version: 2.3.3-2 Severity: normal It's usually difficult to know which file was just downloaded, because there is no feedback on that. Please dump to stdout the name of the saved file. Exemple: nirgal@ऊ:~$ clive 'http://www.youtube.com/watch?v=icZ1rDJgDSY' Checking ...done. %

Bug#681585: lmarbles: Please package lastest upstream version

2012-07-14 Thread Jean-Michel Vourgère
Package: lmarbles Version: 1.0.7-1.1 Severity: wishlist All is in the subject Attached is a debian/watch file to help you track releases. -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (500, 'testing') Architecture: i386 (i686) Kernel: Linux

Bug#671178: clive: Can't locate object method err via package clive::Util::Log

2012-05-02 Thread Jean-Michel Vourgère
Package: clive Version: 2.2.27-1 clive doesn't seem to work with that url: $ clive --format=mp4 'http://www.dailymotion.com/video/xqj6kz_greenpeace-survole-l-usine-nucleaire-de-bugey_news' fetch http://www.dailymotion.com/video/xqj6kz_greenpeace-survole-l-usine-nucleaire-de-bugey_news

Bug#671182: /usr/bin/python: double free or corruption

2012-05-02 Thread Jean-Michel Vourgère
Package: reportbug Version: 6.3.1 Severity: important I tried to report a bug in clive, but reportbug, invoked from the command line, crashed after saying a new version was available in sid, and I decided to submit anyways: nirgal@ऊ:~$ reportbugg *** glibc detected *** /usr/bin/python: double

Bug#610121: Using dpkg-statoverride

2011-10-23 Thread Jean-Michel Vourgère
Actually, some people pointed out chmod/chown is a bad practice in postinst. One should use dpkg-statoverride. So attached is another proposal. diff -Naur lmarbles-1.0.7.keep//debian/postinst lmarbles-1.0.7/debian/postinst --- lmarbles-1.0.7.keep//debian/postinst 2011-01-15 14:37:04.0

Bug#690142: Non english ordering of bind9_9.8.1.dfsg.P1-4.2.diff.gz ?

2012-10-25 Thread Jean-Michel Vourgère
Hello Nico I'm trying to apply a patch to bind9 (see bug #690142). But the diff.gz file I get is very different from previous version, because files are not treated in the same order: $ zgrep ^--- bind9_9.8.1.dfsg.P1-4.2.diff | head -3 --- bind9-9.8.1.dfsg.P1.orig/config.sub ---

Bug#690142: Non english ordering of bind9_9.8.1.dfsg.P1-4.2.diff.gz ?

2012-10-25 Thread Jean-Michel Vourgère
Ops, sorry. Forget it. Neither version is ordered. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#690924: CVE-2012-4527

2012-11-01 Thread Jean-Michel Vourgère
Second patch listed at redhat is ok. But I would either replace #define WIDTH 80 by #define WIDTH (sizeof(tmperr)) or #define WIDTH 80 char tmperr[128]; by #define WIDTH 128 char tmperr[WIDTH]; snprintf does add a \0 at tmperr[WIDTH] WIDTH is a poor choice as a macro name IMHO. signature.asc

Bug#690924: CVE-2012-4527

2012-11-01 Thread Jean-Michel Vourgère
On Thursday 01 November 2012 14:04:55 Jean-Michel Vourgère wrote: snprintf does add a \0 at tmperr[WIDTH] Doh! I mean tmperr[WIDTH-1] of course! -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#690924: CVE-2012-4527

2012-11-01 Thread Jean-Michel Vourgère
-only bug on wheezy. Author: Attila Bogar, Jean-Michel Vourgère jmv_...@nirgal.com Origin: https://bugzilla.redhat.com/show_bug.cgi?id=CVE-2012-4527 Bug: CVE-2012-4527 Bug-Debian: http://bugs.debian.org/690924 Forwarded: no Last-Update: 2012-11-01 Index: mcrypt-2.6.8/src/mcrypt.c

Bug#692065: RFS: mcrypt/2.6.8-1.3 [RC][NMU]

2012-11-01 Thread Jean-Michel Vourgère
/pool/main/m/mcrypt/mcrypt_2.6.8-1.3.dsc Changes since the last upload: mcrypt (2.6.8-1.3) unstable; urgency=medium . * Non-maintainer upload. * CVE-2012-4527: stack-based buffer overflow by encryption / decryption of overly long file names (Closes: #690924) Regards, Jean-Michel

Bug#692065: RFS: mcrypt/2.6.8-1.3 [RC][NMU]

2012-11-01 Thread Jean-Michel Vourgère
* CVE-2012-4527: stack-based buffer overflow by encryption / decryption of overly long file names (Closes: #690924) I've reviewed this and it looks mostly good. However, can you explain why you chose ERRWIDTH=PATH_MAX+1024 vs. the redhat patch WIDTH=80? I don't know exactly. It

Bug#692065: RFS: mcrypt/2.6.8-1.3 [RC][NMU]

2012-11-02 Thread Jean-Michel Vourgère
On Friday 02 November 2012 17:58:08 you wrote: (...) Again, determining the right solution would be best to discuss on the redhat bug or preferably with upstream. Ok. I did post that on redhat bugzilla. upstream did not publish anything for years... Until those things happen, I am going to

Bug#689715: Review of eyefiserver-2.3~rc1-1

2012-10-11 Thread Jean-Michel Vourgère
Salut On Wednesday 10 October 2012 19:59:11 Vincent Fourmond wrote: * I don't see the point of including session dumps - especially 4MB of them ! Absolutely. Done. I did not change the orig tarball however. Right? Hmmm... I'm unsure about that, since the copyright status of the dumps

Bug#689715: Review of eyefiserver

2012-10-18 Thread Jean-Michel Vourgère
Salut! On Tuesday 16 October 2012 07:01:11 Vincent Fourmond wrote: On Wednesday 10 October 2012 19:59:11 Vincent Fourmond wrote: * I don't see the point of including session dumps - especially 4MB of them ! Absolutely. Done. I did not change the orig tarball however. Right?

Bug#690843: libapache2-mod-dnssd: Uninstall yields prerm: line 12: a2dismod: command not found

2012-10-18 Thread Jean-Michel Vourgère
Package: libapache2-mod-dnssd Severity: normal When uninstalling libapache2-mod-dnssd, I get this error: Removing libapache2-mod-dnssd ... /var/lib/dpkg/info/libapache2-mod-dnssd.prerm: line 12: a2dismod: command not found Purging configuration files for libapache2-mod-dnssd ... I suppose you

Bug#656455: jwchat: unowned files after purge (policy 6.8, 10.8)

2012-10-18 Thread Jean-Michel Vourgère
tags 656455 + patch thank you Hello Log shows ucf is purged before jwchat. postrm purge tests -x /usr/bin/ucf before rm these files. It should not test that and do it unconditionnaly. Trivial patch is attached. --- debian/postrm.orig 2012-10-18 19:32:47.0 +0200 +++ debian/postrm

Bug#635279: jwchat: please rename/symlink .da files to use default apache2 .dk extension

2012-10-18 Thread Jean-Michel Vourgère
tags 635279 + moreinfo thank you jwchat does not contains any .da nor .dk file. Can you be more precise? If you give us the full name of a file, maybe we can find which package you are talking about... signature.asc Description: This is a digitally signed message part.

Bug#663971: apache2.2-common: /etc/init.d/apache2 requires facility $named, causing failed install/upgrade

2012-10-19 Thread Jean-Michel Vourgère
It seems to me this is exactly what the little known Should-Start: from LSB is for. https://wiki.debian.org/LSBInitScripts signature.asc Description: This is a digitally signed message part.

Bug#288615: apache2: Issue with language negotiation exceptions

2012-10-20 Thread Jean-Michel Vourgère
tags 288615 + patch thank you I could reproduce the problem. The proposed patch fixes it. :) signature.asc Description: This is a digitally signed message part.

Bug#288615: Patch file

2012-10-21 Thread Jean-Michel Vourgère
: Richard Atterer atte...@debian.org Origin: http://bugs.debian.org/288615 Bug: https://issues.apache.org/bugzilla/show_bug.cgi?id=54033 Bug-Debian: http://bugs.debian.org/288615 Reviewed-By: Jean-Michel Vourgère jmv_...@nirgal.com Last-Update: 2006-11-25 Index: apache2-2.2.22/modules/mappers

Bug#691319: pgstatus: maintainer names with underscores

2012-10-24 Thread Jean-Michel Vourgère
Package: buildd.debian.org Hello Buildd status for packages I maintain is not working. https://buildd.debian.org/status/package.php?p=jmv_deb%40nirgal.comcomaint=yes Page says Buildd status for packages maintained by jmv I expected page to say Buildd status for packages maintained by

Bug#691365: apache2: invoke-rc.d apache2 status fails

2012-10-24 Thread Jean-Michel Vourgère
Package: apache2 Version: 2.4.3-1 Severity: normal Tags: patch After a complex and wrong upgrade to latest debian git snapshot of apache2, I ended up with old 2.2 apache2-mpm-worker in broken state. I purge it and had to kill -9 the processes. http://paste.debian.net/203339/ Anyways, after that

Bug#691306: iptables add 4 rules instead just one in kernel INPUT chain

2012-10-24 Thread Jean-Michel Vourgère
Hint: If you comment out ::1 localhost in /etc/hosts, the problem disappears. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#691306: iptables add 4 rules instead just one in kernel INPUT chain

2012-10-24 Thread Jean-Michel Vourgère
I traced that behavior to host_to_ipaddr function in libxtables/xtables.c --source hostname will match all the IP addresses of hostname. Same thing for --destination The problem is that gethostbyname(localhost) returns two results: 127.0.0.1 and 127.0.0.1 (yes, the same thing). So this might

Bug#691306: Bug #691306: iptables add 4 rules instead just one in kernel INPUT chain

2012-10-24 Thread Jean-Michel Vourgère
Hi libc folks, iptables --insert --source localhost ... is currently creating duplicate entries. I believe this is because of a problem in gethostbyname(localhost) that returns duplicate answers. Attached is a tiny test case. ./main 127.0.0.1 120.0.0.1 If I comment out ::1 localhost in my

Bug#680879: Repository at collab-maint

2012-08-23 Thread Jean-Michel Vourgère
Developement version is available at: http://anonscm.debian.org/gitweb/?p=collab-maint/eyefiserver.git signature.asc Description: This is a digitally signed message part.

Bug#693372: ubuntu-dev-tools: requestsync fails ValueError: IV must be 16 bytes long

2012-11-15 Thread Jean-Michel Vourgère
Package: ubuntu-dev-tools Version: 0.143 Severity: normal When running $ requestsync -d unstable -ns eyefiserver A browser opens and I can login in an Ubuntu website then I enter a password 3 times then I get a python ValueError exception with a backtrace. $ requestsync -d unstable -ns

Bug#694014: debian-maintainers: Please add Jean-Michel Vourgère to Debian Maintainers keyring

2012-11-22 Thread Jean-Michel Vourgère
-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Comment: Add Jean-Michel Vourgère jmv_...@nirgal.com as a Debian Maintainer Date: Fri, 23 Nov 2012 01:52:10 +0100 Action: import Recommended-By: Vincent Fourmond fourm...@debian.org Agreement: https://lists.debian.org/debian-newmaint/2012/10

Bug#683955: gnome-sudoku: Note textentry hide data

2012-08-05 Thread Jean-Michel Vourgère
Package: gnome-sudoku Version: 1:3.4.2-2 Severity: important Like many persons, I use the note (shortcut key N) to list possible choices in a given cell. However, that function now opens an entry that hide other cells, making it impossible to know what to write in the note! The only way to use

Bug#684981: autoconf: make install leaves ld cache outdated

2012-08-15 Thread Jean-Michel Vourgère
Package: autoconf Version: 2.69-1 Severity: normal This issue was originally reported at https://github.com/brianb/mdbtools/issues/12 mdbtools build several libraries that get installed in /usr/local/lib make install rules generated by configure generated by autoconf runs ldconfig -n

Bug#500558: Can't listen to 127.0.0.1 if eth0 has IPv6 only

2012-02-17 Thread Jean-Michel Vourgère
reassign 500558 libapr1 tags 500558 + upstream thanks I traced back that problem to a call in libapr1, in call_resolver() getaddrinfo(127.0.0.1, NULL, hints, ...) that returns -9 because hints contains AI_ADDRCONFIG Without that hint, it works. Extract from call_resolver: if (family ==

Bug#646208: BTS Pending tags

2012-02-17 Thread Jean-Michel Vourgère
Hello I've been sorting bugs. Count is now bellow 100, but that still is a long list of things to watch. So I plan to tag pending the ones fixed in the next git branch of Alioth, without fixed version since 2.4.1 will be officially out in a couple of days, and I'm not sure what next version will

Bug#629899: ap_get_local_host is broken

2012-02-18 Thread Jean-Michel Vourgère
Hello I recently found some strange behavior in dns resolution at boot time, when eth0 has no ipv4 yet. (#500558) Do you use DHCP for your other interface (eth0)? What is the result of `grep CONCURRENCY /etc/init.d/rc' ? Does the bug is still there if you set up a static ip address in

Bug#639408: python2.X/3.X-dev ships windows binaries which are not built at build time

2012-02-19 Thread Jean-Michel Vourgère
which are not built at build time (Closes: 639408). + + -- Jean-Michel Vourgère jmv_...@nirgal.com Sat, 18 Feb 2012 19:07:18 +0100 + python3.2 (3.2.2-4) unstable; urgency=low * The static library belongs into the -dev package. diff -Naur python3.2-3.2.2/debian/orig-tar.sh python3.2-3.2.2+dfsg

Bug#639407: python2.X/3.X-dev ships windows binaries which are not built at build time

2012-02-19 Thread Jean-Michel Vourgère
which are not built at build time (Closes: +639407). + + -- Jean-Michel Vourgère jmv_...@nirgal.com Sun, 19 Feb 2012 15:00:50 +0100 + python2.7 (2.7.2-8) unstable; urgency=low * Update to 2030, taken from the 2.7 branch. diff -Naur python2.7-2.7.2/debian/orig-tar.sh python2.7-2.7.2+dfsg

Bug#629899: ap_get_local_host is broken

2012-02-19 Thread Jean-Michel Vourgère
tags 629899 + moreinfo thanks On Sunday 19 February 2012 11:22:42 Vincent Lefevre wrote: Does the bug is still there if you set up a static ip address in /etc/network/interfaces for your eth0 interface? I haven't tried. In any case, there's no reason to use eth0 (some machines do not even

Bug#208831: apache2 - mod_action forgets authenticated user

2012-02-19 Thread Jean-Michel Vourgère
tags 208831 + unreproducible + moreinfo thanks Hello Bastian I tried to reproduce that 9 year old bug, but could not. My settings: AddType nirgal/test .nirgal Action nirgal/test /cgi-bin/echo Location / AuthType Basic AuthName By

Bug#632884: apache2 upgrade does not upgrade libapr1

2012-02-19 Thread Jean-Michel Vourgère
This is very likely related to fixed bug 591286. I almost merged it. However, the report is about a missing dependency. 2.2.29 can only be testing or newer. And nowodays, testing depends on libapr1 1.4.2. Since 591286 was fixed on libapr1 1.4.2-4 I suggest we force Depends: libapr1 = 1.4.2-4

Bug#604647: a2ensite should handle default-ssl specially

2012-02-20 Thread Jean-Michel Vourgère
Hello Actually, there is a significant difference between default and default-ssl: default defines a VirtualHost on *:80 default-ssl defines a VirtualHost on _default_:443 If I understand correctly https://httpd.apache.org/docs/2.4/en/vhosts/examples.html#default it means default-ssl has no

Bug#660479: getaddrinfo(127.0.0.1, AI_ADDRCONFIG) fails when eth0 has ipv6 only

2012-02-21 Thread Jean-Michel Vourgère
I confirm RFC 2553 specifically exclude loopback interface for AI_ADDRCONFIG. Attached is a trivial patch. --- man3/getaddrinfo.3.orig 2012-02-21 09:38:54.0 +0100 +++ man3/getaddrinfo.3 2012-02-21 09:39:42.0 +0100 @@ -378,6 +378,8 @@ only if the local system has at least one

Bug#500558: apache2.2-common: Apache fails to start on boot after upgrade Etch - Lenny

2012-02-21 Thread Jean-Michel Vourgère
severity 500558 important found 500558 1.4.2-6+squeeze3 found 500558 1.4.5-1.1 tags 500558 + patch thanks Reverting severity level after return from eglibc Patch posted upstream at https://issues.apache.org/bugzilla/attachment.cgi?id=28360action=diff signature.asc Description: This is a

Bug#655435: libapr1: apr_hash vulnerable to oCert-2011-003 style DOS attacks

2012-02-21 Thread Jean-Michel Vourgère
Fixed in 1.4.6 If we need to backport in the future, the changes are located in only two files: tables/apr_hash.c test/testhash.c Attached patch was taken from upstream trunk, it can also apply to previous versions, including 1.4.2-6+squeeze3 --- tables/apr_hash.c 2012/01/17 06:45:02 1232320 +++

Bug#629899: ap_get_local_host is broken (can't always determine the server's FQDN)

2012-02-21 Thread Jean-Michel Vourgère
One can reproduce that bug by having IPv6 only on eth0 and no other interface with IPv4 but localhost, and /etc/hosts matching `hostname` on any lo address. Call (forward) trace: ap_get_local_host apr_sockaddr_info_get find_addresses call_resolver getaddrinfo ... that fails since it is using

Bug#629899: ap_get_local_host is broken (can't always determine the server's FQDN)

2012-02-22 Thread Jean-Michel Vourgère
Hello Vincent On Wednesday 22 February 2012 03:39:38 Vincent Lefevre wrote: I'll try to do some tests when I have the time. But there's no reason eth0 would be IPv6 only here (unless there's a huge bug in the kernel). I'm prety sure your apache tries to start before dhclient has given the

Bug#594166: apxs2 generated Makefile command fails to restart apache

2012-02-24 Thread Jean-Michel Vourgère
found 594166 2.2.16-6+squeeze6 found 594166 2.2.22-1 thanks Actually, you also need to load the module to reproduce: echo LoadModule foo_module /usr/lib/apache2/modules/mod_foo.so /etc/apache2/mods-available/foo.load a2enmod foo error.log repports: [Fri Feb 24 22:56:34 2012] [notice] SIGHUP

Bug#606141: apache2ctl configtest and apache2ctl graceful gives incorrect warning about a (non existent) Alias in VirtualHost

2012-02-24 Thread Jean-Michel Vourgère
tags 606141 +moreinfo thanks Hello What is the content of your /etc/apache2/conf.d/bugconfig.conf file? What aliases do you have on your server? You can get these with: find /etc/apache2/ -exec grep -H Alias {} \; signature.asc Description: This is a digitally signed message part.

Bug#659499: bash fails to properly read /proc files

2012-02-25 Thread Jean-Michel Vourgère
Thank you Cyril for the mini-test. I can reproduce this bug with linux-image-3.2.0-1-686-pae If I reboot on linux-image-3.1.0-1-686-pae lseeking in /proc works fine. signature.asc Description: This is a digitally signed message part.

Bug#650267: watch file

2012-02-26 Thread Jean-Michel Vourgère
Attached is a debian/watch file for uscan. Hope you like it. watch Description: application/fluid signature.asc Description: This is a digitally signed message part.

Bug#631865: patch

2012-02-15 Thread Jean-Michel Vourgère
: Jean-Michel Vourgère jmv_...@nirgal.com Bug-Debian: http://bugs.debian.org/631865 Forwarded: https://bugtrack.alsa-project.org/alsa-bug/view.php?id=5536 Reviewed-By: name and email of someone who approved the patch Last-Update: 2012-01-15 Index: alsa-utils-1.0.24.2/aplay/aplay.c

Bug#402567: apache2: BRF files support

2012-02-16 Thread Jean-Michel Vourgère
tags 402567 - upstream + patch thanks Upstream does not ship with the AddCharset directives. They are Debian specific from source file debian/config-dir/mods-available/mime.conf. We just need to add that line to it: AddCharset BRF .brf signature.asc Description: This is a digitally signed

Bug#564722: htdigest -c should bail if the target digest file exists

2012-02-03 Thread Jean-Michel Vourgère
man page says: -c Create the passwdfile. If passwdfile already exists, it is deleted first. And from older version, one can guess file was effectively truncated: fopen(w). https://svn.apache.org/viewvc/httpd/httpd/trunk/support/htdigest.c?r1=85063r2=85064 (That was 11

Bug#444048: We won't get details

2012-02-05 Thread Jean-Michel Vourgère
Bug repporter email address is invalid. SMTP error from remote mail server after RCPT TO:npr...@q-cat.com: host mail.q-cat.com [64.247.176.135]: 550 5.1.1 sorry, no mailbox here by that name (chkuser) Original bug repport uses: CustomLog /home/nick/access_log combined ScriptLog

Bug#645874: We won't get details

2012-02-05 Thread Jean-Michel Vourgère
Bug repporter email address is invalid. SMTP error from remote mail server after RCPT TO:j.ku...@leonit.nl: host mail.leonit.nl [93.186.180.174]: 550 sorry, no mailbox here by that name. (#5.7.17) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of

Bug#630888: ssl renegotiation hangs with long ca lists

2012-02-05 Thread Jean-Michel Vourgère
Please, can you provide the information required at https://issues.apache.org/bugzilla/show_bug.cgi?id=46952#c24 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#631230: apache2 startup fails with missing log directory

2012-02-05 Thread Jean-Michel Vourgère
While I agree there is a problem with missing log directory repporting - see https://issues.apache.org/bugzilla/show_bug.cgi?id=29941 - I don't think creating this directory automatically would be wise. For exemple, if your directory is not mounted yet, you'll end up writing in / partition and

Bug#653060: KeyError: 'could not find foreign type Surface'

2011-12-23 Thread Jean-Michel Vourgère
Package: gnome-sudoku Version: 1:3.0.2-2 Severity: normal gnome-sudoku does not start anymore. Here's the console output: CRITICAL:root:set_bg_color handed Bad color: black Traceback (most recent call last): File /usr/lib/python2.7/dist-packages/gnome_sudoku/gsudoku.py, line 59, in

Bug#653060: KeyError: 'could not find foreign type Surface'

2011-12-23 Thread Jean-Michel Vourgère
On Friday 23 December 2011 10:58:02 Josselin Mouette wrote: Does it work if you install python-gi-cairo? Yes it does. Thank you. :) signature.asc Description: This is a digitally signed message part.

Bug#669837: RFS: phppgadmin_5.1-1.1 [NMU]

2013-07-22 Thread Jean-Michel Vourgère
, dh_apache2 does it + automatically. +. Build-depends on dh-apache2. +. Use {$misc:Suggests} so apache2 is no longer a direct Depends. + + -- Jean-Michel Vourgère jmv_...@nirgal.com Mon, 22 Jul 2013 13:01:15 + + phppgadmin (5.1-1) unstable; urgency=low * New upstream release. diff

Bug#710688: jwchat: fails to install with apache 2.4

2013-07-22 Thread Jean-Michel Vourgère
Control: tags -1 + patch Dear jwchat maintainers Attached is a patch for your package. Regards diff -Nru jwchat-1.0+dfsg/debian/postinst jwchat-1.0+dfsg/debian/postinst --- jwchat-1.0+dfsg/debian/postinst 2010-09-17 18:11:48.0 + +++ jwchat-1.0+dfsg/debian/postinst 2013-07-22

Bug#669837: phppgadmin: transition towards Apache 2.4

2013-07-22 Thread Jean-Michel Vourgère
Control: tags -1 - moreinfo On Monday 22 July 2013 17:46:12 Christoph Berg wrote: We need a source package that is also compatible with squeeze and lucid so it is backportable just by rebuilding - I believe the new require local directive doesn't work there and needs at least Version magic or

Bug#717621: lintian: apache2 AllowOverride matches obsolete Allow directive check

2013-07-22 Thread Jean-Michel Vourgère
Package: lintian Version: 2.5.14 Severity: normal Dear Maintainer, I'm working on phppgadmin package that include a apache2.4 config file containing: AllowOverride None lintian is repporting apache2-deprecated-auth-config Allow I believe that AllowOverride directive is ok, and not

Bug#717621: lintian: apache2 AllowOverride matches obsolete Allow directive check

2013-07-23 Thread Jean-Michel Vourgère
Control: tags -1 +patch Attached is a patch: It will check there is a white space type character, or a new line, after. It will also check the directive is a the begining of a line, and will now ignore comments and the name of a directive in a text, a url, and so on. Note that all this space

Bug#669837: phppgadmin: transition towards Apache 2.4

2013-07-24 Thread Jean-Michel Vourgère
We need a source package that is also compatible with squeeze and lucid so it is backportable just by rebuilding - I believe the new require local directive doesn't work there and needs at least Version magic or similar treatment. The rest of the changes is probably ok if we also leave the

Bug#718166: a2disconf fails on purge when conf already gone

2013-07-28 Thread Jean-Michel Vourgère
I spent some time on that issue, and could not find a reasonable aswner: I noticed that dh_apache2 generates snipets with: apache2_invoke enconf phppgadmin || exit $? Adding the || exit $? really does help (I could not believe it at first). This may somehow be linked to the fact that

Bug#717476: updates lost Alias modules

2013-07-28 Thread Jean-Michel Vourgère
This is probably a duplicate of bugs.debian.org/cgi-bin/bugreport.cgi?bug=711925 Can you check if you auto-purge packages on updates? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#718249: javascript-common: postinst fails trying to remove non-existent apache conf.d

2013-07-29 Thread Jean-Michel Vourgère
-Michel Vourgère jmv_...@nirgal.com Mon, 29 Jul 2013 10:16:39 + + javascript-common (10) unstable; urgency=low * Downgrading webservers relation from 'recommends' to 'suggests' diff -Nru javascript-common-10/debian/javascript-common.postinst javascript-common-11/debian/javascript

Bug#710412: mod_access_compat apparently not working

2013-07-29 Thread Jean-Michel Vourgère
Control: tags -1 + patch As of 2.4.6-1, authn_core is enabled by default, so your problem should now be fixed in most cases. See bugs.debian.org/cgi-bin/bugreport.cgi?bug=717496 Nevertheless, the dependency should be properly registered: ---

Bug#711454: libapache2-mod-php5: MPM prefork not found ... == ... Syntax error on line 9 of /etc/apache2/mods-enabled/php5.conf: Invalid command 'Order'

2013-07-29 Thread Jean-Michel Vourgère
reassign 711454 libapache2-mod-php5 found 711454 5.5.0~rc2+dfsg-2 thanks The problem is that apache2 is not configured before libapache2-mod-php5. apache2_switch_mpm prefork fails, because conffiles from apache2 are not yet there, including mods-available/mpm_prefork.load. Is a Pre-Depends

Bug#711493: No default site enabled after fresh (re)install

2013-07-29 Thread Jean-Michel Vourgère
Control: found -1 2.4.6-2 Looks like apache2.preinst is always creating /etc/apache2/.apache2_mpm_selected because if [ $? -eq 0 ] on line 109 is always true. This results in we_are_upgrading_from_wheezy in apache2.postinst to succeed where it shouln't. Attached is a log of apache2 maintainer

Bug#718318: [Pkg-javascript-devel] Bug#718318: javascript-common does not install

2013-07-30 Thread Jean-Michel Vourgère
Apparently, your system is having a dangling symlink /etc/apache2/conf-available/javascript-common.conf - /etc/javascript-common/javascript-common.conf that prevents the installation of package javascript-common. javascript-common has been broken for a while in sid, since apache2.4 is out. So

Bug#684050: SuppressHTMLPreamble also discards data in the directory listing

2013-07-30 Thread Jean-Michel Vourgère
Control: tags -1 + unreproducible Hello I tried to reproduce your problem on apache 2.2.22, but I could not. Are you sure you did not forget the + ? Using IndexOptions SuppressHTMLPreamble gives excatly the problem you repported. But IndexOptions +SuppressHTMLPreamble works as expected. --

Bug#698787: apache2-mpm-worker multiarch install problem on Debian Wheezy amd64

2013-07-30 Thread Jean-Michel Vourgère
reassign 698787 ssl-cert retitle 698787 ssl-cert: Please mark Multi-Arch: foreign thanks It looks to me that this is more a problem with mime-support and ssl-cert packages that were not flaged as supporting multi-arch. [1] mime-support package was fixed in version 3.53. [2] So I'm reassigning

Bug#718387: apache2-data: Please mark Multi-Arch: foreign

2013-07-30 Thread Jean-Michel Vourgère
Package: apache2-data Version: 2.4.6-2 Dear Maintainer, Please mark apache2-data as Multi-Arch: foreign This is preventing the cross-installation of apache2. Step to reproduce from an i386 arch: # dpkg --add-architecture amd64 # apt-get update # apt-get install apache2:amd64 Reading package

Bug#518370: libapache2-mod-macro: please enable macro module during install

2013-07-31 Thread Jean-Michel Vourgère
Hello This package does not enable the module when it is installed libapache2-mod-macro is no longer part of jessie. It will be reintroduced very soon, but only as a transitionnal package. mod_macro is now part of core apache modules, as of version 2.4. This also breaks other packages that

Bug#718475: ocsinventory-server: transition towards Apache 2.4

2013-08-01 Thread Jean-Michel Vourgère
Package: ocsinventory-server Severity: important User: debian-apa...@lists.debian.org Usertags: apache24webapptransition Control: block 661958 by -1 Dear maintainer, your package ocsinventory-server is a web application which supports the Apache2 web server. We're upgrading Apache to the new

Bug#718483: simpleid: transition towards Apache 2.4

2013-08-01 Thread Jean-Michel Vourgère
Package: simpleid Severity: important User: debian-apa...@lists.debian.org Usertags: apache24webapptransition Dear maintainer, your package simpleid is a web application which supports the Apache2 web server. We're upgrading Apache to the new upstream version 2.4 [1] (tracked as transition bug

Bug#718465: mdbtools: Don't build-depend on rarian-compat

2013-08-01 Thread Jean-Michel Vourgère
Control: tags -1 + pending Hi Jeremy rarian-compat is obsolete and unmaintained; please don't build-depend on it. I did that in git. http://anonscm.debian.org/gitweb/?p=collab-maint/mdbtools.git Harmless errors about scrollkeeper are ignored. The whole gmdb2 gnome help makefiles needs to be

Bug#710934: apache2: please include a way to automatically move conf.d files to conf-available

2013-08-02 Thread Jean-Michel Vourgère
To ease the transition to apache 2.4, could you have a script that automatically copies files in /etc/apache2/conf.d to /etc/apache2/conf-available? I advise caution if this is implemented, because it might break unrelated software. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=718318 is

Bug#711493: No default site enabled after fresh install if /etc/apache2 exists

2013-08-03 Thread Jean-Michel Vourgère
Control: tags -1 + patch Attached is a patch for that problem. I believe it's important because there is a lot of packages that does not depend on apache2, meaning they can be installed first, but that will install some conffiles in /etc/apache/*-available/ as requested. Note that cut will eat

Bug#714341: apache2: Disabled Sites Still Served

2013-08-03 Thread Jean-Michel Vourgère
Hello In my experience, that happens when I reload the wrong httpd, getting confused by the VMs. Maybe that happened to you... Can you please: - Check that the link has been removed from /etc/apache2/sites-enabled/ ? In version 2.2, a2dissite is just deleting the link there. - Did you try a

Bug#296493: apache2ctl -t should warn about non-initial #

2013-08-04 Thread Jean-Michel Vourgère
Kevin Bullock wrote: (...) I therefore suggest that 'apache2ctl -t' warn the user about the mid-line comment which could be an error. (...) Stefan Fritsch wrote: The documentation states: Comments may not be included on a line after a configuration directive. '#' signs are considered

Bug#670973: w3c-markup-validator: transition towards Apache 2.4

2013-08-05 Thread Jean-Michel Vourgère
Control: tags -1 + patch Attached is a patch for your package that fails to install on jessie. diff -Nru w3c-markup-validator-1.3+dfsg/debian/postinst w3c-markup-validator-1.3+dfsg/debian/postinst --- w3c-markup-validator-1.3+dfsg/debian/postinst 2012-10-04 00:48:48.0 +0200 +++

Bug#719265: apache2: Cleanup after a2_invoke enable third party mod|conf|site

2013-08-09 Thread Jean-Michel Vourgère
Package: apache2 Version: 2.4.6-2 Sometimes a package needs to enable a third party module. For example, jwchat enables module proxy_http on postinst. According to policy, a package should clean after itself. But right now, there is no way to do that: For example, it would be wrong for jwchat

Bug#719266: apache2_invoke enconf ignores dependencies disabled by admin

2013-08-09 Thread Jean-Michel Vourgère
Package: apache2 Version: 2.4.6-2 If sysop disabled a module, then a maintainer script enable a module that depends on the first one, they are both enabled, ignoring admin choice. Example (from a fresh wheezy install upgraded to jessie): # a2enmod proxy Enabling module proxy. # a2dismod proxy

Bug#719280: Error/warn: Config variable .... is not defined - Invalid Mutex directory in argument file:${APACHE_LOCK_DIR}

2013-08-10 Thread Jean-Michel Vourgère
Hello André It looks like you did not define required variables APACHE_LOCK_DIR, APACHE_PID_FILE and so on. This is exactly what envvar file is for. You *need* to source it before runing apache. apache2 is a service. On Debian, you need to start services using service(8) on the command line: #

Bug#719280: Error/warn: Config variable .... is not defined - Invalid Mutex directory in argument file:${APACHE_LOCK_DIR}

2013-08-10 Thread Jean-Michel Vourgère
Control: severity -1 normal Control: tags -1 +moreinfo +unreproducible Please do read the documentation, many of your questions are explained there. If service apache2 start doesn't work, please provide error messages. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with

Bug#719280: Error/warn: Config variable .... is not defined - Invalid Mutex directory in argument file:${APACHE_LOCK_DIR}

2013-08-10 Thread Jean-Michel Vourgère
On Saturday 10 August 2013 17:54:31 Andre Verwijs wrote: service apache2 start doesn't work because of above reported bug..!! The errors you are talking about are what you get when you enter apache2 on the command line. apache2 is not supposed to work without the proper environement being set up

  1   2   3   4   >