Bug#340245: xmms-goom: [INTL:sv] Swedish PO-template translation
On Tue, 2005-11-22 at 01:53 +0100, Daniel Nylander wrote: > Here is the swedish translation of xmms-goom > Thank you for the translation. The binary you translated isn't included in Debian's xmms-goom package but I'll forward your work to the upstream author. John signature.asc Description: This is a digitally signed message part
Bug#340070: yate: FTBFS: compile error: invalid conversion
tags 340070 + patch thanks pri_set_error and pri_set_message in libpri.h from libpri-dev have changed since this version of yate was released. Updating yate to a newer upstream would likely fix the FTBFS error. Barring that, the attached dpatch gets the yate package compiling again in its current form. I assume the changes will make no difference in the functioning on yate, but I don't use the software and have not tested it. John zapchan.dpatch Description: application/shellscript
Bug#336840: space-orbit: doesn't start
tags 336840 + patch
thanks
Reordering the call to glutInit in orbit.c fixes this error.
John
diff -ur space-orbit-1.01.orig/src/orbit.c space-orbit-1.01/src/orbit.c
--- space-orbit-1.01.orig/src/orbit.c 2005-12-13 18:22:35.0 -0600
+++ space-orbit-1.01/src/orbit.c 2005-12-13 18:22:22.0 -0600
@@ -41,6 +41,9 @@
/* Set up the player viewpoint, etc */
InitPlayer();
+ /* glutInit is required for InitStuff() */
+ glutInit(&argc, argv);
+
/* Initialize all sorts of other stuff */
InitStuff();
@@ -63,8 +66,10 @@
*/
{
char *p;
-
- glutInit (&argc, argv);
+
+ /* This is being called in main() now */
+ /* glutInit (&argc, argv); */
+
glutInitDisplayMode (GLUT_RGBA | GLUT_DOUBLE | GLUT_DEPTH);
glutInitWindowPosition (0, 0);
Bug#306488: libdv: DEB_BUILD_OPTIONS=noopt ignored
Package: libdv Severity: minor In debian/rules you need to either export the CFLAGS after setting them or include the CFLAGS on the configure/make lines. As is, compiling with noopt will result in -O2 binaries. Thanks John -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: powerpc (ppc) Kernel: Linux 2.6.10-powerpc Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#308669: kino: ffmpeg export scripts should tag video as DivX
Package: kino Version: 0.75-7 Severity: wishlist The ffmpeg divx export scripts export a valid mpeg4 file but tag this file as "FMP4". This causes problems with some free software video players which don't understand that "FMP4" is functionally identical to "DIVX". test.avi is tagged as "FMP4" [EMAIL PROTECTED]:~$ wxvlc kino/test.avi VLC media player 0.8.2-svn Janus libhal.c 767 : org.freedesktop.DBus.Error.ServiceDoesNotExist raised "Service "org.freedesktop.Hal" does not exist" [0245] main playlist: adding playlist item `kino/test.avi' ( kino/test.avi )[0264] main decoder error: no suitable decoder module for fourcc `FMP4'. VLC probably does not support this sound or video format. [0245] main playlist: deleting playlist item `kino/test.avi' test5.avi is tagged as "DIVX" [EMAIL PROTECTED]:~$ wxvlc kino/test5.avi VLC media player 0.8.2-svn Janus libhal.c 767 : org.freedesktop.DBus.Error.ServiceDoesNotExist raised "Service "org.freedesktop.Hal" does not exist" [0245] main playlist: adding playlist item `kino/test5.avi' ( kino/test5.avi ) [0245] main playlist: deleting playlist item `kino/test5.avi' The fix for this problem is a trivial change to the export scripts. "2" ) ffmpeg -f dv -i - -f avi -deinterlace -s 320x240 -b 500 -acodec "$acodec" -ab 64 -y "$file".avi ;; Should be changed to: "2" ) ffmpeg -f dv -i - -f avi -deinterlace -s 320x240 -b 500 -acodec "$acodec" -ab 64 -vtag divx -y "$file".avi ;; The only change to the output with "-vtag divx" is the "FMP4" tag. [EMAIL PROTECTED]:~$ ls -al kino/test*.avi -rw-r--r-- 1 john john 1941318 2005-05-11 14:47 kino/test5.avi -rw-r--r-- 1 john john 1941318 2005-05-11 14:32 kino/test.avi [EMAIL PROTECTED]:~$ xdelta delta kino/test.avi kino/test5.avi diffs [EMAIL PROTECTED]:~$ ls -al diffs -rw-r--r-- 1 john john 217 2005-05-11 15:11 diffs Some discussion about this issue in the Kino forums: http://kino.schirmacher.de/dcforum/dcforum?az=show_topic&forum=101&topic_id=3194&mesg_id=3194&page= Thanks for your excellent packaging work on Kino. John -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kernel: Linux 2.6.11-1-k7 Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1) Versions of packages kino depends on: ii libart-2.0-2 2.3.17-1Library of functions for 2D graphi ii libatk1.0-0 1.8.0-4 The ATK accessibility toolkit ii libavc1394-0 0.4.1-2 control IEEE1394 audio/video devic ii libbonobo2-0 2.8.1-2 Bonobo CORBA interfaces library ii libbonoboui2-0 2.8.1-2 The Bonobo UI library ii libc62.3.2.ds1-21GNU C Library: Shared libraries an ii libdv4 0.103-2 software library for DV format dig ii libgcc1 1:3.4.3-13 GCC support library ii libgconf2-4 2.8.1-6 GNOME configuration database syste ii libglade2-0 1:2.4.2-2 library to load .glade files at ru ii libglib1.2 1.2.10-10 The GLib library of C routines ii libglib2.0-0 2.6.4-1 The GLib library of C routines ii libgnome2-0 2.8.1-2 The GNOME 2 library - runtime file ii libgnomecanvas2-02.8.0-1 A powerful object-oriented display ii libgnomeui-0 2.8.1-3 The GNOME 2 libraries (User Interf ii libgnomevfs2-0 2.8.4-3 The GNOME virtual file-system libr ii libgtk2.0-0 2.6.4-2 The GTK+ graphical user interface ii libice6 4.3.0.dfsg.1-12.0.1 Inter-Client Exchange library ii liborbit21:2.12.2-1 libraries for ORBit2 - a CORBA ORB ii libpango1.0-01.8.1-1 Layout and rendering of internatio ii libpopt0 1.7-5 lib for parsing cmdline parameters ii libquicktime10.9.3-2 A library for reading and writing ii libraw1394-5 0.10.1-1.1 library for direct access to IEEE ii libsamplerate0 0.1.1-2 audio rate conversion library ii libsm6 4.3.0.dfsg.1-12.0.1 X Window System Session Management ii libstdc++5 1:3.3.6-4 The GNU Standard C++ Library v3 ii libx11-6 4.3.0.dfsg.1-12.0.1 X Window System protocol client li ii libxext6 4.3.0.dfsg.1-12.0.1 X Window System miscellaneous exte ii libxml2 2.6.16-7GNOME XML library ii libxv1 4.3.0.dfsg.1-12.0.1 X Window System video extension li ii xlibs4.3.0.dfsg.1-12 X Keyboard Extension (XKB) configu ii zlib1g 1:1.2.2-4 compression library - runtime -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#293550: xmms-infinity: new upstream version available
On Fri, 2005-02-04 at 11:11 +0100, Martin Schuster wrote: > Package: xmms-infinity > Version: 0.5.9-1 > Severity: normal > > New upstream version is available (0.6.0) > http://freshmeat.net/projects/infinity-plugin/?branch_id=50068&release_id=178677 > My understanding was that the 0.6.0 release was mostly the portability fixes I applied to 0.5.9 and forwarded upstream. I'll take a look at the upstream version though and see if I was mistaken. I should have time to check this weekend, probably Sunday. John signature.asc Description: This is a digitally signed message part
Bug#285274: libvisual
retitle 285274 RFP: libvisual -- abstraction library for visualization plugins thanks > Is there any progress on this ITP? I'd really like to see libvisual > in Debian. I haven't made any progress with it and I'd assume that Frederik hasn't either. Feel free to take over the ITP if you'd like. John signature.asc Description: This is a digitally signed message part
Bug#285274: libvisual
retitle 285274 ITP: libvisual -- abstraction library for visualization plugins thanks On Fri, 2005-05-06 at 20:25 +0200, Frederik Dannemare wrote: > I have actually been doing packages since 0.1.7, since I need it for > LiVES which I'm also maintaining (also not yet in Debian). > All I need is a sponsor for them. Sorry about that. I figured someone would have sponsored these packages by now. I use Kino and Cinelerra very often and I already maintain several packages that need libvisual, so I guess it would make sense for me to sponsor libvisual and lives. You'll have to give me some slack though. I haven't sponsored anyone before and I'll be fairly busy for the next few weeks. John signature.asc Description: This is a digitally signed message part
Bug#285274: libvisual
On Fri, 2005-05-06 at 21:50 +0200, Frederik Dannemare wrote: > On Friday 06 May 2005 21:07, John Lightsey wrote: > > You'll have to give me some slack though. I haven't sponsored anyone > > before and I'll be fairly busy for the next few weeks. > > No problem. I think we should actually wait until Sarge is out to not > stress the buildd with packages that will not make it into Sarge > anyways. > > Maybe I could ping you in 4-5 weeks or so to hear if your workload has > lowered a bit by then, if that's okay with you? That also leaves me > some time to write a little man page and do a few other minor tweaks > for LiVES, before it will be fully compliant with Debian Policy. Sounds good to me. John signature.asc Description: This is a digitally signed message part
Bug#293550: xmms-infinity: new upstream version available
On Fri, 2005-02-04 at 18:40 +0100, Martin Schuster wrote: > On Fri, Feb 04, 2005 at 08:15:52AM -0600, John Lightsey wrote: > > My understanding was that the 0.6.0 release was mostly the portability > > fixes I applied to 0.5.9 and forwarded upstream. I'll take a look at > > the upstream version though and see if I was mistaken. > > > I didn't check the differences :) just wanted to tell you about the new > version. Sorry for taking so long to get back to you on this bug report. I've had some time to package and test infinity 0.6.0 and talk with the upstream author, Duilio Protti. 0.6.0 does have some feature improvements. The most noticeable is that the config dialog has been redone to include a FPS adjustment. Unfortunately, 0.6.0 also has some multithreading problems. I can get it to crash consistently by resizing the visualization window. Duilio is aware of the problems with 0.6.0 and is in the process of working on new release. Once the new version is out, I'll package it up and try again. John signature.asc Description: This is a digitally signed message part
Bug#325609: gnome-games: aisleriot instant win
Package: gnome-games Version: 1:2.10.1-5 Severity: normal Play a game of AisleRiot until you get the "Congratulations You have won!!!" dialog. >From this point you can intentionally or accidentally win another game instantly. Click the close button on the dialog. Click one of the open spaces so the dialog reappears. Repeat until you get bored. Check your game statistics and you'll find that each time the dialog reappears you are credited with another win. Thanks John -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.8-11-amd64-generic Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1) Versions of packages gnome-games depends on: ii gnome-games-data 1:2.10.1-5 data files for the GNOME games ii guile-1.6-libs1.6.7-1Main Guile libraries ii libart-2.0-2 2.3.17-1 Library of functions for 2D graphi ii libc6 2.3.5-4GNU C Library: Shared libraries an ii libesd0 0.2.36-1 Enlightened Sound Daemon - Shared ii libgcc1 1:4.0.1-6 GCC support library ii libgconf2-4 2.10.1-1 GNOME configuration database syste ii libglade2-0 1:2.5.1-2 library to load .glade files at ru ii libglib2.0-0 2.8.0-1The GLib library of C routines ii libgnome2-0 2.10.1-1 The GNOME 2 library - runtime file ii libgnomecanvas2-0 2.10.2-2 A powerful object-oriented display ii libgnomeui-0 2.10.1-1 The GNOME 2 libraries (User Interf ii libgnomevfs2-02.10.1-5 The GNOME virtual file-system libr ii libgtk2.0-0 2.6.9-1The GTK+ graphical user interface ii libpango1.0-0 1.8.2-1Layout and rendering of internatio ii libstdc++64.0.1-6The GNU Standard C++ Library v3 ii libxml2 2.6.20-1 GNOME XML library ii zlib1g1:1.2.3-3 compression library - runtime Versions of packages gnome-games recommends: ii gnome-games-extra-data2.10.0-2 games for the GNOME desktop (extra ii python2.3.5-3An interactive high-level object-o -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#294635: xmms-jess: Inactive Jess uses Memory
On Thu, 2005-02-10 at 21:42 +0100, Thorsten Haude wrote: > Jess uses 50M memory although inactive and never uses in current session: > 50924k: PID 4259 (/usr/lib/xmms/Visualization/libjess.so) > You saw this output from memstat, correct? I can get the same sort of wacky output, but if I remove xmms-jess then I get: 40360k: PID 1527 (/usr/lib/libSDL-1.2.so.0.7.1) With it installed I show jess in two places... [EMAIL PROTECTED]:~$ memstat | grep xmms 40480k: PID 1642 (/usr/lib/xmms/Visualization/libjess.so) ... 40k: /usr/lib/xmms/Visualization/libjess.so 1642 I would guess that this is a bug in memstat since the man page claims: "First, the processes are listed. An amount of memory is shown along with a process ID and the name of the executable which the process is running." libjess.so is not the name of the executable. It should say "/usr/bin/xmms" like ps does. John signature.asc Description: This is a digitally signed message part
Bug#285274: ITP: libvisual
On Tue, 2005-01-11 at 00:12 +0100, Frederik Dannemare wrote: > On Saturday 08 January 2005 00:39, John Lightsey wrote: > > On Sat, 2005-01-08 at 00:00 +0100, Frederik Dannemare wrote: > > > OK, I can confirm my interest in this package. I have pretty much > > > already made the package (split into a lib- and a dev-package). > > > > > > However, I will need a sponsored upload, since I'm currently in the > > > NM queue. I was thinking of asking my AM to work with me on > > > libvisual (and LiVES which makes use of libvisual). > > > > Are you planning on packaging the XMMS libvisual plugin or any of the > > other material outside libvisual itself? > > I wasn't planning on it (at least not initially), but maybe we could > start co-maintaining them or something if you're interested? > I'll package up xmms-libvisual if you don't want it. I'd be happy to work together as much as possible. If I start to fall behind on xmms-libvisual you should feel free to hijack it (and I'll do the same with libvisual.) > > > > OTOH, the libvisual developers who piped in on the Goom-Devel > > > > mailing list told me they'd like to do some cleanup before it's > > > > packaged. > > > > > > I'll keep this in mind. > > > > > > > There is a 0.2.0 release planned for sometime in January, and I > > > > was waiting on that release before creating packages. > > > > > > > > So, feel free to package libvisual. If you lose interest please > > > > let me know. > > > > > > If you can wait a little while for the package to be uploaded I > > > would very much like to take it. > > > > Any chance you could send me a copy to look over? > > Yes, it's now available from > <http://sentinel.dk/linux/debian/packages/libvisual/>. > > And apt-get'able with: > deb http://sentinel.dk/debian/ unstable main contrib non-free > deb-src http://sentinel.dk/debian/ unstable main contrib non-free > > It is lintian clean and the packages I have made available are built in > a pbuilder chroot environment, since building libvisual on my 'normal' > installation complains a bit because I have a mix of nvidia and > nvidia-dev packages installed alongside xlibmesa-gl and > xlibmesa-gl-dev. Don't feel bad. Even if you stick to Free Software compatible video cards you still get sucked into a mixed up system. I'm using a DRI libGL myself. > > As this is my very first attempt of packaging a library and dev package, > critique and comments are very welcome. > > Btw, congratulations on your status as new Debian Developer, I noticed > you were approved last week. > Thanks. I'll grab a copy of your packages, look them over and start building an xmms-libvisual package. I'd sponsor your packages, but I'm still learning myself. I haven't done a regular upload yet. > Best regards, Let me know if I can help. John signature.asc Description: This is a digitally signed message part
Bug#297299: xmms-synaesthesia: FTBFS (amd64/gcc-4.0): invalid lvalue in assignment
On Mon, 2005-02-28 at 15:45 +0100, Andreas Jochens wrote:
...
> With the attached patch 'xmms-synaesthesia' can be compiled
> on amd64 using gcc-4.0.
>
> Regards
> Andreas Jochens
>
> diff -urN ../tmp-orig/xmms-synaesthesia-0.0.3/syna_core.c ./syna_core.c
> --- ../tmp-orig/xmms-synaesthesia-0.0.3/syna_core.c 2005-02-28
> 15:42:08.651295041 +0100
> +++ ./syna_core.c 2005-02-28 15:42:03.28933 +0100
> @@ -204,9 +204,9 @@
> gint x, y, i, j, start, end;
> gint step = outWidth * 2;
>
> - lastLastOutput = lastOutput;
> - lastOutput = output;
> - output = t;
> + synx_output[2] = lastOutput;
> + synx_output[1] = output;
> + synx_output[0] = t;
>
> for (x = 0, i = 0, j = outWidth * (outHeight - 1) * 2; x < outWidth;
> x++, i += 2, j += 2)
> {
> @@ -278,9 +278,9 @@
> gint x, y, i, j, start, end;
> gint step = outWidth * 2;
>
> - lastLastOutput = lastOutput;
> - lastOutput = output;
> - output = t;
> + synx_output[2] = lastOutput;
> + synx_output[1] = output;
> + synx_output[0] = t;
>
> for (x = 0, i = 0, j = outWidth * (outHeight - 1) * 2; x < outWidth;
> x++, i += 2, j += 2)
> {
>
Could you try the version at http://www.nixnuts.net/files/test/ The
only differences to your patch is that lastOutput and output are being
cast to (guint16 *) and config.{sub|guess} have been updated.
I don't have an AMD64 box to test it on.
John
signature.asc
Description: This is a digitally signed message part
Bug#688009: unreproducable
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 09/20/2012 11:39 AM, Henri Salo wrote: > I could not reproduce this issue in squeeze with amd64-machine > using monkey package 0.9.3-1. Could you tell me more about your > virtualization environment? I used a KVM VM running Squeeze with an AMD Athlon(tm) II X4 640 Processor and with the enabled processor features copied from the host in virt-manager. The hypervisor was running linux-image-3.2.0-2-amd64 version 3.2.20-1. It really didn't look like an issue that came up because of my virtualization though, and the VMs I tested with are very solid in my experience. If you'd like, I can get a full backtrace. It takes some effort since the monkey package doesn't handle DEB_BUILD_OPTIONS correctly. It's possible it has to hit the glibc 2.0 compatibility code in either m_build_buffer() or m_build_buffer_from_buffer(). This seemed to be consistent when I was looking at the problem. It's possible my system hit this reliably because of length of the hostname or something along those lines. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAlBbYm8ACgkQBYeybkXz+/kOcACg32HJjqIiiKgCsGyBH+gmOETv MuwAoKChmNNvO8TvR4xhNaIhRxZ0eJfE =qvYO -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#688009: unreproducable
On Fri, 2012-09-21 at 16:07 +0300, Henri Salo wrote: > On Thu, Sep 20, 2012 at 01:37:35PM -0500, John Lightsey wrote: > > If you'd like, I can get a full backtrace. It takes some effort since > > the monkey package doesn't handle DEB_BUILD_OPTIONS correctly. > > I think full backtrace is needed, but at the moment this monkey-package is > unmaintained[1] and contains at least two unfixed security > vulnerabilities[2][3]. Security team is going to request this packages > removal from wheezy. Are you using this in production? Is it something that > only monkey can handle or can it be any www-server software in Debian? I'm not using this software. Closing this bug with the removal of monkey would be the best course of action. signature.asc Description: This is a digitally signed message part
Bug#655435: libapr1: apr_hash vulnerable to oCert-2011-003 style DOS attacks
Package: libapr1 Version: 1.4.5-1.1 Severity: important Tags: security APR's hash implementation is vulnerable to the same types of algorithmic complexity attacks disclosed in oCert-2011-003. Discussion of the problem on the apr-dev mailing list is available here: http://www.mail-archive.com/dev%40apr.apache.org/msg24439.html -- System Information: Debian Release: wheezy/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 3.1.0-1-amd64 (SMP w/4 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages libapr1 depends on: ii libc6 2.13-24 ii libuuid1 2.20.1-1.1 libapr1 recommends no packages. libapr1 suggests no packages. -- 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#688007: monkey: Fails to drop supplemental groups when lowering privileges
Package: monkey Version: 0.9.3-1 Severity: grave Tags: security Justification: user security hole Monkey webserver fails to drop supplemental groups when lowering privileges. This allows any local user on the system to read any fine that root's supplemental groups can access. Monkey does perform a filesystem access check to make sure that its EUID/EGID can access the target file, but this check is subject to TOCTOU flaws. -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (500, 'testing') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-3-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.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#688008: monkey: CGI scripts executed without dropping RUID/RGID root
Package: monkey Version: 0.9.3-1 Severity: grave Tags: security Justification: user security hole The Monkey webserver retains RUID/RGID root so that it can regain root as needed to perform privileged operations. Unfortunately, monkey does not drop RUID/RGID root before executing CGI scripts. This allows any user with write access to a cgi-bin directory to gain local root. It would also allow a remote attacker to do the same in combination with a CGI/PHP script that has any remote code execution bug. -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (500, 'testing') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-3-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.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#688009: monkey: All requests generate segfaults on AMD64 systems
Package: monkey Version: 0.9.3-1 Severity: important In testing the 0.9.3-1 monkey package on an AMD64 KVM VM, all requests to the server resulted in segfaults in the thread trying to handle the request. A 386 test VM did not have the same problem. The segfault occurs when m_build_buffer_from_buffer() calls vsnprintf() while building up the response headers. -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (500, 'testing') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-3-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.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#610384: libparallel-forkmanager-perl: new upstream version
On 03/28/2013 09:58 AM, Damyan Ivanov wrote: > -=| John Lightsey, 18.05.2011 20:25:09 -0500 |=- >> tags 610384 + wontfix >> thanks >> >> The handling of files in /tmp with Parallel::Forkmanager 0.7.6+ is very >> insecure. >> >> http://rt.cpan.org/Ticket/Display.html?id=68298 > > Dear John, > > It seems to me that the current upstream version (1.03) of > Parallel::ForkManager is better in handling temporary files. Although > all the files still use predictable names, they are all created in > a directory created by File::Temp::tmpdir, which should be safe enough > AIUI. > > Maybe you would consider uploading an updated package? Experimental > should be fine if you don't want to disturb the freeze. > > > As an alternative, in case you don't have time for this package, > I offer to take it over to the pkg-perl team (which you are welcome to > join too). If the Debian Perl team would like to take over the three Perl modules I'm maintaining, please feel free to do so. My git repos for these packages are here: http://nixnuts.net/git/libyaml-tiny-perl.git http://nixnuts.net/git/libfinance-quotehist-perl.git http://nixnuts.net/git/libparallel-forkmanager-perl.git I have the 1.02 version of Parallel::Forkmanager packaged already and have been holding off on an upload due to the release freeze. John signature.asc Description: OpenPGP digital signature
Bug#637969: libhttp-dav-perl: New 0.44 upstream version available
Package: libhttp-dav-perl Version: 0.38-1 Severity: normal There have been several releases of HTTP::DAV since the last refresh of the Debian package in 2009. The 0.40 version in particular fixed an annoying bug where you have to specify the realm along with authentication credentials. -- System Information: Debian Release: 6.0.2 APT prefers stable APT policy: (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-amd64 (SMP w/4 CPU cores) Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages libhttp-dav-perl depends on: ii libwww-perl5.836-1 Perl HTTP/WWW client/server librar ii libxml-dom-perl1.44-1Perl module for building DOM Level ii perl 5.10.1-17squeeze2 Larry Wall's Practical Extraction Versions of packages libhttp-dav-perl recommends: ii perl [libdigest-md5-pe 5.10.1-17squeeze2 Larry Wall's Practical Extraction Versions of packages libhttp-dav-perl suggests: ii libcrypt-ssleay-perl 0.57-2 Support for https protocol in LWP -- 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#638002: Improper seteuid() calls in src/log.c and src/masqmail.c
Package: masqmail Version: 0.2.21-4 Severity: critical Tags: security Justification: root security hole Reporting publicly since this has already been disclosed on the masqmail list. In src/log.c there are two logging functions that use this logic: uid_t saved_uid; saved_uid = seteuid(conf.mail_uid); write to a log file... seteuid(saved_uid); The first seteuid() call here isn't returning the previous EUID, it's returning 0 on success and -1 on failure. The net result should be that any time masqmail writes to the log, it's resetting the EUID to root. This would undo the effect of other code in masqmail that drops root privileges. The most recent upstream version of masqmail (0.3.2) contains identical code to the version I audited (Debian stable's version 0.2.27). Per information provided by the upstream author, src/masqmail.c contains additional code with the same type of flaw. -- System Information: Debian Release: 6.0.2 APT prefers stable APT policy: (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-amd64 (SMP w/4 CPU cores) Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (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#636270: FTBFS: test failure: Failed test 'direct split (yahoo) (rows)'
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 08/01/2011 04:00 PM, Dominic Hargreaves wrote: > Source: libfinance-quotehist-perl > Version: 1.16-1 > Severity: serious > Justification: fails to build from source (but built successfully in the past) > > This package FTBFS with a clean sid chroot: Thanks for the report. I'm going to disable all of the remote tests in this package. That should bring it more in line with the policies the debian-perl team is following. http://pkg-perl.alioth.debian.org/policy.html#test_suites -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk43PGIACgkQBYeybkXz+/ns0wCfU0Ye+YlNAW8DSRjlSna0jMtr OZkAnito+nvhNXs2noKaQv3h0MkX2+WB =hASW -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#242613: Switch apt-watch to x-terminal-emulator alternative
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 block 242613 by 531702 thanks I was going to update apt-watch-gnome to use the x-terminal-emulator alternative instead of xterm with the apt-watch 0.4.0 release, but it turns out that gnome-terminal isn't usable in a su-to-root context because of bug 531702. Until gnome-terminal is fixed, I'll leave apt-watch using xterm instead. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBAgAGBQJOWmmYAAoJEORPgBbTYw+JfHsP/3pGpwcUkl1ghG6jHva5C5ZJ zIjK2bn79/VKD34kS6UBBQisE/D9IVFfJ9Ho7nRYhJs5QueHtu1KNMZiCN8RSghQ P7lICtLpvCcp4BtQzcBnZRXOLryN/AK4tPxpWokJRqABnjunp2tdM91deKadTmcm 7wSDl6N6uVQ/1S8UCuOnKe96uXO0YpUrAoDVgRpfcRv4L9rkDbXp/OFAVGLk8w71 v7sWLVsQ4iSdlc6/24UyjU1CkTTQd+24ssawSEpMYUC3A53LgX86WVo1oX6fFn9W y9EAMU7HW1zzS3d26uHXMhFIBlrrIE/yyOTN6cj+L4OtovMzFSaAVNFHdlcvv4HP QSflrKWQw8XRGThIGfGaWYeR/sISRmDQR9duoB5iRqVNOy9T/SJdSC+cAd6hsdPm qHgswufPnYjEiu5/RVF0IYOEZAGJNieZgfjg1KmsJm5iVKaRG38PH8IFQcU9egiS b+7K8fc+a4yW79DKiKgnCHj4eQfz48M44By3Xy9yNyX//k0mvyDbpP+PlrLxHjqd O8HC6kUWvo0rcqT0mkzADGldfDQVrGs3K54wsy3d6tmj5n/j8LyVSxh3fm/q98Qy 7RxeLQj8mq5Ks5EG6jWSUaY96o5BJjqbYC14195uUsY6hxoM9yT3sitZXjaU5Wjo IeGhwYe5lolwGvt92vOf =8Pyt -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#648492: apt-watch: Please add support for build-arch and build-indep targets plus enable hardening flags
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/12/2011 03:03 AM, Niels Thykier wrote: > Please see attached patch as an example of how to do this. > > Please note that the buildds are still using "build" followed by > "binary-arch". > > For more information, please see [1]. > > ~Niels > > [1] https://wiki.debian.org/ReleaseGoals/BuildArchTarget Thanks for the patch. I'll roll this into the next update to apt-watch. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBAgAGBQJOvq4pAAoJEORPgBbTYw+JP+MQAL5Qkz4J13BGTKVI65+aAuG4 FW4EbjAA1Ek23+46pLfER6ENBNuc2ks2j5s18aqTHR6Fmod4Iv5fEbBfA39Dw6ch 8mX0WQ55bWWRc1rdTTNBAQSjnhHUrQhIE+vhxOdLPhaOzLsGf9PSudmCGPZcELpi kOUlVWRpFJfPnxYMGDFCchOQ203UgFhItp/5MY8Drp1judfbRrryRnj+OT3XmJkV EHl41kGbfkpaHWMtrUv4uUI2LpiK7jVn/8Jk3VAHC75cEYQxvHEw3sowR79QWe/+ m5VpTl0TaQZK2TMSWF3bSeBOI7nLzkjDhhk/8G2+G+2aYuHtsNqbBtpMcMYKHyd1 Rpc/ghRfMrkxFSWabyBs9kWBBt1jCRmO5st4p64oLz94B2AQCSVhIYfsziaAmxIx pQMl11Lai69+A0w77quGNktwc0o0Zoi52zPGCth45GXZn4rfkt0nvj4l0MMjIw8a 6Zz9unknWZoopJ9PeocPUQ4NQCmjqEIV7cXTB7GllyRwM9H8NI8UIZmxa0z/PJdA lyL+bZpJsxfL6Koy8pajSHQle9Th333j7+ksMrRRFpQ3oP1Z6dVjTzsIUyyBfS2a MAeTpPSWUg6gLwxmnounwd5jjAWWdVMtUeyx8qtoKuCRONNlBdTziNOlb+oOoh/j /MhJRu46ya3ofU1JM/zP =it16 -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#655044: glib2.0: ghashtable vulnerable to oCert-2011-003 DOS attacks
Source: glib2.0 Severity: important Tags: security The standard hashing functions provided with the ghashtable implementation in glib are vulnerable to the algorithmic complexity attacks described in oCert-2011-003 http://www.ocert.org/advisories/ocert-2011-003.html This was reported upstream in 2003 when Perl fixed their hashing implementation by introducing a random hash seed. The upstream discussion is archived here: http://mail.gnome.org/archives/gtk-devel-list/2003-May/msg00111.html -- System Information: Debian Release: wheezy/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Kernel: Linux 3.1.0-1-amd64 (SMP w/4 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.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#655044: glib2.0: ghashtable vulnerable to oCert-2011-003 DOS attacks
On 01/07/2012 10:34 PM, Michael Biebl wrote:
> On 08.01.2012 02:28, John Lightsey wrote:
> This discussion is from 2003 and had no real conclusion.
> Have you checked if the current code base is still vulnerable?
Yes, I looked at their upstream repo and it appears to me that the
standard hashing functions still have this problem.
guint
g_str_hash (gconstpointer v)
{
const signed char *p;
guint32 h = 5381;
for (p = v; *p != '\0'; p++)
h = (h << 5) + h + *p;
return h;
}
This is a harder to reverse than the standard "h = h * 33 + *p", but the
collisions are predictable.
The hash functions for int64 and double just truncate the keys.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#674678: [nwall] segfault when stderr isn't attached to a tty
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Package: nwall Version: 1.32+debian-4.1 Severity: normal - --- Please enter the report below this line. --- In main(), the return value of ttyname(2) isn't checked which results in a segfault if fd 2 isn't attached to a tty: strace -f perl -e 'open STDERR, ">", "/dev/null"; `nwall -n`' - --- System information. --- Architecture: amd64 Kernel: Linux 3.2.0-2-amd64 Debian Release: wheezy/sid 500 unstableftp.us.debian.org - --- Package information. --- Depends(Version) | Installed -+- libc6 (>= 2.2.5) | 2.13-32 libncurses5 (>= 5.5-5~) | 5.9-7 libreadline6(>= 6.0) | 6.2-8 Package's Recommends field is empty. Package's Suggests field is empty. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk/BBb4ACgkQBYeybkXz+/ldOwCgx+2yHM93yJJhbz6FstRBEbc3 9eAAn2c0+0VBCIOIitvirX/jVPn6FmKM =lgVE -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#672080: apt-watch: FTBFS: apt-watch-common.cc:16:34: error: 'write' was not declared in this scope
Thanks for the bug report on apt-watch. It should be straightforward to fix. signature.asc Description: OpenPGP digital signature
Bug#638074: apt-watch and the ongoing GNOME 3 transition
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 10/19/2011 07:08 PM, Michael Biebl wrote: > Hi John, > > as you might have noticed, the GNOME 3 transition is now ongoing in unstable. > Could you please upload apt-watch 0.4.0 from experimental to unstable now as > otherwise apt-watch will block this transition. > > In case you are currently busy, I can offer to NMU, if you are ok with that. Feel free to NMU. I tried to do the rebuild a few days ago but libpanel-applet-4-dev and libapt-pkg-dev weren't installable. If it's buildable now, please feel free to upload. No changes are necessary from the previous build other than the release target. If no NMU takes place I should be able to get this uploaded tonight assuming its dependencies are now installable. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk6gUEsACgkQBYeybkXz+/k5mQCfTWz44jCOPpHiBW7SPms5wKYd CLQAoOI0vmkTD3WpoUOc9O3gCD/lfcRp =mCku -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#772785: RM: apt-watch -- ROM; obsolete
Package: ftp.debian.org Severity: normal src:apt-watch hasn't served much purpose since the switch to Gnome3 and the deemphasis of panel applets. It will FTBFS shortly (#772617) and should be removed from the archive. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#772617: apt-watch: uses deprecated panel_applet_get_preferences_key
On Tue, 2014-12-09 at 08:34 +, Dmitry Shachnev wrote: > Package: apt-watch-gnome > Version: 0.4.0-2.1 > Severity: important > Justification: will FTBFS soon Thanks for the report. I've requested removal of apt-watch. It hasn't been very relevant since the switch to Gnome3. signature.asc Description: This is a digitally signed message part
Bug#868988: phamm: CVE-2017-0378 reflected XSS in phamm
Source: phamm Severity: important Tags: upstream security While looking through codesearch.debian.net I noticed that phamm's views/helpers.php uses $_SERVER['PHP_SELF'] in a way that is vulnerable to reflected XSS attacks. To reproduce the problem, load a URL like this in Firefox: http://127.0.0.1/phamm/main.php/%22%3E%3Cscript%3Ealert%28123%29%3C/script%3E The Debian Security team assigned this issue CVE-2017-0378 Upstream bug report is here: https://github.com/lota/phamm/issues/21 -- System Information: Debian Release: 9.0 APT prefers stable APT policy: (500, 'stable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.9.0-3-amd64 (SMP w/8 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system)
Bug#848587: castle-combat does not start because numpy.oldnumeric has been removed
On Sun, 2016-12-18 at 18:15 +0100, koopa wrote: > numpy.oldnumeric has been removed in 1.9 release so castle-combat does not > start > https://docs.scipy.org/doc/numpy-dev/release.html#numpy-1-9-0-release-notes > > so castle-combat does not start > Thanks for pointing this out. castle-combat hasn't been developed upstream in many years, so it's unlikely it will be updated for this transition. I'll request removal of the package. signature.asc Description: This is a digitally signed message part
Bug#849171: RM: castle-combat -- ROM; abandoned upstream
Package: ftp.debian.org Severity: normal Please remove the castle-combat package. There has been no upstream development of castle-combat in many years, and it no longer functions due to the removal of the oldnumeric module from numpy. signature.asc Description: This is a digitally signed message part
Bug#425578: apt-watch: should set owner to root.root when moves files to /var/cache/apt
On Tue, 2007-05-22 at 18:09 +0200, Giovanni Mascellani wrote: > While having a look into /var/cache/apt, I noted that some files are > owned by root.root and others by giovanni.giovanni (my username). I > believe that apt-watch moves the .debs from its cache to the system > cache, but doesn't set right permissions. Couldn't this be a security > hole? > > -rw-r--r-- 1 giovanni giovanni 662K 2007-04-14 14:47 yelp_2.18.1-1_i386.deb > -rw-r--r-- 1 root root 44K 2007-04-22 11:47 ytalk_3.3.0-3_i386.deb Sorry for taking so long to respond to this bug report. I agree that the packages in /var/cache/apt should be owned by root. Apt-watch is basically doing a mv of the debs from the user's home directory to the system package cache and it should change the ownership once they have been moved. I'll look at changing this behavior when I have free time to work on apt-watch. John signature.asc Description: This is a digitally signed message part
Bug#369551: castle-combat: Unhandled error in Deferred when shooting
On Tue, 2006-06-20 at 20:07 +0200, Karl Bartel wrote: > I just uploaded a new release. The code worked out quite different > than in the patch, because I wanted to add a "Sound on/off" switch > without adding an additional conditional around each line where a > sound is played. This bug should be fixed nevertheless. > I didn't manage to disable my sound card (well, I didn't try very > hard...) to verify this, so it would be great if you could let me > know whether it works for you. The new version will crash if you don't have a sound card and toggle the sound on/off switch in the config menu. You can't call pygame.mixer.music.stop() without a working mixer. I've updated the 16-no_sound_device.dpatch to fix the problem and it's attached here. There are also a couple of typos still present in 0.8.1: data/doc/rules.html Line 21: s/seperated/separated/ src/gamephases.py Line 242: s/enemey's/enemy's/ John 16-no_sound_device.dpatch Description: application/shellscript
Bug#369551: castle-combat: Unhandled error in Deferred when shooting
On Tue, 2006-05-30 at 18:07 +0300, Timo Juhani Lindfors wrote: > Package: castle-combat > Version: 0.8.0-2 > Severity: grave > Justification: renders package unusable > > When two local players play the game it throws ... > File "/usr/share/games/castle-combat/scripts/cannon.py", line 21, in > __init__ > sound.cannon.play() > exceptions.AttributeError: 'module' object has no attribute 'cannon' > > when player #1 tries to fire the cannon. If you can't reproduce this > please ask for more info. Thanks for pointing this out. This problem comes up when you play castle-combat without any sound device. The cannon sound is loaded in sound.py after "if pygame.mixer.get_init():". This will be false when there is no sound device and the cannon sounds will not be loaded. When the cannon (and wall hit) sounds are played in cannon.py there is no test to see if the sounds were successfully loaded. I'm not certain if calling pygame.mixer.get_init() each time a cannon shot sound is played would cause a notable performance hit and my knowledge of python is still very limited, so I'll just add an "enabled" flag to sound.py and check it in cannon.py before the sounds are played. A patch with the changes I'm applying to the next Debian release of castle-combat is attached to this message. John 16-no_sound_device.dpatch Description: application/shellscript
Bug#377325: [patch] Support updated Python policy
On Sat, 2006-07-08 at 12:52 +0200, Matthias Klose wrote: > Package: castle-combat > Tags: patch > > I'm not uploading this before Jul 15 as a NMU, because it's a change > to the build system as well (introducing debhelper). While I certainly appreciate the help in updating to the new Python policy, I don't understand why switching to debhelper would be the preferred way of doing so. The Debian Python Policy states that python-central can be used without debhelper, though it doesn't provide detailed guidance on how this is done. All three of the examples referenced in the policy guide use debhelper. Do you know where I could find an example of the new python policy implemented without debhelper? Thanks for your help John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#368553: ITP: libparallel-forkmanager-perl -- A simple parallel processing fork manager for Perl
Package: wnpp Severity: wishlist Owner: John Lightsey <[EMAIL PROTECTED]> * Package name: libparallel-forkmanager-perl Version : 0.7.5 Upstream Author : Szabó, Balázs <[EMAIL PROTECTED]> * URL : http://search.cpan.org/~dlux/Parallel-ForkManager-0.7.5/ * License : Perl (GPL/Artistic) Description : A simple parallel processing fork manager for Perl This Perl module is intended for use in operations that can be done in parallel where the number of processes to be forked off should be limited. Typical use is a downloader which will be retrieving hundreds/thousands of files. . Homepage: http://search.cpan.org/~dlux/Parallel-ForkManager-0.7.5/ -- System Information: Debian Release: 3.1 Architecture: i386 (i686) Kernel: Linux 2.6.12-1-686 Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1)
Bug#524474: FollowSymlinks / SymlinksIfOwnerMatch ignored with server-side-includes
This shouldn't be tagged as a grave security issue. The symlink tests in Apache are trivial to overcome with timing attacks and the Apache documentation explicitly states that the symlink tests should not be considered a security restriction. http://httpd.apache.org/docs/2.2/mod/core.html#options John signature.asc Description: This is a digitally signed message part
Bug#431324: FTBFS: undefined reference to `glade_xml_signal_autoconnect'
tags 431324 + pending thanks I isolated the problem with apt-watch yesterday and I'll have a new version uploaded this evening with the fix. John signature.asc Description: This is a digitally signed message part
Bug#421556: ITP: libyaml-tiny-perl -- Read/Write YAML files with as little code as possible
Package: wnpp Severity: wishlist Owner: John Lightsey <[EMAIL PROTECTED]> * Package name: libyaml-tiny-perl Version : 1.04 Upstream Author : Adam Kennedy <[EMAIL PROTECTED]> * URL : http://search.cpan.org/~adamk/YAML-Tiny-1.04/ * License : Perl (GPL/Artistic) Programming Lang: Perl Description : Read/Write YAML files with as little code as possible The YAML specification is huge. Like, really huge. It contains all the functionality of XML, except with flexibility and choice, which makes it easier to read, but with a full specification that is more complex than XML. The pure-Perl implementation of YAML costs just over 4 megabytes of memory to load. Just like with Windows .ini files (3 meg to load) and CSS (3.5 meg to load) the situation is just asking for a YAML::Tiny module, an incomplete but correct and usable subset of the functionality, in as little code as possible. -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18-4-686 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#478440: Should castle-combat be removed from Debian?
On Sun, 2009-09-27 at 15:48 +0200, Luca Falavigna wrote: > I was looking at castle-combat trying to see if it can be ported to > python-numpy because python-numeric* packages have been removed. > > This is not a trivial task because some things have changed and it > crashes every now and then, with twisted too. > > Do you think it can be scheduled for removal from Debian, or will > upstream take care of this? The upstream author of castle-combat is still active and responsive. Unless there's a compelling reason to ask for immediate removal, please give me a little more time to get this resolved. John signature.asc Description: This is a digitally signed message part
Bug#548909: xen-tools: xen-create-image creates world readable disk image files
Package: xen-tools Version: 3.9-4 Severity: grave Tags: security Justification: user security hole I'm tagging this security, though common best practices would suggest that access to the Dom0 should be severely restricted to begin with. When xen-create-image is used to create a file based DomU, the disk image files will have world readable permissions on a typical system with default umask settings. This means that all accounts on the Dom0 will have full access to the data on the DomU. The fix is to simply to alter createLoopbackImages() to chmod 0600 the image files after they are created with DD and before the filesystem is initialized or to simply to adjust the umask before running dd. This problem exists in both the stable 3.9 version of xen-tools and the unstable 4.1 version. -- System Information: Debian Release: 5.0.3 APT prefers stable APT policy: (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.26-1-xen-amd64 (SMP w/4 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages xen-tools depends on: ii debootstrap 1.0.10lenny1Bootstrap a basic Debian system ii libconfig-inifiles-perl 2.39-5 Read .ini-style configuration file ii libtext-template-perl1.44-1.2Text::Template perl module ii perl-modules 5.10.0-19lenny2 Core Perl modules Versions of packages xen-tools recommends: ii libexpect-perl 1.20-1Expect.pm - Perl Expect interface ii reiserfsprogs 1:3.6.19-6User-level tools for ReiserFS file ii rinse 1.3-2 RPM installation environment ii xen-hypervisor-3.2-1-amd64 3.2.1-2.jd1 The Xen Hypervisor on AMD64 ii xen-shell 1.8-3 Console based Xen administration u ii xfsprogs 2.9.8-1lenny1 Utilities for managing the XFS fil xen-tools suggests no packages. -- 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#479222: nsd3: Cron job shouldn't generate output
Package: nsd3 Version: 3.0.7-2~bpo40+1 Severity: minor Please update the cron job for NSD3 so that it doesn't generate output when there is no error. IMHO, this message shouldn't be sent: nsdc: no patch necessary. -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18-6-686 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Versions of packages nsd3 depends on: ii adduser3.102 Add and remove users and groups ii libc6 2.3.6.ds1-13etch5 GNU C Library: Shared libraries ii libssl0.9.80.9.8c-4etch1 SSL shared libraries ii lsb-base 3.1-23.2etch1 Linux Standard Base 3.1 init scrip nsd3 recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#491883: nsd3 init script should not provide "reload" action
Package: nsd3 Version: 3.0.7-2~bpo40+1 Severity: normal The NSD3 init script provides a "reload" action which calls "nsdc reload". As discussed in the upstream bug report, "nsdc reload" does not reread the nsd.conf configuration file, so it's behavior falls far short of the policy manual's guidelines that reload "cause the configuration of the service to be reloaded without actually stopping and restarting the service." http://www.nlnetlabs.nl/bugs/show_bug.cgi?id=193 http://www.debian.org/doc/debian-policy/ch-opersys.html#s9.3.2 NSD isn't capable of "reloading" in the normal sense; only a full restart will cause changes to nsd.conf to take effect. Because of this, it's misleading for the NSD init script to provide a reload action. -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18-6-686 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Versions of packages nsd3 depends on: ii adduser3.102 Add and remove users and groups ii libc6 2.3.6.ds1-13etch5 GNU C Library: Shared libraries ii libssl0.9.80.9.8c-4etch3 SSL shared libraries ii lsb-base 3.1-23.2etch1 Linux Standard Base 3.1 init scrip nsd3 recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#275401: apt-watch: Forced check for updates sometimes returns Archive directory is missing.
tags 275401 +patch
thanks
I can easily replicate this bug...
update apt-watch (works)
launch synaptic from apt-watch
close synaptic
update apt-watch (error)
The attached patch seems to fix the problem with 3.2-2. I haven't
looked at apt-watch closely enough to follow exactly what it's doing,
and I didn't check that this section of code is identical between -2 and
-4.
John
diff -Nur apt-watch-0.3.2.orig/backend/apt-watch-slave.cc apt-watch-0.3.2/backend/apt-watch-slave.cc
--- apt-watch-0.3.2.orig/backend/apt-watch-slave.cc 2004-04-18 10:05:12.0 -0500
+++ apt-watch-0.3.2/backend/apt-watch-slave.cc 2006-01-16 12:47:13.0 -0600
@@ -384,6 +384,7 @@
static void do_update(int outfd)
{
setup_list_dir(outfd);
+ setup_archive_dir(outfd);
SlaveProgress progress(outfd);
signature.asc
Description: This is a digitally signed message part
Bug#354179: apt-watch: I'm sending you an home-made swirl icon, and animation
On Fri, 2006-02-24 at 00:38 +, Eduardo Neves Heleno Silva wrote: > Hi, I made myself a new icon and animation for apt-watch. Unfortunally > it's very hard to make the animation look better, because its size is > very small (24x24) but I think it's a bit beeter than the current one. Personally I like the current swirl graphic. If I remember correctly though, there were other comments asking for RedHat and Ubuntu style icons. My next priority is to get a KDE apt-watch applet working, but I'll look at adding an option to choose from among the various possible graphics. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#364573: apt-watch-backend: Wrong description
On Mon, 2006-04-24 at 12:21 +0200, Martin Schulze wrote: > Package: apt-watch-backend > One of these descriptions is most probably wrong... I'll update the short descriptions with the next version. Thanks for pointing this out. John signature.asc Description: This is a digitally signed message part
Bug#304570: ITP: Codeblocks
Hi there, The other day I was looking for a cross platform C++ IDE that's a bit lighter and better integrated than Eclipse + CDT. The description and screenshots of Codeblocks on their website looked quite good, and I'm a fan of wxWidgets to begin with. After that I saw your ITP on the WNPP list and went to your website to look at the packages you created. Unfortunately, the link you gave is dead now, and it doesn't look like you've updated the ITP in the last 9 months or so. Do you still intend to package codeblocks? Do you have any packages built from the -rc2 release? Thanks John signature.asc Description: This is a digitally signed message part
Bug#594353: apt-watch: FTBFS with apt 0.8.0
tags 594353 + pending thanks I'll upload a new version tonight without -Werror in CXXFLAGS. The package builds fine aside from the deprecation warnings. Thanks for your bug report. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#612914: Fails to build from source
On 02/11/2011 08:28 AM, Moritz Muehlenhoff wrote: > Hi John, > I've tried to rebuild java-imaging-utilities for Univention Corporate Server, > a Debian derived distribution based on Debian stable (currently Lenny, our > next > release will be based on Squeeze). > > libfinance-quotehist-perl fails to build from source, see the following log. > Apparently some of the fetched stock data has changed? Very strange. Thanks for reporting this. I'll dig into it today. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#516315: apt-watch: should check system cache ( especially with cron-apt installed )
On Fri, 2009-02-20 at 15:45 +0100, colliar wrote: > especially with cron-apt installed apt-watch often downloads files for a > second time instead of checking the system cache. Thanks for reporting this issue. This problem will be addressed in the next release of apt-watch. John signature.asc Description: This is a digitally signed message part
Bug#444708: ITA: parrot -- virtual machine for dynamic languages
Hi there, Do you still intend to adopt the Debian Parrot packages? From what I can see, there was some talk in December about setting up group maintenance of Parrot, but it doesn't look like that ever took off. John signature.asc Description: This is a digitally signed message part
Bug#470427: Clicking help in main menu freezes the program
On Tue, 2008-03-11 at 07:23 +0200, Joona Kiiski wrote: > Package: castle-combat > Version: 0.8.1.dfsg.1-0.1 > Severity: normal > When I click help-button in the main menu, nothing happens, but > program becomes unresponsive for mouse > clicks or keypresses. Thanks for pointing this out. It's popping up a browser window with the documentation and shifting input focus to it without leaving fullscreen mode. I'll see what we can do about it this weekend. John -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#592117: debbugsconfig fails due to missing /etc/debbugs/indices/ directory
Package: debbugs
Version: 2.4.2~exp1
Severity: normal
Tags: patch
Hi there, I rebuilt the packages from experimental on my testing system, but
it looks like this problem may exist in previous versions as well. We doing a
fresh install of debbugs and running debbugsconfig for the first time, it dies
while trying to copy the /usr/share/doc/debbugs/examples/sources file to
/etc/debbugs/indices/ since that directory does not exist. This stops
debbugsconfig from running to completion.
-- System Information:
Debian Release: squeeze/sid
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (i686)
Kernel: Linux 2.6.32-5-686 (SMP w/2 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages debbugs depends on:
ii exim4 4.72-1 metapackage to ease Exim MTA (v4)
ii exim4-daemon-light [mail-tran 4.72-1 lightweight Exim MTA (v4) daemon
ii libdebbugs-perl 2.4.2~exp1 modules used by the active Debian
ii perl 5.10.1-13 Larry Wall's Practical Extraction
Versions of packages debbugs recommends:
ii debbugs-web 2.4.2~exp1 web scripts for the active Debian
Versions of packages debbugs suggests:
pn libcgi-alert-perl (no description available)
pn spamassassin (no description available)
-- no debconf information
=== modified file 'debian/debbugsconfig'
--- debian/debbugsconfig 2005-01-16 22:27:09 +
+++ debian/debbugsconfig 2010-08-07 16:03:35 +
@@ -79,6 +79,10 @@
sub template {
my ($name, $destdir) = @_;
if (! -f "$destdir/$name") {
+ if (! -d $destdir ) {
+File::Path::make_path($destdir, {'mode' => 0755 });
+print "created $destdir/.\n";
+ }
system("cp /usr/share/doc/debbugs/examples/$name $destdir/$name") == 0 ||
die "$!";
print "created $destdir/$name from template.\n";
Bug#594090: libyaml-tiny-perl: please remove hype from the package description
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Thanks for your bug report. I'll improve the long description when I package the next upstream release of YAML::Tiny. John -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkxy4RcACgkQBYeybkXz+/kMKQCgwVv5qId5pRdimgoPSaAXL2b8 YC0AmwZCvK5O6SeK2qZNXb6St0TESA7D =Mp20 -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#588401: ITP: libfinance-quotehist-perl -- Perl modules for fetching historical stock quotes from the web
Package: wnpp Severity: wishlist Owner: John Lightsey * Package name: libfinance-quotehist-perl Version : 1.14 Upstream Author : Matthew Sisk * URL : http://search.cpan.org/dist/Finance-QuoteHist/ * License : Perl (Artistic + GPL) Programming Lang: Perl Description : Perl modules for fetching historical stock quotes from the web -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#612914: Fails to build from source
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 fixed 612914 1.14-1+squeeze1 thanks The fixed version of libfinance-quotehist-perl has been accepted into stable-proposed-updates and will be included in the next stable release. The patch applied is: diff --git a/debian/rules b/debian/rules index 2d33f6a..20f4b36 100755 - --- a/debian/rules +++ b/debian/rules @@ -2,3 +2,5 @@ %: dh $@ + +override_dh_auto_test: -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk3gQWEACgkQBYeybkXz+/lPwgCcDJDuTkh7VRmnknsQvgQQ89yc XSwAnRDUQMLztkJ8umri9V/asVCOW9K/ =Hggj -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#610384: libparallel-forkmanager-perl: new upstream version
tags 610384 + wontfix thanks The handling of files in /tmp with Parallel::Forkmanager 0.7.6+ is very insecure. http://rt.cpan.org/Ticket/Display.html?id=68298 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#627742: pu: package libfinance-quotehist-perl/1.14-1+squeeze1
Package: release.debian.org Severity: normal User: release.debian@packages.debian.org Usertags: pu Hi, The unit tests in libfinance-quotehist-perl 1.14-1 fail because they assume that adjusted historical stock quotes queried from public sources will remain constant. This assumption is faulty and the upstream author has overhauled the unit tests in 1.16 to correct this. For 1.14-1 in stable, I'd like to disable the unit tests to close out the FTBFS bug #612914. diff --git a/debian/changelog b/debian/changelog index 2c1d80d..0db8541 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,3 +1,9 @@ +libfinance-quotehist-perl (1.14-1+squeeze1) stable; urgency=low + + * Disable faulty unit tests. (Closts: #612914) + + -- John Lightsey Mon, 23 May 2011 23:58:00 -0500 + libfinance-quotehist-perl (1.14-1) unstable; urgency=low * Initial Release. (Closes: #588401) diff --git a/debian/rules b/debian/rules index 2d33f6a..20f4b36 100755 --- a/debian/rules +++ b/debian/rules @@ -2,3 +2,5 @@ %: dh $@ + +override_dh_auto_test: -- System Information: Debian Release: 6.0.1 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-amd64 (SMP w/4 CPU cores) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#841144: kernel BUG at linux-4.7.5/fs/ocfs2/alloc.c:1514!
tags 841144 + patch
thanks
I'm attaching the patch we used at cPanel to fix this issue with the
4.9 Debian Stable kernel.
I forwarded a version of this patch to the ocfs2-devel mailing list
already.
From Mon Sep 17 00:00:00 2001
From: John Lightsey
Date: Mon, 20 Nov 2017 13:55:05 -0600
Subject: [PATCH] Fix OCFS2 extent split estimation for dio allocators locking.
The dw_zero_count tracking was assuming that w_unwritten_list would
always contain one element. The actual count is now tracked whenever
the list is extended.
---
fs/ocfs2/aops.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index f2961b1..18a311d 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -775,6 +775,8 @@ struct ocfs2_write_ctxt {
struct ocfs2_cached_dealloc_ctxt w_dealloc;
struct list_head w_unwritten_list;
+
+ unsigned int w_unwritten_count;
};
void ocfs2_unlock_and_free_pages(struct page **pages, int num_pages)
@@ -864,6 +866,7 @@ static int ocfs2_alloc_write_ctxt(struct ocfs2_write_ctxt **wcp,
ocfs2_init_dealloc_ctxt(&wc->w_dealloc);
INIT_LIST_HEAD(&wc->w_unwritten_list);
+ wc->w_unwritten_count = 0;
*wcp = wc;
@@ -1364,6 +1367,7 @@ static int ocfs2_unwritten_check(struct inode *inode,
desc->c_clear_unwritten = 0;
list_add_tail(&new->ue_ip_node, &oi->ip_unwritten_list);
list_add_tail(&new->ue_node, &wc->w_unwritten_list);
+ wc->w_unwritten_count++;
new = NULL;
unlock:
spin_unlock(&oi->ip_lock);
@@ -2238,7 +2242,7 @@ static int ocfs2_dio_get_block(struct inode *inode, sector_t iblock,
ue->ue_phys = desc->c_phys;
list_splice_tail_init(&wc->w_unwritten_list, &dwc->dw_zero_list);
- dwc->dw_zero_count++;
+ dwc->dw_zero_count += wc->w_unwritten_count;
}
ret = ocfs2_write_end_nolock(inode->i_mapping, pos, len, len, NULL, wc);
--
2.11.0
signature.asc
Description: This is a digitally signed message part
Bug#841144: Info received (kernel BUG at linux-4.7.5/fs/ocfs2/alloc.c:1514!)
The problem turned out to be slightly more complicated than I initially
imagined.
In addition to calculating the number of extents in the write
incorrectly, the directio write logic can truncate the extent tree
while the write is taking place.
Changwei Ge came up with fully functional patches for the issue.
I'm attaching versions of the upstream patches that have been
backported for the Debian 4.9.0-5 kernel.
From 3e4bcc15f950bf762b48a5c7ce18c93e675de16f Mon Sep 17 00:00:00 2001
From: Changwei Ge
Date: Mon, 8 Jan 2018 01:36:05 +
Subject: [PATCH 1/2] ocfs2: make metadata estimation accurate and clear
Current code assume that ::w_unwritten_list always has only one item on.
This is not right and hard to get understood.
So improve how to count unwritten item.
Reported-by: John Lightsey
Signed-off-by: Changwei Ge
---
fs/ocfs2/aops.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index f2961b13..09e02040 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -775,6 +775,7 @@ struct ocfs2_write_ctxt {
struct ocfs2_cached_dealloc_ctxt w_dealloc;
struct list_head w_unwritten_list;
+unsigned int w_unwritten_count;
};
void ocfs2_unlock_and_free_pages(struct page **pages, int num_pages)
@@ -1364,6 +1365,7 @@ static int ocfs2_unwritten_check(struct inode *inode,
desc->c_clear_unwritten = 0;
list_add_tail(&new->ue_ip_node, &oi->ip_unwritten_list);
list_add_tail(&new->ue_node, &wc->w_unwritten_list);
+wc->w_unwritten_count++;
new = NULL;
unlock:
spin_unlock(&oi->ip_lock);
@@ -2238,7 +2240,7 @@ static int ocfs2_dio_get_block(struct inode *inode, sector_t iblock,
ue->ue_phys = desc->c_phys;
list_splice_tail_init(&wc->w_unwritten_list, &dwc->dw_zero_list);
- dwc->dw_zero_count++;
+ dwc->dw_zero_count += wc->w_unwritten_count;
}
ret = ocfs2_write_end_nolock(inode->i_mapping, pos, len, len, NULL, wc);
--
2.11.0
From d901abb1c95614333a78607fc3e9b02172e18242 Mon Sep 17 00:00:00 2001
From: Changwei Ge
Date: Mon, 8 Jan 2018 01:36:11 +
Subject: [PATCH 2/2] ocfs2: try to reuse extent block in dealloc without
A crash issue was reported by John.
The call trace follows:
ocfs2_split_extent+0x1ad3/0x1b40 [ocfs2]
ocfs2_change_extent_flag+0x33a/0x470 [ocfs2]
ocfs2_mark_extent_written+0x172/0x220 [ocfs2]
ocfs2_dio_end_io+0x62d/0x910 [ocfs2]
dio_complete+0x19a/0x1a0
do_blockdev_direct_IO+0x19dd/0x1eb0
__blockdev_direct_IO+0x43/0x50
ocfs2_direct_IO+0x8f/0xa0 [ocfs2]
generic_file_direct_write+0xb2/0x170
__generic_file_write_iter+0xc3/0x1b0
ocfs2_file_write_iter+0x4bb/0xca0 [ocfs2]
__vfs_write+0xae/0xf0
vfs_write+0xb8/0x1b0
SyS_write+0x4f/0xb0
system_call_fastpath+0x16/0x75
The BUG code told that extent tree wants to grow but no metadata
was reserved ahead of time.
From my investigation into this issue, the root cause it that although
enough metadata is not reserved, there should be enough for following use.
Rightmost extent is merged into its left one due to a certain times of
marking extent written. Because during marking extent written, we got many
physically continuous extents. At last, an empty extent showed up and the
rightmost path is removed from extent tree.
Add a new mechanism to reuse extent block cached in dealloc which were
just unlinked from extent tree to solve this crash issue.
Criteria is that during marking extents *written*, if extent rotation
and merging results in unlinking extent with growing extent tree later
without any metadata reserved ahead of time, try to reuse those extents
in dealloc in which deleted extents are cached.
Also, this patch addresses the issue John reported that ::dw_zero_count is
not calculated properly.
After applying this patch, the issue John reported was gone.
Thanks for the reproducer provided by John.
And this patch has passed ocfs2-test(29 cases) suite running by New H3C Group.
Change from v2:
1)Separate improvement for how to estimate metadata into respective patch
2)Solve the issue reported by John in last version of this patch - try to
reuse metadata in dealloc ahead of claiming new metadata.
3)Add a method ocfs2_find_preferred_free_list() to find space of local slot
at higher priority.
Reported-by: John Lightsey
Signed-off-by: Changwei Ge
---
fs/ocfs2/alloc.c | 208 ---
fs/ocfs2/alloc.h | 1 +
fs/ocfs2/aops.c | 6 ++
3 files changed, 205 insertions(+), 10 deletions(-)
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 06089bec..41f9d0e8 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -164,6 +164,13 @@ static int ocfs2_dinode_insert_check(struct ocfs2_extent_tree *et,
struct ocfs2_extent_rec *rec);
static int ocfs2_dinode_sanity_check(struct ocfs2_extent_tree *et);
static void ocfs2_dinode_fill_root_el(struct ocfs2_extent_tree *et);
+
+static int ocfs2_reuse_blk_f
