Bug#300882: gpgsm: fails to invoke gpg-protect-tool
Package: gpgsm Version: 1.9.15-5 Severity: normal gpgsm fails to invoke gpg-protect-tool because it looks in wrong place. For instance if I want to import secret key: $ gpgsm --call-protect-tool --p12-import --store certkey.p12 gpgsm: error executing `/usr/lib//gpg-protect-tool': No such file or directory secmem usage: 0/16384 bytes in 0 blocks It works just fine if I manually create symlink /usr/lib/gpg-protect-tool -> gnupg2/gpg-protect-tool. -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (990, 'unstable'), (1, 'experimental') Architecture: i386 (i686) Kernel: Linux 2.6.10-1-k7 Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Versions of packages gpgsm depends on: ii libc6 2.3.2.ds1-20 GNU C Library: Shared libraries an ii libgcrypt111.2.0-11 LGPL Crypto library - runtime libr ii libgpg-error0 1.0-1 library for common error values an ii libksba8 0.9.9-2 X.509 and CMS support library ii libopensc1 0.9.4-8 SmartCard library with support for ii libpcsclite1 1.2.9-beta7-4 Middleware to access a smart card ii libusb-0.1-4 2:0.1.10a-6 userspace USB programming library 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#300882: gpgsm: works after fixing probable typo in debian/rules
Package: gpgsm Version: 1.9.15-5 Followup-For: Bug #300882 It seems to work after this trivial patch. --- gnupg2-1.9.15_orig/debian/rules 2005-03-23 14:04:12.0 +0200 +++ gnupg2-1.9.15/debian/rules 2005-03-23 14:06:00.0 +0200 @@ -33,7 +33,8 @@ chmod +x configure env CFLAGS="$(CFLAGS)" \ ./configure --enable-maintainer-mode \ - --prefix=/usr --libexecdir=/usr/lib/ --with-included-gettext \ + --prefix=/usr --libexecdir=/usr/lib/gnupg2 \ + --with-included-gettext \ --with-zlib=/usr --infodir=/usr/share/info/ touch config-stamp I assume it was a simple typo or oversight but if there was a reason for having libexecdir point to /usr/lib then it is another story. -- Andres Alla -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#319604: sl-modem: suboptimal dependencies.
Package: sl-modem Version: 2.9.9a-1 Severity: important sl-modem-modules-_KVERS_ package built from sl-modem-source depends on kernel-image-_KVERS_ but recent (>=2.6.12) Debian kernels provide linux-image instead. That makes sl-modem-modules not installable with recent kernels, hence important. Instead of depending from, it should either recommend kernel-image | linux-image (like eg. ndiswrapper-modules) or not be bothered with it at all (like eg. hostap-modules). Also sl-modem-daemon depends on sl-modem-modules-new | sl-modem-source (>> 2.9.6-1) but can operate without them (ALSA intel8x0m). I suppose just recommending sl-modem-modules-new would be enough. -- Andres Alla -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (990, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.12-1-k7 Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#321049: gs-esp: Failing testcase
Package: gs-esp
Version: 8+8.15rc3.dfsg.1-1
Followup-For: Bug #321049
I'm not the original submitter but the problem seems to be the same.
gs-esp fails if document contains certain fonts. I stumbled upon this
bug with spreadsheet that had pages with Arial printing fine but pages
with Helvetica not. I can reproduce this bug 100% with documents that
contain Helvatica, so that is my submitted testcase.
Also, as in original report, with gs-esp/testing (7.07.1-9) everything
is OK.
I also include snippet of cups error log at level debug while trying to
print failing document.
-- Andres Alla
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (990, 'unstable'), (500, 'testing'), (500, 'stable'), (1,
'experimental')
Architecture: i386 (i686)
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.12-1-k7
Locale: LANG=C, LC_CTYPE=C (charmap=ISO-8859-15) (ignored: LC_ALL set to et_EE)
Versions of packages gs-esp depends on:
ii gs-common 0.3.8 Common files for different Ghostsc
ii libc6 2.3.5-3GNU C Library: Shared libraries an
ii libcupsimage2 1.1.23-11 Common UNIX Printing System(tm) -
ii libcupsys2-gnutls10 1.1.23-11 Common UNIX Printing System(tm) -
ii libfreetype6 2.1.10-1 FreeType 2 font engine, shared lib
ii libice6 6.8.2.dfsg.1-5 Inter-Client Exchange library
ii libjpeg62 6b-10 The Independent JPEG Group's JPEG
ii libpaper1 1.1.14-3 Library for handling paper charact
ii libpng12-01.2.8rel-1 PNG library - runtime
ii libsm66.8.2.dfsg.1-5 X Window System Session Management
ii libstdc++64.0.1-4The GNU Standard C++ Library v3
ii libtiff4 3.7.3-1Tag Image File Format (TIFF) libra
ii libx11-6 6.8.2.dfsg.1-5 X Window System protocol client li
ii libxext6 6.8.2.dfsg.1-5 X Window System miscellaneous exte
ii libxt66.8.2.dfsg.1-5 X Toolkit Intrinsics
ii xlibs 6.8.2.dfsg.1-5 X Window System client libraries m
ii zlib1g1:1.2.3-3 compression library - runtime
Versions of packages gs-esp recommends:
ii gsfonts 8.14+v8.11+urw-0.2 Fonts for the Ghostscript interpre
ii psfontmgr 0.11.8-0.1 PostScript font manager -- part of
-- no debconf information
321049_Helvetica12.pdf
Description: Adobe PDF document
{\rtf1\ansi\deff0\adeflang1025
{\fonttbl{\f0\froman\fprq2\fcharset0 Times New Roman;}{\f1\froman\fprq2\fcharset0 Times New Roman;}{\f2\fswiss\fprq2\fcharset0 Helvetica;}{\f3\fnil\fprq2\fcharset0 Arial;}{\f4\fnil\fprq2\fcharset0 Tahoma{\*\falt Lucidasans};}{\f5\fnil\fprq0\fcharset0 Tahoma{\*\falt Lucidasans};}}
{\colortbl;\red0\green0\blue0;\red128\green128\blue128;}
{\stylesheet{\s1\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af4\afs24\lang255\ltrch\dbch\af3\afs24\langfe255\loch\f0\fs24\lang1061\snext1 Default;}
{\s2\sa120\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af4\afs24\lang255\ltrch\dbch\af3\afs24\langfe255\loch\f0\fs24\lang1061\sbasedon1\snext2 Text body;}
{\s3\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af5\afs24\lang255\ltrch\dbch\af3\afs24\langfe255\loch\f0\fs24\lang1061\sbasedon2\snext3 List;}
{\s4\sb120\sa120\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af5\afs20\lang255\ai\ltrch\dbch\af3\afs20\langfe255\ai\loch\f0\fs20\lang1061\i\sbasedon1\snext4 Caption;}
{\s5\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af5\afs24\lang255\ltrch\dbch\af3\afs24\langfe255\loch\f0\fs24\lang1061\sbasedon1\snext5 Index;}
}
{\info{\author Andres Alla}{\creatim\yr2005\mo8\dy15\hr13\min11}{\operator Andres Alla}{\revtim\yr2005\mo8\dy15\hr13\min12}{\printim\yr1601\mo1\dy1\hr0\min0}{\comment StarWriter}{\vern6450}}\deftab720
{\*\pgdsctbl
{\pgdsc0\pgdscuse195\pgwsxn11905\pghsxn16837\marglsxn1134\margrsxn1134\margtsxn1134\margbsxn1134\pgdscnxt0 Default;}}
\paperh16837\paperw11905\margl1134\margr1134\margt1134\margb1134\sectd\sbknone\pgwsxn11905\pghsxn16837\marglsxn1134\margrsxn1134\margtsxn1134\margbsxn1134\ftnbj\ftnstart1\ftnrstcont\ftnnar\aenddoc\aftnrstcont\aftnstart1\aftnnrlc
\pard\plain \ltrpar\s1\cf0{\*\hyphen2\hyphlead2\hyphtrail2\hyphmax0}\rtlch\af4\afs24\lang255\ltrch\dbch\af3\afs24\langfe255\loch\f2\fs24\lang1061 {\loch\f2\fs24\lang1061\i0\b0 321049}
\par }
cups_error_log_job1.gz
Description: Binary data
Bug#1077806: libreoffice-core: libreoffice (i386) crashes on startup
Package: libreoffice-core Version: 4:24.2.5-1 Severity: important Dear Maintainer, libreoffice (i386) crashes on startup (backtrace attached). Splash is shown and then nothing, application window does not show up. It is the same in safe-mode. I suspect it may be i386 specific. System is Debian unstable, mix of i386 and amd64; i386 is primary architecture, most of X desktop applications are i386, as is libreoffice. I am unable to pinpoint what change caused it. Machine was not rebooted since february but was gradually upgraded. Libreoffice has been open for weeks. It is possible that actual breaking change happened several versions past but went unnoticed because previous versions of libraries were used by still running program. Problem did not appear until reboot. I see pdf-s generated by 24.2 in the end of june but there is no more specific version in metadata. Reinstalled libreoffice and stripped down to just libreoffice-writer, -calc and strictly required dependencies, no DE or language-specific packages, no java- support etc. Still crashing. Tried all 24.2 versions (with corresponding dependencies) that have ever been in unstable from https://snapshot.debian.org/package/libreoffice/ and 4:24.8.0~rc1-1 from experimental. All crashing. 4:7.4.7-1+deb12u3 from stable works somewhat, writer at least, calc crashes because of some libxml discrepancy. Installing amd64 version in that kind of mixed environment is not trivial (aptitude could not find workable solution) but I did it through some abuse of dpkg (+f1 packages in the list below, just added Multi-Arch: foreign to control and repacked). Works just fine. List of packages below that reportbug generated corresponds to currently installed and working amd64 version but is essentially same for i386. Attached backtrace (of crashing i386 version) looks rather sparse to me, gdb supposedly loaded symbols through debuginfod. Happy to debug further but would need some guidance on where to look. -- Andres Alla -- Package-specific info: All deployed bundled extensions: All deployed shared extensions: All deployed user extensions: Experimental features enabled: false Installed VCLplugs: Desired=Unknown/Install/Remove/Purge/Hold | Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend |/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad) ||/ Name Version Architecture Description +++---- = un libreoffice-gtk3 (no description available) un libreoffice-kf5(no description available) un libreoffice-qt5(no description available) -- System Information: Debian Release: trixie/sid APT prefers unstable APT policy: (990, 'unstable'), (1, 'experimental') Architecture: i386 (x86_64) Foreign Architectures: amd64, armel Kernel: Linux 6.9.10-amd64 (SMP w/4 CPU threads; PREEMPT) Kernel taint flags: TAINT_FIRMWARE_WORKAROUND Locale: LANG=, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8), LANGUAGE not set Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages libreoffice-core depends on: ii fontconfig 2.15.0-1.1 ii fonts-opensymbol4:102.12+LibO24.2.5-1 ii libabsl20230802 20230802.1-4 ii libargon2-1 0~20190702+dfsg-4+b1 ii libboost-locale1.83.0 1.83.0-3+b2 ii libc6 2.39-6 ii libcairo2 1.18.0-3+b1 ii libclucene-contribs1t64 2.3.3.4+dfsg-1.2 ii libclucene-core1t64 2.3.3.4+dfsg-1.2 ii libcmis-0.6-6t640.6.2-2.1+b1 ii libcups2t64 2.4.10-1 ii libcurl4t64 8.9.0-3 ii libdbus-1-3 1.14.10-4+b1 ii libdconf1 0.40.0-4+b2 ii libeot0 0.01-5+b1 ii libepoxy0 1.5.10-1+b2 ii libexpat1 2.6.2-1 ii libexttextcat-2.0-0 3.4.7-1 ii libfontconfig1 2.15.0-1.1 ii libfreetype62.13.2+dfsg-1+b4 ii libgcc-s1 14.1.0-5 ii libglib2.0-0t64 2.80.4-1 ii libgpgmepp6t64 1.18.0-4.1+b2 ii libgraphite2-3 1.3.14-2 ii libgstreamer-plugins-base1.0-0 1.24.6-dmo1 ii libgstreamer1.0-0 1.24.6-1 ii libharfbuzz-icu08.3.0-2+b1 ii libharfbuzz0b 8.3.0-2+b1 ii libhunspell-1.7-0 1.7.2+really1.7.2-10+b2 ii libhyphen0 2.8.8-7+b1 ii libice6 2:1.0.10-1+b1 ii libicu7272.1-5 ii libjpeg62-turbo 1:2.1.5-3 ii liblcms2-2 2.14-2+b1 ii libldap-2.5-0 2.5.18+dfsg-2 ii libmythes-1.2-0 2:1.2
Bug#1077806: libreoffice-core: libreoffice (i386) crashes on startup
Hi
reede, 2. august 2024 20:01:10 EEST kirjutas Rene Engelhard:
> Am 02.08.24 um 17:25 schrieb Andres Alla:
> > []
> > I suspect it may be i386 specific.
> I suspect, too. Works definitely on amd64.
It may be fixed in 24.8.0 but see below.
After some detouring I stumbled on upstream commit
https://github.com/LibreOffice/core/commit/542e590
that supposedly fixes this bug.
I now upgraded to 4:24.8.0~rc3-2 from experimental and everything works fine
with SAL_USE_VCLPLUGIN set either kf5 or gen.
4:24.2.5-3 from unstable also worked if libreoffice-plasma installed and
SAL_USE_VCLPLUGIN set to kf5; crashed if gen.
I am not sure why it did not show up on amd64 because it seems to be rather
KDE- than i386-specific.
Also I think that I tried 24.8.0~rc1 from experimental (includes fix) as first
thing after encountering crash of then sid-version and that crashed too but
maybe it was something else, can not reproduce it now anyway.
As a hint to future bug-hunters. I understood that backtrace included in
original report is nearly useless but it did not occur to me why gdb could not
do better. Until I realized to purge debuginfod cache. Then gdb started to
give meaningful output.
New backtrace of crash attached.
Also, in an effort to have more consistent environment between runs I started
LO with switch --norestore. And it did not crash. That was awkward since it
shouldn't have made a difference since I ran LO with pristine user profile
anyway (that I failed to mention in original report).
So
$ libreoffice -env:UserInstallation=file:///tmp/testlo --backtrace
crashed, but
$ libreoffice -env:UserInstallation=file:///tmp/testlo --norestore --backtrace
did not.
Comparing strace output of both ways, I saw that only relevant files first one
touched before crashing code that second one did not were:
/run/user/*/iceauth_*
/tmp/.ICE-unix/*
/usr/lib/libreoffice/share/config/soffice.cfg/svx/ui/docrecoverysavedialog.ui
That suggests something to do with session management as is crashing function
SessionManagerClient::open().
So I ran LO under gdb to watch more closely what happens before crash and
discovered that rtl_string_newFromStr() was called with 2nd parameter (const
char *) as null. That of course leads to strlen of it segfaulting.
Googleing "sm.cxx null" led to git commit referred above.
> > System is Debian unstable, mix of i386 and amd64; i386 is primary
> > architecture, most of X desktop applications are i386, as is libreoffice.
>
> *sigh*. Why are you using libreoffice for i386 then?
Because installing amd64 version of LO on this kind of system is not
straightforward and it has been working well as is.
> And why people do think this mix makes sense in any wy?
Because it works and there is no pressing need to fix what works.
It is very old install, lived through dozen-ish machines. Gradually updated,
it actually works very well and any kind of breakage is surprisingly rare.
Larger changes - like cross-grading i386 to amd64 would be - are generally
done along with moving to a new machine but arch-transition probably did not
seem important during last couple of moves, maybe next time.
Of course software that clearly benefits from or requires that, scilab for
instance is run as amd64.
Possibility of running system like this with little to no drama speaks very
highly of Debian ("unstable" no-less).
I deeply appreciate your work making it happen.
> []
> > 4:7.4.7-1+deb12u3 from stable works somewhat, writer at least, calc
> > crashes because of some libxml discrepancy.
> What exactly? That sounds like the (broken) libxml2 in unstable also needs
> an other conflict...
Have not yet got time to look into this again, maybe later.
> []
> > Architecture: i386 (x86_64)
> > Foreign Architectures: amd64, armel
> even armel there?
Sometimes useful to run binaries from embedded-devices with qemu.
As I wrote above, it works, why not? ;-)
> []
> > Versions of packages libreoffice-core depends on:
> > []
> > ii ure 4:24.2.5-1+f1
> What on earth is this? This is some random version which doesn't exist?
As I wrote in original report +f1-versioned packages in the list are due to my
dirty hack to get working amd64 version to install on this system.
It is just original binary package extracted, added "Multi-Arch: foreign" to
control file, adjust version and then dpkg-deb --build.
I understand that making LO truly multi-arch installable is much more involved
and I am not pestering you into this (unless I'd come up with patch or sth).
Now that I got it working, it is back to all original Debian, without these
weird packages.
> []
--
Andres Alla
warning: Currently logging to gdbtrace.log. Turn the logging off and on to make the new setting effective.
[Thread debuggin
Bug#1077806: libreoffice-core: libreoffice (i386) crashes on startup
Hi pühapäev, 25. august 2024 17:46:26 EEST kirjutas Rene Engelhard: > Am 22.08.24 um 23:01 schrieb Andres Alla: > > > > [] worked if libreoffice-plasma installed and > > SAL_USE_VCLPLUGIN set to kf5; crashed if gen. > > Well, as I said I tried with "gen" in my VM, too. > > > I am not sure why it did not show up on amd64 because it seems to be > > rather KDE- than i386-specific. > > As said hee it didn't even affect a clean i386 VM. It depends on what you used as X session manager. As far as I understand the bug triggers only if LO is ran with VCL=gen under fairly recent version of ksmserver (from package plasma-workspace) managed session. > [] > But should we close it with 4:24.8.0~rc3-2 then? Yes. I am certain that commit I referred to fixes the bug and is included in 24.8.0. Maybe also to change title to better reflect essence of bug, so it would be easier to find if somebody else stumbles on it. As we know now it could affect any earlier version of LO under recent Plasma. I propose: libreoffice crashes if run under KDE Plasma with SAL_USE_VCLPLUGIN set to gen. -- Andres Alla tel +372 513 5252 e-post and...@alla.ee
