no task description

Attached to Project: Archlinux32
Opened by Andreas Baumann - 13.12.2017
Last edited by Erich Eckner - 03.01.2018
 FS#21  - [ceph] unit tests failing or segfault

5/142 Test   #3: test_objectstore_memstore.sh ............***Failed    1.13 sec
7/142 Test   #1: run-rbd-unit-tests.sh ...................***Failed    1.49 sec

46/142 Test #57: unittest_util ………………………*Failed 0.05 sec
51/142 Test #59: unittest_lru ……………………….*Failed 0.57 sec
92/142 Test #100: unittest_erasure_code_shec_arguments ….*Failed 0.63 sec
96/142 Test #102: unittest_journal ……………………*Exception: SegFault 0.90 sec
98/142 Test #106: unittest_mds_sessionfilter …………..*Exception: SegFault 1.02 sec
106/142 Test #116: unittest_bluefs …………………….*Exception: SegFault 0.66 sec
107/142 Test #117: unittest_bluestore_types …………….*Exception: SegFault 1.25 sec
108/142 Test #119: unittest_memstore_clone ……………..*Exception: SegFault 2.51 sec
114/142 Test #124: unittest_osdscrub …………………..*Exception: SegFault 0.63 sec
115/142 Test #125: unittest_pglog ……………………..*Exception: SegFault 1.23 sec
116/142 Test #126: unittest_hitset …………………….*Failed 0.52 sec
125/142 Test #134: test_ceph_argparse.py ……………….*Failed 1.64 sec
130/142 Test #6: run-tox-ceph-disk …………………..*Failed 94.75 sec
132/142 Test #142: unittest_rbd_mirror …………………*Exception: SegFault 0.51 sec
140/142 Test #2: run-cli-tests ………………………*Failed 167.72 sec
141/142 Test #110: mgr-dashboard-smoke.sh ………………*Failed 301.60 sec

Closed by Erich Eckner
03.01.2018 21:15
Reason for closing: Won’t fix
Additional comments about closing:

blacklisted ceph

Comments (9)

Admin
Erich Eckner commented on 17.12.2017 19:11

I’ll build it w/o check(), but we should definitely look into this (later)
Admin
Andreas Baumann commented on 19.12.2017 12:09

ceph goes heavy non-i686:

shell#> cmake ….

Error at cmake/modules/BuildDPDK.cmake:61 (message):

not able to build DPDK support: unsupported target.
"i686-native-linuxapp-gcc" not listed in

Call Stack (most recent call first):

cmake/modules/BuildDPDK.cmake:83 (do_build_dpdk)
cmake/modules/BuildSPDK.cmake:4 (build_dpdk)
CMakeLists.txt:239 (build_spdk)

I suspect libvirt and the other packages can use ceph, but do not require it
really. I would make ceph an optdepend.
Admin
Andreas Baumann commented on 19.12.2017 12:10

Using it without check is no option IMHO.
Admin
Andreas Baumann commented on 03.01.2018 19:29

New try, same bugs:

        1 - run-rbd-unit-tests.sh (Failed)
        2 - run-cli-tests (Failed)
        3 - test_objectstore_memstore.sh (Failed)
       59 - unittest_lru (Failed)
      100 - unittest_erasure_code_shec_arguments (Failed)
      102 - unittest_journal (SEGFAULT)
      106 - unittest_mds_sessionfilter (SEGFAULT)
      116 - unittest_bluefs (SEGFAULT)
      117 - unittest_bluestore_types (SEGFAULT)
      119 - unittest_memstore_clone (SEGFAULT)
      124 - unittest_osdscrub (SEGFAULT)
      125 - unittest_pglog (SEGFAULT)
      126 - unittest_hitset (Failed)
      134 - test_ceph_argparse.py (Failed)
      142 - unittest_rbd_mirror (SEGFAULT)

Decision: remove ceph dependency on libvirt (more?), then blacklist the package.

Reason: we cannot maintain all software for companies upstream.
Admin
Andreas Baumann commented on 03.01.2018 19:45

mmh, a (most likely incomplete) list of software using ceph:

- libvirt: seems to be an optional storage method
- qemu: obvious, if used in combination with libvirt to store the disk image with ceph
- pifpaf: “Suite of tools and fixtures to manage daemons for testing”, ceph used in tests only
- fio: “Scriptable I/O tool for storage benchmarks and drive testing”, ceph seems to be optional

So, all seem to work fine without ceph.
Admin
Andreas Baumann commented on 03.01.2018 19:53

ceph on gihub: a project with pull requests only and no bug reports.. ok then.
Admin
Erich Eckner commented on 03.01.2018 20:59

agreed: I’ll blacklist it, once I compiled a list of ceph dependent packages.
Admin
Erich Eckner commented on 03.01.2018 21:04

I only see libvirt, python-pifpaf and qemu depending on ceph
Admin
Erich Eckner commented on 03.01.2018 21:07

ah, that’s because you already removed the dependencies

Google Cache

Bing Cache

I scheduled a rebuild of viewnior, let me know if viewnior 1.6-3.1 works.

no task description

Attached to Project: Archlinux32
Opened by Andreas Baumann - 02.02.2018
Last edited by Andreas Baumann - 02.02.2018
 FS#26  - [texlive] partially linked against old libmpfr

( 9/18) Updating TeXLive filename database… texlua: error while loading shared libraries: libmpfr.so.4: cannot open shared object file: No such file or directory

Triggered a forced rebuild of textlive-core and texlive-bin.
Closed by Andreas Baumann
02.02.2018 10:15
Reason for closing: Fixed

  Comments (0)

Google Cache

Attached to Project: Archlinux32
Opened by Andreas Baumann - 02.02.2018
Last edited by Andreas Baumann - 03.02.2018
 FS#27  - man breaks on gdbm sobump

man
man: error while loading shared libraries: libgdbm.so.4: cannot open shared object file: No such file or directory

either you must link against libgdbm_compat.so.4 or against libgdbm.5
Closed by Andreas Baumann
03.02.2018 07:47
Reason for closing: Fixed
Additional comments about closing:

fixed in man-db-2.7.6.1-3.1, installing that package from testing as temporary is the better
solution than adding a symlink you make forget to delete..

  Comments (4)
  Related Tasks (0/0)

Admin
Andreas Baumann commented on 02.02.2018 17:50

temporary workaround: ln -fs libgdbm_compat.so.4 /usr/lib/libgdbm.so.4
Admin
Andreas Baumann commented on 02.02.2018 17:53

weird: Archlinux has:

/usr/bin/man is owned by man-db 2.7.6.1-3
gdbm 1.14.1-1

Archlinux 32 has:

man-db 2.7.6.1-3.0
gdbm 1.14.1-1.0

This seems to be pretty much the same version.
Admin
Erich Eckner commented on 02.02.2018 18:15

it’s amazing what packages we managed to break I scheduled man-db for a rebuild
Admin
Andreas Baumann commented on 02.02.2018 20:38

man-db is rebuilding on my slave. *fingers crossed*

Google Cache

12.04.2018 - io_queue_run.os: In function `io_queue_run’: io_queue_run.c:(.text+0×71): undefined reference to `__stack_chk_fail_local’ io_getevents.os: In …

Attached to Project: Archlinux32
Opened by Swift Geek - 16.03.2018
Last edited by Andreas Baumann - 26.03.2018
 FS#31  - librsvg fails with invalid opcode on 2.42.1 and newer

traps: gtk3-demo[365] trap invalid opcode ip:aedd3e15 sp:bffc64c0 error:0 in librsvg-2.so.2.42.3[aed27000+11a000]

Downgrading to 2.40.19 seems to help.
https://archive.archlinux32.org/repos/2017/11/01/extra/os/i686/librsvg-2:2.40.19-1-i686.pkg.tar.xz

This could be related to machine not having SSE2 (Pentium III-M Tualin)

Related bbs thread: https://bbs.archlinux32.org/viewtopic.php?id=1369 Closed by Andreas Baumann
26.03.2018 15:02
Reason for closing: Fixed

  Comments (1)
  Related Tasks (0/0)

Paul Gover commented on 26.03.2018 11:03

For me, librsvg-2:2.42.3-1.2 fixes the problem

Google Cache

11.04.2018 - samba-4.7.5-1.0 smbclient-4.7.5-1.0 libwbclient-4.7.6-1.0. Downgrading those three packages to 4.7.4 as a temporary workaround.

no task description

Attached to Project: Archlinux32
Opened by Andreas Baumann - 10.05.2018
Last edited by Andreas Baumann - 23.01.2019
 FS#37  - ISO 2018.05.01 is not bootable with qemu or Virtualbox

I tried the i686 image. Booting leads to kernel panic.
Adding an explicit init=/lib/systemd/systemd parameter also panics.

qemu-system-i386 -cdrom archlinux-2018.05.01-i686.iso

It also panics in Virtualbox.

The same ISO works with libvirtd though.
Closed by Andreas Baumann
23.01.2019 18:29
Reason for closing: Fixed
Additional comments about closing:

works

  Comments (6)
  Related Tasks (0/0)

Admin
Tyzoid commented on 10.05.2018 10:49

i686 iso works fine in virtualbox on both a windows and linux host from my testing.
Admin
Tyzoid commented on 10.05.2018 11:04

Issue confirmed on qemu-system-i386 and qemu-system-x86_64. archlinux-2018.04.01-i686.iso fails to boot correctly as well.

Screenshot of the issue on i386 qemu: https://i.imgur.com/KRt21gI.png Screenshot of the issue on x86_64 qemu: https://i.imgur.com/wozM2hb.png Admin
Andreas Baumann commented on 11.05.2018 09:02

So /lib/systemd/systemd is either not found or cannot be executed, maybe because of a wrong
shared library dependency? I doubt it has to do with the kernel itself, as the log message
indicates the initial ram disk has been loaded.
Admin
Andreas Baumann commented on 16.06.2018 18:35

The write error indicates that the ramdisk could not be extracted.

So, if I start qemu with:

qemu-system-i386 -enable-kvm -cpu pentium3 -m 2048 -cdrom archlinux-2018.06.01-i686.iso

it works.

The default in qemu is 128MB Admin
Andreas Baumann commented on 16.06.2018 18:48

Sadly this is not the cause for VirtualBox, there it still fails even with 2048MB memory.
Admin
Andreas Baumann commented on 23.01.2019 18:18

The 2019.01.04 ISO works with Virtualbox 6.0.0.

Google Cache

Attached to Project: Archlinux32
Opened by Luke Shumaker - 13.06.2018
Last edited by Erich Eckner - 28.06.2018
 FS#41  - [extra/vlc] is uninstallable because [extra/ffmpeg2.8] was removed

[extra/ffmpeg2.8] (which provided ffmpeg=2.8; [extra/ffmpeg] is 3.4) was removed, despite that it was still needed by [extra/vlc].

[extra/vlc] is an old version (2.2); the current version (3.0), which no longer depends on ffmpeg2.8, is currently in [testing].

(I am unsure if [testing/vlc] depends on extra/ffmpeg=3.4 or testing/ffmpeg=4.0)
Closed by Erich Eckner
28.06.2018 20:25
Reason for closing: Fixed

  Comments (3)
  Related Tasks (0/0)

Admin
Tyzoid commented on 13.06.2018 21:39

Checking https://packages.archlinux32.org/extra/i686/vlc/ vs https://packages.archlinux32.org/testing/i686/vlc/, it appears that the extra/ffmpeg satisfies the testing/vlc requirements.

Here’s from vlc in testing:
libavcodec.so.58 (ffmpeg)
libavformat.so.58 (ffmpeg)
libavutil.so.56 (ffmpeg)

vs vlc in extra:
not satisfiable dependency: “libavcodec.so.56” not satisfiable dependency: “libavformat.so.56” not satisfiable dependency: “libavutil.so.54” Admin
Erich Eckner commented on 18.06.2018 09:37

vlc 3.0.3 is now in extra
Admin
Andreas Baumann commented on 21.06.2018 11:36

vlc 3.0.3 works fine on testing.

Google Cache

no task description

20.06.2018 - In file included from /root/qt5-webengine/src/qtwebengine-everywhere-src-5.11.0/src/plugins/qwebengineview/qwebengineview_plugin.cpp:40:

21.06.2018 - Both the gtk2 and the qt5 version fail with: [FORMS.PP] ExceptionOccurred Sender=EAccessViolation Exception=Access violation Stack trace: …

Attached to Project: Archlinux32
Opened by Levi - 20.08.2018
Last edited by Erich Eckner - 29.10.2018
 FS#49  - ffmpeg update failures in release and testing

When I have done a pacman -Syu recently, it offers to update libx264 with x264:

:: Replace libx264 with testing/x264? [Y/n]
resolving dependencies… looking for conflicting packages… error: failed to prepare transaction (could not satisfy dependencies)
:: removing libx264 breaks dependency ‘libx264.so=152-32’ required by ffmpeg

As can be seen there, it fails because a dependency is not met.

This is reportedly also a problem on the release branches.

Using pacman’s query functionality I can see that x264 claims to provide libx264.so=155-32 rather than 152-32 i.e. x264 is too new for the current version of ffmpeg.

I consider this a high priority defect because ffmpeg is used by rather a lot of a/v playback and transcoding software.
Steps to reproduce

$ pacman -Syu

Workaround

Press n and enter when it offers to upgrade libx264. Ensure you’re not using multiple -ys on your pacman invocation.
Suggested fix

Provide an updated ffmpeg.
Closed by Erich Eckner
29.10.2018 11:50
Reason for closing: Fixed

  Comments (0)

Google Cache

Bing Cache

24.09.2018 - The current extra/imagemagick depends on ‘perl<5.27’, but core/perl is 5.28. $ pacman -Si imagemagick perl Repository : extra Name …

Attached to Project: Archlinux32
Opened by Paul Gover - 07.09.2018
Last edited by Erich Eckner - 24.09.2018
 FS#51  - ufw needs rebuild for python 3.7

Been using ufw for some time. Today, I wanted to turn logging off.
Command “ufw logging off” should just work. Instead it gives:

ufw logging off
Traceback (most recent call last):

File “/usr/bin/ufw”, line 26, in

import ufw.frontend

ModuleNotFoundError: No module named ‘ufw’

which I trace to the fact ufw is installed in
/usr/lib/python3.6/site-packages/ufw
and there’s no longer a python 3.6 executable; it’s all 3.7 now.
Closed by Erich Eckner
24.09.2018 07:32
Reason for closing: Fixed

  Comments (3)
  Related Tasks (0/0)

Admin
Erich Eckner commented on 07.09.2018 09:46

ufw-0.35-5.0 from [community-testing] should work - I’m moving it to [community] just now
Paul Gover commented on 07.09.2018 09:48

Sorry to add to your workload. I assumed a UFW problem was local to me. Now I’ve read the furore about python 3.7 etc. on the forums, my apologies. I’ll try the community-testing version. Thanks for the speedy response.
Paul Gover commented on 11.09.2018 15:21

To confirm, UFW is working fine for me now. Thanks.

Google Cache

Attached to Project: Archlinux32
Opened by Luke Shumaker - 21.09.2018
 FS#52  - Some packages are built by “Unknown Packager”

Some packages have

packager = Unknown Packager

, which is caused by

makepkg.conf:PACKAGER

not being set.

I analyzed every .pkg.tar.xz file in the Arch 32 repos (all architectures). The earliest Unknown Packager package was built on June 9, 2017, and they’ve been intermittent since then. Here’s a listing of how many times each packager appears:

    2  Dave Reisner 
    2  Fabio Castelli (Muflone) 
    2  Giancarlo Razzolini 
    2  Jelle van der Waa 
    2  Laurent Carlier 
    2  S?bastien Luttringer 
    2  schuay 
    3  Florian Pritz 
    3  Johannes L?thberg 
    4  Alexander R?dseth 
    4  Andreas Radke 
    4  Eli Schwartz 
    4  Tobias Powalowski 
    5  Gaetan Bisson 
    5  Jonathan Steel 
    6  Bart?omiej Piotrowski 
    6  Erich Eckner 
    7  Kyle Keen 
    8  Evangelos Foutras 
    8  Maxime Gauduin 
    9  Jaroslav Lichtblau 
    9  Lukas Fleischer 
   18  BlackEagle 
   21  Sergej Pupykin 
   22  Guillaume ALAUX 
   29  Sven-Hendrik Haase 
   32  Levente Polyak 
   35  Jan Alexander Steffens (heftig) 
   39  Felix Yan 
   43  Jan de Groot 
   44  Andreas Baumann 
   75  Ball? Gy?rgy 
   81  Antonio Rojas 
 6592  Unknown Packager
20577  Erich Eckner 
------------------------------------------------------------------
27707  total

This causes a problem for us in Parabola, because we just started applying the usual upload checks from

db-update

to packages that we import.

  Comments (3)
  Related Tasks (0/0)

bill auger commented on 24.09.2018 05:20

i was trying to make sense of this today and my results were pretty strange

i wrote a script to extract ‘Unknown Packager’ from the db caches in /var/lib/pacman/sync/ and i got very different counts then what luke reported - i ran the same script on both parabola i686 and again on arch32 - it did show that all of the packages by ‘Unknown Packager’ are in [extra], [community], and [core]

<pre>
$ wc -l ./unknown-packagers-parabola
1800 ./unknown-packagers-parabola

$ wc -l ./unknown-packagers-arch32
2250 ./unknown-packagers-arch32
</pre>

one thing i noted was that one of the packages (4ti2) that arch32 has by ‘Unknown Packager’ for v1.6.7 shows as v1.6.9 packaged by ‘Erich Eckner’ on parabola and the parabola package has a higher version number and later build date - maybe my mirrorlist is wonky on arch32 but it is the default one that was installed

<pre>
arch32:
package: 4ti2-1.6.7-1
build date: 1497644502
packager: Unknown Packager

parabola:
package: 4ti2-1.6.9-1.0
build date: 1536224793
packager: Erich Eckner
</pre>

the script i used is attached to the corresponding parabola bug report
https://labs.parabola.nu/issues/1652 https://labs.parabola.nu/attachments/444/find-unknown-packagers Admin
Erich Eckner commented on 24.09.2018 08:47

The buildmaster now rejects package built by “Unknown Packager” - what remains is to rebuild all packages, that are already committed.
Admin
Erich Eckner commented on 24.09.2018 08:54

btw: operating on the package database is much faster than extracting every single package:

find . -name ‘*.db.tar.gz’ | \

while read -r repo; do
  tar -Oxzf "${repo}" --wildcards '*/desc' | \
    sed -n '/^%FILENAME%$\|^%PACKAGER%$/{N;s/\n/ /;p;d;}'
done | \
sed '/^%FILENAME% /{N;/\n%PACKAGER% Unknown Packager$/!d;s/\n.*//;s/^%FILENAME% //;}'

Google Cache

26.09.2018 - using openbox (not gnome) $ gnome-terminal # Couldn’t register with accessibility bus: Did not receive a reply. Possible causes include: the …

no task description

Attached to Project: Archlinux32
Opened by Raphael Scholer - 11.11.2018
Last edited by Erich Eckner - 14.11.2018
 FS#56  - libreoffice-fresh 6.1.3-1.0 in [extra] is not build against packages in it

libreoffice-fresh seems not to be build against packages in [extra], but rather against packages in [*testing].

Running libreoffice produces the following error message:

/usr/lib/libreoffice/program/soffice.bin: error while loading shared libraries: liborcus-0.14.so.0: cannot open shared object file: No such file or directory

extra/liborcus is 0.13.4-3.1
testing/liborcus is 0.14.1-1.0
Closed by Erich Eckner
14.11.2018 18:05
Reason for closing: Fixed

  Comments (1)
  Related Tasks (0/0)

Admin
Erich Eckner commented on 12.11.2018 10:37

I moved liborcus and libixion from testing to extra - now it should work.

Google Cache

Attached to Project: Archlinux32
Opened by Andreas Baumann - 10.01.2019
Last edited by Andreas Baumann - 10.01.2019
 FS#59  - unison segfault

Calling unison, a file synchronizer written in Ocaml. Segfaults.
Most likely a string optimization inside Ocaml itself or a non-matching
library somewhere (recompile ocaml?)

(gdb) bt
#0 0x006782df in caml_string_equal ()
#1 0x005ce0d8 in camlUpdatearchiveMode_802888 ()
#2 0x005ce6e5 in camlUpdatesetArchiveData_802926 ()
#3 0x005da668 in camlGlobalsfun_802549 ()
#4 0x005fed99 in camlLwt_utilmap_1080 ()
#5 0x005fedac in camlLwt_utilmap_1080 ()
#6 0x005da5b9 in camlGlobalsallRootsMap_301891 ()
#7 0x005cea7f in camlUpdateloadArchives_802958 ()
#8 0x005d4637 in camlUpdatefindUpdatesOnPaths_2603478 ()
#9 0x005b36f5 in camlUitextsynchronizeOnce_1102725 ()
#10 0x005b4041 in camlUitextloop_1203129 ()
#11 0x005b41d6 in camlUitextsynchronizeUntilDone_1203134 ()
#12 0x005b442e in camlUitextstart_1303143 ()
#13 0x005acc54 in camlMainBody_401249 ()
#14 0x005ac19d in camlLinktextentry ()
#15 0x005a8220 in caml_program ()
#16 0x0068b64d in caml_start_program ()
#17 0x0068b9f1 in caml_startup_common ()
#18 0x0068ba6a in caml_startup ()
#19 0x005a7c25 in main ()

Closed by Andreas Baumann
10.01.2019 14:15
Reason for closing: Not a bug
Additional comments about closing:

Cannot reprodduce but on a single machine. Might as well be something completly different..

  Comments (6)
  Related Tasks (0/0)

Admin
Erich Eckner commented on 10.01.2019 08:31

easy trials first: let’s recompile ocaml
Admin
Andreas Baumann commented on 10.01.2019 09:30

Some things seem to be wrong here:

Warning 3: deprecated: Stdlib.String.capitalize
Use String.capitalize_ascii instead.
File “/build/unison/src/unison-2.51.2/build/src/uigtk2.ml”, line 1:
Error: /usr/lib/ocaml/lablgtk2/pango.cmi
is not a compiled interface for this version of OCaml.
It seems to be for an older version of OCaml.
make[1]: * [Makefile.OCaml:423: uigtk2.cmx] Error 2
make[1]: Leaving directory ‘/build/unison/src/unison-2.51.2/build/src’ make: * [Makefile:14: src] Error 2

Admin
Andreas Baumann commented on 10.01.2019 09:56

rebuilding lablgtk2 now.
Admin
Andreas Baumann commented on 10.01.2019 11:29

rebuilding unison didn’t help.
Admin
Andreas Baumann commented on 10.01.2019 11:35

Ok, let’s rebuild ocaml itself.
Admin
Andreas Baumann commented on 10.01.2019 12:51

works on two vms (libvirt and Virtualbox), but not on my real eeepc.

Google Cache

Attached to Project: Archlinux32
Opened by bill auger - 13.05.2019
Last edited by Andreas Baumann - 02.10.2019
 FS#73  - [gajim] fails to launch when (optional) gupnp-igd is installed

$ gajim
No translations found
Dirs searched: [PosixPath(’/usr/local/share’), PosixPath(’/usr/share’), PosixPath(’/usr/share/gdm’), PosixPath(’/var/lib/menu-xdg’)]

(gajim:1588): dbind-WARNING : 15:14:29.481: Couldn’t connect to accessibility bus: Failed to connect to socket /tmp/dbus-Nv1LTXmb8O: Connection refused (gajim:1588): WARNING **: 15:14:29.534: Failed to load shared library ‘libgupnp-igd-1.0.so.4’ referenced by the typelib: libgupnp-1.2.so.0: cannot open shared object file: No such file or directory
/usr/lib/python3.7/site-packages/gajim/common/app.py:281: Warning: cannot retrieve class for invalid (unclassed) type ‘void’

gupnp_igd = GUPnPIgd.SimpleIgd()

Traceback (most recent call last):

File "/usr/lib/python3.7/site-packages/gajim/application.py", line 185, in _startup
  app.detect_dependencies()
File "/usr/lib/python3.7/site-packages/gajim/common/app.py", line 281, in detect_dependencies
  gupnp_igd = GUPnPIgd.SimpleIgd()

TypeError: could not get a reference to type class
Traceback (most recent call last):

File "/usr/lib/python3.7/site-packages/gajim/application.py", line 220, in _activate
  self.interface = Interface()
File "/usr/lib/python3.7/site-packages/gajim/gui_interface.py", line 2615, in __init__
  cfg_was_read = parser.read()
File "/usr/lib/python3.7/site-packages/gajim/common/optparser.py", line 86, in read
  self.update_config(old_version, new_version)
File "/usr/lib/python3.7/site-packages/gajim/common/optparser.py", line 158, in update_config
  caps_cache.capscache.initialize_from_db()

AttributeError: ‘NoneType’ object has no attribute ‘initialize_from_db’ Traceback (most recent call last):

File "/usr/lib/python3.7/site-packages/gajim/application.py", line 269, in do_shutdown
  app.logger.commit()

AttributeError: ‘NoneType’ object has no attribute ‘commit’

Closed by Andreas Baumann
02.10.2019 19:36
Reason for closing: Fixed
Additional comments about closing:

Seems to work on pentium4/stable. Feel free to reopen if you it crashes
for you..

Google Cache

no task description

Attached to Project: Archlinux32
Opened by Andreas Baumann - 21.09.2019
Last edited by Andreas Baumann - 21.09.2019
 FS#87  - firefox doesn’t rebuild

OOM when build gkrust:

patching out debug_level=2 in moz.configure/toolchain.configure

thanks BoidLinux for the hint.

  Comments (3)
  Related Tasks (0/0)

Admin
Andreas Baumann commented on 21.09.2019 17:05

A test is executed testing the generated Gecko rust library (a static library), whether
it contains references to networking functions, this fails on IA32:

https://hg.mozilla.org/mozilla-central/rev/47cd8671b12c

$(call py_action,check_binary,–target –networking $@)

Solution is just not to execute this check in config/makefiles/rust.mk
Admin
Andreas Baumann commented on 21.09.2019 17:08

187:52.88 toolkit/library/libxul.so
190:00.47 readelf: Error: Unable to seek to 0x801db328 for section headers
190:01.36 out of memory allocating 3761314508 bytes after a total of 1595923192 bytes
190:01.47 File “/build/firefox/src/firefox-69.0.1/python/mozbuild/mozbuild/action/check_binary.py”, line 361, in

=&gt; python/readelf combo reads whole generated libxul.so into memory in order to do
some ELF fidding.

190:01.48 make[4]: *** [/build/firefox/src/firefox-69.0.1/config/rules.mk:661: libxul.so] Error 1

shows us:

$(SHARED_LIBRARY): $(OBJS) $(RESFILE) $(RUST_STATIC_LIB) $(STATIC_LIBS) $(EXTRA_DEPS) $(GLOBAL_DEPS)

      $(REPORT_BUILD)

ifndef INCREMENTAL_LINKER

      $(RM) $@

endif

      $(MKSHLIB) $($@_$(OBJS_VAR_SUFFIX)) $(RESFILE) $(LDFLAGS) $(STATIC_LIBS) $(RUST_STATIC_LIB) $(SHARE
      $(call py_action,check_binary,--target $@)

Trying to comment that out too (the python check).
Admin
Andreas Baumann commented on 22.09.2019 07:46

Now libxul.so misses some libraries, like libmozsandbox.so, liblgpllibs.so, … The libraries exists, but most likely libxul.so cannot find them (rpath ELF fiddling?).

As a workaround I’ll set LD_LIBRARY_PATH=/usr/lib/firefox in the startup wrapper
scripts /usr/bin/firefox.

Google Cache

no task description

no task description

no task description

Attached to Project: Archlinux32
Opened by Markus Schaaf - 22.10.2019
 FS#92  - nfs mounts with sec=krb5 fail with ‘stale file handle’ for everyone but root

rpc.gssd bug due to missing syscall (setgroups). See https://bugzilla.linux-nfs.org/show_bug.cgi?id=340 `journalctl –unit=rpc-gssd.service` shows:

WARNING: unable to drop supplimentary groups!
WARNING: failed to change identity: Function not implemented

Until a fixed upstream release, use
$ cat abs/packages/nfs-utils/trunk/setgroups32.patch
— nfs-utils-2.4.1/utils/gssd/gssd_proc.c.orig 2019-10-22 11:26:48.059877484 +0200
+++ nfs-utils-2.4.1/utils/gssd/gssd_proc.c 2019-10-22 11:28:03.874553996 +0200
@@ -437,7 +437,7 @@

    int res;

    /* drop list of supplimentary groups first */

- if (syscall(SYS_setgroups, 0, 0) != 0) {
+ if (syscall(SYS_setgroups32, 0, 0) != 0) {

            printerr(0, "WARNING: unable to drop supplimentary groups!");
            return errno;
    }

Google Cache

looks, like the new devtools do not strip packages anymore - maybe it was only an intermediate version, that lacked automatic stripping.

Installing wireguard-dkms fails with:

(2/2) Install DKMS modules
==> dkms install wireguard/0.0.20200318 -k 5.5.8-arch1-1.0
Error! Bad return status for module build on kernel: 5.5.8-arch1-1.0 (i686)
Consult /var/lib/dkms/wireguard/0.0.20200318/build/make.log for more information.

The file /var/lib/dkms/wireguard/0.0.20200318/build/make.log contains the following messages:

DKMS make.log for wireguard-0.0.20200318 for kernel 5.5.8-arch1-1.0 (i686)
Fri 03 Apr 2020 10:11:21 PM CEST
make: Entering directory '/usr/lib/modules/5.5.8-arch1-1.0/build'
  AR      /var/lib/dkms/wireguard/0.0.20200318/build/built-in.a
  CC [M]  /var/lib/dkms/wireguard/0.0.20200318/build/main.o
cc1: error: incompatible gcc/plugin versions
cc1: error: fail to initialize plugin ./scripts/gcc-plugins/structleak_plugin.so
make[1]: *** [scripts/Makefile.build:266: /var/lib/dkms/wireguard/0.0.20200318/build/main.o] Error 1
make: *** [Makefile:1693: /var/lib/dkms/wireguard/0.0.20200318/build] Error 2
make: Leaving directory '/usr/lib/modules/5.5.8-arch1-1.0/build'

The packages for syslog-ng since version 3.27.1-1.0 appears to have been built without systemd support. Although their PKGBUILD runs configure with –enable-systemd, running syslog-ng –version shows Enable-Systemd: off. The official Arch x86_64 package (at version 3.28.1-1) doesn’t exhibit this problem.

I tried building the package locally on i686 using the official Arch PKGBUILD and could not reproduce the problem. My locally built version correctly finds libsystemd during configure, and gets built with systemd support. This leads me to believe the problem happens somewhere in the Archlinux32 build process.

The practical upshot of the problem is that the systemd unit file supplied with syslog-ng lists it as a Type=notify service, but when it fails to notify systemd after startup, systemd eventually kills it and restarts it repeatedly:

Jul 24 07:32:31 wolfie systemd[1]: Starting System Logger Daemon "default" instance...
Jul 24 07:32:32 wolfie syslog-ng[18977]: syslog-ng starting up; version='3.28.1'
Jul 24 07:34:01 wolfie systemd[1]: syslog-ng@default.service: start operation timed out. Terminating.
Jul 24 07:34:01 wolfie syslog-ng[18977]: syslog-ng shutting down; version='3.28.1'
Jul 24 07:34:01 wolfie systemd[1]: syslog-ng@default.service: Failed with result 'timeout'.
Jul 24 07:34:01 wolfie systemd[1]: Failed to start System Logger Daemon "default" instance.
Jul 24 07:34:02 wolfie systemd[1]: syslog-ng@default.service: Scheduled restart job, restart counter is at 1.
Jul 24 07:34:02 wolfie systemd[1]: Stopped System Logger Daemon "default" instance.
Jul 24 07:34:02 wolfie systemd[1]: Starting System Logger Daemon "default" instance...

# pacman -Sy lxde
reboot and start lxsession
lxpanel will crash immediately, with this error in dmesg

pcmanfm[501]: segfault at 2d ip b74823cf sp bf9dd0d0 error 4 in libpango-1.0.so.0.4600.0[b7466000+29000]

lxpanel does not have this problem with GTK3
# pacman -Sy lxde-gtk3
works as expectd

start leafpad
everything looks fine
type or paste anything and it segfaults with this error in dmesg

leafpad[550]: segfault at 1 ip b77adf13 sp bfd0eaec error 4 in libgobject-2.0.so.0.6400.3[b7787000+2f000]

gnome-shell segfaults in libmutter.

                                                               Stack trace of thread 1009:
                                                               #0  0x00000000b64d5436 n/a (libmutter-cogl-6.so.0 + 0xe436)
                                                               #1  0x00000000b64d809c n/a (libmutter-cogl-6.so.0 + 0x1109c)
                                                               #2  0x00000000b653edb9 n/a (libmutter-cogl-6.so.0 + 0x77db9)
                                                               #3  0x00000000b64f8fc5 cogl_flush (libmutter-cogl-6.so.0 + 0x31fc5)
                                                               #4  0x00000000b6549154 cogl_onscreen_swap_buffers_with_damage (libmutter-cogl-6.so.0 + 0x82154)
                                                               #5  0x00000000b6c59f73 n/a (libmutter-clutter-6.so.0 + 0x10ff73)
                                                               #6  0x00000000b6c53cf2 n/a (libmutter-clutter-6.so.0 + 0x109cf2)
                                                               #7  0x00000000b6c54747 n/a (libmutter-clutter-6.so.0 + 0x10a747)
                                                               #8  0x00000000b6bf52ad n/a (libmutter-clutter-6.so.0 + 0xab2ad)
                                                               #9  0x00000000b7703bde g_main_context_dispatch (libglib-2.0.so.0 + 0x4dbde)
                                                               #10 0x00000000b7752962 n/a (libglib-2.0.so.0 + 0x9c962)
                                                               #11 0x00000000b7702299 g_main_loop_run (libglib-2.0.so.0 + 0x4c299)
                                                               #12 0x00000000b697f216 meta_run (libmutter-6.so.0 + 0xb9216)
                                                               #13 0x000000000043144e n/a (gnome-shell + 0x244e)
                                                               #14 0x00000000b7cd900e __libc_start_main (libc.so.6 + 0x1f00e)
                                                               #15 0x0000000000431635 n/a (gnome-shell + 0x2635)

The ArchLinux32 mirror at http[s]://mirror.math.princeton.edu/pub/archlinux32/ appears to have stopped syncing properly in October. I contacted web@math.princeton.edu on January 1 to alert them to the problem, and received the following response from a Princeton sysadmin:

Here is the configuration I am using:

archlinux32:
   remotesrc: “rsync://mirror.archlinux32.org/archlinux32/”
   localdest: “/var/www/html/pub/archlinux32”
   extraopts: “–port=22873”

Looks like at some point this stopped working from upstream:

Raw standard error:
rsync: failed to connect to mirror.archlinux32.org (85.10.198.216): Connection refused (111)
rsync: failed to connect to mirror.archlinux32.org (2a01:4f8:a0:5264::2): Network is unreachable (101)
rsync error: error in socket IO (code 10) at clientserver.c(125) [Receiver=3.1.2]

Can you recommend a better upstream target I can use? And if there are additional constraints on checkin time or frequency.

I informed him that I am not an official member of the ArchLinux32 team, but suggested that using the regular rsync port of 873 might work as mirror.archlinux32.org appears to accept connections to that port from the Internet. As of the time of this writing, it appears that the mirror is now again up to date (although it has other issues returning HTTP error 403 for some of its files), but I promised to bring his additional questions to the official team.

So, to iterate, is this the appropriate target to mirror, or should they use some other parameters?

Thank you.

In my quest to find a working mirror close to me, I’ve come across another issue with at least a couple of mirrors. Some mirrors respond with a HTTP 403 Forbidden for certain files, notably many of the Pacman database files. At the time of this writing, I’m aware of this problem happening with the princeton.edu and the clarkson.edu mirrors. I’ve been in contact with the admin of one of the mirrors, and they’ve pointed out that the permission bits are set to 0600 at the source. If the primary mirror could have their files be world readable, it would likely fix this problem automatically across more than one mirror.

Is that something we can do?

Steps to verify:

% rsync -av rsync://mirror.archlinux32.org/archlinux32/i686/extra/extra.db.tar.gz .
receiving incremental file list
extra.db.tar.gz

sent 43 bytes  received 2,044,453 bytes  371,726.55 bytes/sec
total size is 2,043,851  speedup is 1.00

% ls -l extra.db.tar.gz
-rw------- 1 user user 2043851 Jan  8 11:51 extra.db.tar.gz

libhandy0: /usr/share/vala/vapi/libhandy-0.0.deps exists in filesystem (owned by libhandy)
libhandy0: /usr/share/vala/vapi/libhandy-0.0.vapi exists in filesystem (owned by libhandy)

:: installing mutter (3.38.1-1.1) breaks dependency ‘libmutter-6.so=0-32’ required by gnome-shell

breaks in more SIMD optimizations.

root@arch32-stable-pentium4 ~]# iotop
No module named 'iotop'
To run an uninstalled copy of iotop,
launch iotop.py in the top directory

iotop is a Python script, presumably some Python modules are missing.

newsboat: error while loading shared libraries: libjson-c.so.4: cannot open shared object file: No such file or directory

either building 1.49 via 1.48, via 1.47 or we have to bootstrap from a binary version again.

:: removing user-manager breaks dependency ‘user-manager’ required by plasma-meta

no task description