On Sun, 2017-09-17 at 12:22 +0200, Dominique Dumont wrote: > On mardi 12 septembre 2017 17:24:36 CEST you wrote: > > 00000000 08:02 918537 > > /lib/x86_64-linux-gnu/libgpg-error.so.0.21.0 Thread 4 "pool" > > received > > signal SIGABRT, Aborted. > > [Switching to Thread 0x7fffe3fff700 (LWP 24971)] > > I can't reproduce this crash on my system. This bug looks like this > upstream > report: > > https://bugzilla.gnome.org/show_bug.cgi?id=754921 > > Can you try the workaroung shown there ? > > All the best
Thanks for that idea. Unfortunately the groups files as suggested in
that bug didn't do anything. However... I have found that removing the
article-cache directory causes pan to launch successfully. On selecting
an article, however, it crashed again.
I noticed during a couple of iterations that the same filename, e6e618a
a-387f-d253-4bf4-e9ab27001...@gmail.com.msg, was always present in the
article-cache directory after a crash. This file was one of three
mentioned in tasks.nzb:
<file poster="Ghislain Vaillant < ghisv...@gmail.com >"
date="1505144897" subject="Bug#875432: s/python3-sphinx-intl/sphinx-intl">
<path>/home/sam</path>
<paused>0</paused>
<groups>
<group>gmane.linux.debian.devel.general</group>
<group>gmane.linux.debian.devel.wnpp</group>
</groups>
<segments>
<segment bytes="7932" number="1">
e6e618aa-387f-d253-4bf4-e9ab27001...@gmail.com </segment>
</segments>
</file>
At this point I blew away tasks.nzb and now the crash is gone. The bad
news is that viewing each article in the problematic thread no longer
caused the crash! However, if I restore the problematic tasks.nzb file
then the crash is back.
I've attached that file in case it's useful. I'm unable to reproduce
the crash from a fresh (non-existing) tasks.nzb file, even if I view
each of the articles that are listed in the attached file.
--
Sam Morris <https://robots.org.uk/>
PGP: rsa4096/CAAA AA1A CA69 A83A 892B 1855 D20B 4202 5CDA 27B9
tasks.nzb
Description: application/nzb
signature.asc
Description: This is a digitally signed message part
