abrt 1.0.9 detected a crash.
Attached file: backtrace
reason: Process /usr/bin/liferea was killed by signal 6 (SIGABRT)
release: Fedora release 12 (Constantine)
How to reproduce: unknown
It crashed 3 times in 2 days. Suspected a bad feed that I'm hit intermittently,
possibly related to Java. But even so, 1. why crash, 2. no way to work around.
Created attachment 417075 [details]
I looked at other 2 traces, the seem pretty much the same: java plugin,
BTW, "/etc/alternatives/../../bin/java" does not exist, never existed
(it's in /usr at the least).
Same here, and it's Java related:
** (liferea:10188): DEBUG: NP_Initialize
** (liferea:10188): DEBUG: NP_Initialize succeeded
/builddir/build/BUILD/icedtea6-1.8/plugin/icedteanp/IcedTeaNPPlugin.cc:1508: thread 0x857c038: Error: Failed to spawn applet viewer: Failed to execute child process "/usr/lib/mozilla/plugins/../../bin/java" (No such file or directory)
*** glibc detected *** liferea: free(): invalid pointer: 0x08a38318 ***
BTW, that path doesn't exist:
ls: cannot access /usr/lib/mozilla/plugins/../../bin: No such file or directory
Not said before, but Fedora 12 here.
Workaround here: https://bugzilla.redhat.com/show_bug.cgi?id=592563#c3
Just disable the browser plugins. You'll get the error, but liferea won't crash.
The "Enable Browser Plugins" checkbox is not marked in my Liferea,
so either the root cause is something else, or we have another bug
of Liferea not obeying the configuration.
The other bug is likely https://bugs.webkit.org/show_bug.cgi?id=39687
Note that the "Component: liferea" is misleading, since like #595466 and #595002 and many more duplicate bug reports this is very likely a bug in WebKitGTK+ and/or the Java plugin.
Just noticed that after webkitgtk-1.2.0-1.fc12.i686 upgrade, liferea seems to work fine again.
I opened a new bug 608180 for F13 where the traceback is a bit different.
Cannot reproduce F12 bugs anymore, closing.