for what it's worth, here's a short snippet from #nautilus discussing the status of this bug with upstream:
10:51 < fatal> alex: has there been any news on http://mail.gnome.org/archives/gvfs-list/2008-October/msg00020.html ? 10:51 < alex> hpj is supposed to be trying to track it down 10:51 < fatal> It looks similar to http://bugs.debian.org/496269 .... note the assertion failure. 10:51 < alex> Seems to be some kind of race or memory corruption bug in the fuse backend 10:52 < alex> Its kinda hard to debug since we can't reproduce it easily though 10:52 < alex> Maybe i should just try amarok 10:53 < fatal> I haven't reproduced it either, but he amarok way that is used by the debian bug reporter might be a good stresstest. :) 10:53 < alex> that seems to be what alexey use too 10:54 < alex> fatal: your comment at the end is correct 10:54 < alex> fatal: and in fact, such a error != NULL check is in svn 10:54 < alex> fatal: but that merely makes us run a short while more 10:55 < alex> file->data_stream shouldn't be NULL or invalid while the file lives though (set on initialize, unref:ed on finalize) 10:55 < alex> so, we believe there is either a corruption bug 10:55 < alex> or the file is unreffed in another thread while the read is running in another thread. 10:57 * alex finds amarok2 src rpm 12:19 < alex> man, i needed to build a new mysql to get amarok2 going 12:20 < alex> the mysql test suite is run on build.... 12:20 < alex> Its... not... fast... -- Andreas Henriksson -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
