On Monday 07 January 2013, you wrote: > Hi. > > On Sun, Dec 23, 2012 at 04:09:48PM +0100, Stefan Fritsch wrote: > > It always happens after there has been an graceful reload (i.e. > > /etc/init.d/apache2 reload), so it will break after every > > logrotate (therefore bumping severity). > > FWIW, I think #691318 is probably a duplicate of #694473, as I can > reproduce systematically a crash in an apache subprocess when > editing a mediawiki page right after a reload. > > However, I've been unable to produce core dumps (see my previous > message to #691318) to confirm for sure. > > In any case, thanks for the hint.
Strange, I can't remember having the same problem. You could try setting /proc/sys/fs/suid_dumpable to 1. But this is a security issue on a production server, read the proc man page to see if it is acceptable. Also, look into /etc/apache2 for coredumps, too. There is a short time period during restart where the core dump directory will be set to the compiled-in server root directory (I think this is a bug). -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
