It seems that the crash is caused by some kind of heap corruption. The backtrace and the ltrace file is no real use as the damage was done somewhat earlier.
The best way to track this down is to make it reproducible on my machine, but for that I probably need your configuration file. Running syslog-ng with unlimited core file limit (ulimit -c unlimited) and mailing me the core file and all the binaries (syslog-ng and everything it depends on) could also help. I'm currently running syslog-ng under valgrind with 4 sources (file, udp, tcp, kernel log), HUPing it every second while also sending some messages. It did not crash so far and the messages of valgrind do not indicate anything serious. So, I'm unable to do anything else right now. -- Bazsi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
