Yeah, so this issue is mis-titled. It looks like every EOF is followed by:
``` Sep 26 13:09:01 softswitch1b /sbin/kamailio[628]: ALERT: <core> [main.c:740]: handle_sigs(): child process 666 exited normally, status=255 ```
And that is one of the TCP workers. So, for some reason they just keep exiting, every X minutes, but not as the result of a crash.