Greg Hendershott wrote at 03/26/2011 01:16 PM:
So in parallel to enabling core dumps and so on, I think I'll want to have some other process check every to see if it's still running. If not, launch it again.
In addition to this watchdog process (or cron job), which is a good idea, you also always have the option of having a simple manager process that retires worker processes after a while, whether or not you could tell they need it. There is a proud tradition of doing this even in very stable servers, as a way of being resilient in case of memory leaks or similar problems. It can also make better use of your hardware resources.
-- http://www.neilvandyke.org/ _________________________________________________ For list-related administrative tasks: http://lists.racket-lang.org/listinfo/users
