Date: Wed, 9 Apr 1997 17:51:47 +0000 From: "Jesus A. Mora Marin" <amora@zoom.es> To: lutz@muc.de (Lutz Albers) Cc: questions@freebsd.org Subject: Re: Becoming a daemon the long way... Message-ID: <199704091558.RAA05431@silvester.zoom.es> In-Reply-To: <199704090839.KAA19213@tavari.muc.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, Lutz! Thanks for replying. > I zombie is a process, which has ended, but for which the exit code has not > been fetchted by the parent process. As long as the parent process lives, > so will the zombie. Agree. > If a process exists which has has spawed child > processes, then these processes will be reassigned to process-id 1 > (init). init will (hopefully) read the exit codes for all exited > processes. That's the really interesting point. Theorically every process whose parent exits without waiting for it to die, would be inherited by init. Otherwise the child process will become a zombie when it dies. In my own experience an ICL DS-90 system, running SysVR4, generates a large number of zombies when it goes REALLY short of memory -and probably of CPU-. This is annoying, since you've got your process table plagued with these nasty guys. Again, are there kernel-related conditions -say race conditions, I don't know- that makes a system prone to generate zombies? Has this been observed in FreeBSD? See you! ---- Jesus A. Mora Marin, MD (aka EA7HAC, ex-EC7DVE) Email: amora@zoom.es
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704091558.RAA05431>