From owner-freebsd-questions Fri Aug 13 8:47:15 1999 Delivered-To: freebsd-questions@freebsd.org Received: from scientia.demon.co.uk (scientia.demon.co.uk [212.228.14.13]) by hub.freebsd.org (Postfix) with ESMTP id 74A3414F47 for ; Fri, 13 Aug 1999 08:47:10 -0700 (PDT) (envelope-from ben@scientia.demon.co.uk) Received: from lithium.scientia.demon.co.uk ([192.168.0.3] ident=exim) by scientia.demon.co.uk with esmtp (Exim 3.024 #3) id 11FJ9w-000B2r-00; Fri, 13 Aug 1999 16:22:20 +0100 Received: (from ben) by lithium.scientia.demon.co.uk (Exim 3.024 #3) id 11FJ9t-0003q3-00; Fri, 13 Aug 1999 16:22:17 +0100 Date: Fri, 13 Aug 1999 16:22:17 +0100 From: Ben Smithurst To: Oscar Bonilla Cc: freebsd-questions@FreeBSD.ORG, yurtesen@ispro.net.tr Subject: Re: processes cant be killed? Message-ID: <19990813162217.A14699@lithium.scientia.demon.co.uk> References: <19990811183046.A1967@fisicc-ufm.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <19990811183046.A1967@fisicc-ufm.edu> Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Oscar Bonilla wrote: > kill -9 won't work. Zombie processes cannot be killed. Their address > spaces have been deallocated so they can't get any signals. The processes below are not zombies -- there is no "Z" in the status column. And I would worry about zombie processes if there are a lot of them; don't they take up space in the process table? If so, they could limit the overall number of real processes which can be forked. In some cases, of course, you may be able to get rid of zombie processes by killing their parent process, but in some cases it might not be appropriate to do that. >>>>> Evren Yurtesen 8/8/99 12:39:51 PM >>> >> >> are these that kind of processes? (well I do not have any clue >> about what is a zombie process. Where can I learn it from?) >> >> root 7991 0.0 0.0 45044 0 p1- D> root 8026 0.0 0.0 40088 0 p0- DE - 0:00.00 (ee) This has probably already been answered, but a zombie is a process which has called exit(), but its parent has not yet called wait4() to pick up the corpse and dispose of it in an appropriate fashion. The reason killing its parent will work is that the dead process will lose its parent, and the system will assign process 1 (init) as its parent, which will call wait4() appropriately. But the two above are not zombies, RTFM for ps(1), it will explain what the letters in the status column mean. -- Ben Smithurst | PGP: 0x99392F7D ben@scientia.demon.co.uk | key available from keyservers and | ben+pgp@scientia.demon.co.uk To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message