From owner-freebsd-hackers Tue Apr 28 04:19:29 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id EAA25171 for freebsd-hackers-outgoing; Tue, 28 Apr 1998 04:19:29 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from phoenix.its.rpi.edu (dec@phoenix.its.rpi.edu [128.113.161.45]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id EAA25158 for ; Tue, 28 Apr 1998 04:19:25 -0700 (PDT) (envelope-from dec@phoenix.its.rpi.edu) Received: from localhost (dec@localhost) by phoenix.its.rpi.edu (8.8.8/8.8.7) with SMTP id HAA15376; Tue, 28 Apr 1998 07:19:17 -0400 (EDT) (envelope-from dec@phoenix.its.rpi.edu) Date: Tue, 28 Apr 1998 07:19:16 -0400 (EDT) From: "David E. Cross" To: "Jordan K. Hubbard" cc: Adrian Filipi-Martin , freebsd-hackers@FreeBSD.ORG Subject: Re: SIGDANGER In-Reply-To: <7364.893741081@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Mon, 27 Apr 1998, Jordan K. Hubbard wrote: > All the SIGDANGER (Will Robinson) signal is meant to do is give a hehe ;) > I don't think the question here is "is this a good idea" - it's a > perfectly reasonable idea and one which has been proposed before. The > question here is really "what are the proposed semantics of this > mechanism?", e.g. how long do you wait from the time you SIGDANGER the > process and actually shoot it down, and what happens if you're also > critically short of resources and don't have much time to wait? > My understanding of this was that it shotguned a set of processes all with SIGDANGER, at least some of them should die.... or,the kernel can tell if a process is ignoring SIGDNAGER (which would be the easiest way for a process to handle it), and would use that as a flag for those to be the highest priority processes. Processes that are using their own handler would be second, and those using the default handler would be first choice. (So don't use SIGDANGER as a signal, per-se, but as a flag to the kernel 'it is OK to kill me'). As per memory overcomitment, I do not see this as an issue, I can still run out of memory even if I am not over commited (say netscape went berkerk or whatever). I think thet in situation where the system is starting to page (or heavily paging) SIGDANER [Will Robinson] would be a usefull tool. -- David Cross To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message