From owner-freebsd-stable Sun Apr 5 22:57:44 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA28838 for freebsd-stable-outgoing; Sun, 5 Apr 1998 22:57:44 -0700 (PDT) (envelope-from owner-freebsd-stable@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA28797; Sun, 5 Apr 1998 22:57:26 -0700 (PDT) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id AAA08923; Mon, 6 Apr 1998 00:57:20 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199804060557.AAA08923@dyson.iquest.net> Subject: Re: swap-leak in 2.2.5 ? In-Reply-To: <199804060543.AAA02003@detlev.UUCP> from Joel Ray Holveck at "Apr 6, 98 00:43:18 am" To: joelh@gnu.org Date: Mon, 6 Apr 1998 00:57:20 -0500 (EST) Cc: dyson@FreeBSD.ORG, dswartz@druber.com, dg@root.com, dag-erli@ifi.uio.no, stable@FreeBSD.ORG, current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk > >> My only quibble with this technique is that it would seem to make it > >> harder to tell if your machine is really running low on swap or not > >> (e.g. swap as backing store for stack/heap/whatever *is* critical and > >> allocation failure can cause application failure, whereas swap being > >> used to cache random cruft is in the "who really cares" department). > >> Or is there some way to tell the difference? > > It is difficult not only to tell if you are low on swap, but also it > > is hard to quantify being low on memory. I have been thinking about > > this over the last year or so. > > I don't know much about the FreeBSD VM system, but it is my > understanding that in stock 4.4BSD, you can tell kinda by a high > number of page outs but mostly by swap outs if you are low on RAM. Is > this not the case also with FreeBSD? > Yes, that is about the only way. I think that high (swap) pageout activity along with a small cache queue length would seem to be a good (and sensitive) indicator. We have no utility that informs users simply though. > > I've been considering writing a utility to watch several > performance-related stats (including *why* pages are being paged out, > and processes swapped out, etc), and watch whether MINFREE is being > approached, or what have you. > Very good idea!!! If you need pointers to the info, let me know. Most of the info is available under sysctl vm and sysctl vfs, along with some of the cnt.* variables (which should likely be accessible with sysctl also.) John To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message