From owner-freebsd-current Sat Nov 7 17:14:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA05278 for freebsd-current-outgoing; Sat, 7 Nov 1998 17:14:43 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from ns1.yes.no (ns1.yes.no [195.204.136.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA05273 for ; Sat, 7 Nov 1998 17:14:37 -0800 (PST) (envelope-from eivind@bitbox.follo.net) Received: from bitbox.follo.net (bitbox.follo.net [195.204.143.218]) by ns1.yes.no (8.9.1a/8.9.1) with ESMTP id CAA27718; Sun, 8 Nov 1998 02:14:15 +0100 (CET) Received: (from eivind@localhost) by bitbox.follo.net (8.8.8/8.8.6) id CAA15217; Sun, 8 Nov 1998 02:14:14 +0100 (MET) Message-ID: <19981108021413.48676@follo.net> Date: Sun, 8 Nov 1998 02:14:13 +0100 From: Eivind Eklund To: John Fieber , Alexander Litvin Cc: Brian Feldman , current@FreeBSD.ORG Subject: Re: The infamous dying daemons bug References: <199811071314.PAA23544@grape.carrier.kiev.ua> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89.1i In-Reply-To: ; from John Fieber on Sat, Nov 07, 1998 at 09:16:27AM -0500 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sat, Nov 07, 1998 at 09:16:27AM -0500, John Fieber wrote: > On Sat, 7 Nov 1998, Alexander Litvin wrote: > > > In article you wrote: > > BF> I haven't had this bug. And a 3.0 box at school doesn't have it either... > > BF> it's not as common as you think. > > > > Have you beaten your systems to "swap_pager: out of swap space"? > > The meaning of this messages is reasonably clean and I have seen > it before (although not since upgrading to 3.0 and meeting the > dying daemons bug), but I have on occasion seen: > > smap_pager: suggest more swap space: 125 MB > > I assume it is a warning that I'm about to run out, but what does > the 125MB mean? I believe (and this is not confirmed) that this mean that you are overcommitting 125MB of memory - that is, there are 125MB of allocated pages that don't have backing swap space if you dirty them. Looking at sys/vm/swap_pager.c, it looks like this guess is correct - but there is a 2x multiplier in there that seems to mean it is suggesting 2x the amount of overcommit. This is the relevant code-fragment: if (!suggest_more_swap && (vm_swap_size < btodb(cnt.v_page_count * PAGE_ SIZE))) { printf("swap_pager: suggest more swap space: %d MB\n", (2 * cnt.v_page_count * (PAGE_SIZE / 1024)) / 1000); suggest_more_swap = 1; } Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message