From owner-freebsd-hackers Sat May 6 04:16:30 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA05038 for hackers-outgoing; Sat, 6 May 1995 04:16:30 -0700 Received: from news.iadfw.net (news.iadfw.net [204.178.72.99]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id EAA05032 for ; Sat, 6 May 1995 04:16:28 -0700 Received: (from jbryant@localhost) by news.iadfw.net (8.6.11/8.6.9) id GAA02679; Sat, 6 May 1995 06:13:50 -0500 From: Jim Bryant Message-Id: <199505061113.GAA02679@news.iadfw.net> Subject: Re: problems with 0412 To: davidg@Root.COM Date: Sat, 6 May 1995 06:13:48 -0500 (CDT) Cc: freebsd-hackers@FreeBSD.org In-Reply-To: <199505050620.XAA00226@corbin.Root.COM> from "David Greenman" at May 4, 95 11:20:47 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1954 Sender: hackers-owner@FreeBSD.org Precedence: bulk In reply: > To: Jim Bryant > cc: dyson@root.com > Subject: Re: problems with 0412 > From: David Greenman > Date: Thu, 04 May 1995 23:20:47 -0700 > > >i've noticed a few quirks in 0412. > > > >case: swap-bound under continous heavy network load. > > > >[64M RAM, 128M swap space, average 80-100% into swap during peak load > >periods] > > > >after long sessions of swap-bound activities, not all of the vm space is > >being reclaimed, thus requiring reboot. garbage-collection problem? > > How much space? > > -DG in the 20-30% range. also, a few hours after i wrote that, i swapped the 16meggers for 32meggers for 128meg. the kernel seems to only see 64megs of that on boot. it also seems [possible disk problem] at around 50% swap on the secondary scsi controller. this curses lotsa kernel prints, including one about pagein, it has since spontaneously rebooted various times, and required operator intervention twice. i'm sure this has to do with the vm... my personal system had 8 days of uptime with moderate network/disk/cpu activity today. outside of the memory problems right now, it is really looking nice. another interesting thing, since adding a second scsi controller and hanging swap off of both of those drives, top seems to want to do a floating point exception prior to getting the second set of samples. 1). how do i get 128megs working right? I've never had to change the kernel before [bounce buffers was always defined]. The bios and cmos both say 128megs. pentium/pci, aha0, aha1, ed0, etc... i really need some help with this one. i also need sleep. Jim -- All opinions expressed are mine, if you | "I will not be pushed, stamped, think otherwise, then go jump into turbid | briefed, debriefed, indexed, or radioactive waters and yell WAHOO !!! | numbered!" - #1, "The Prisoner" jbryant@server.iadfw.net, System administrator, Internet America