From owner-freebsd-current Sun Nov 8 09:59:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id JAA02636 for freebsd-current-outgoing; Sun, 8 Nov 1998 09:59:43 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from lupo.thebarn.com (lupo.lcse.umn.edu [128.101.182.105]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA02624 for ; Sun, 8 Nov 1998 09:59:29 -0800 (PST) (envelope-from cattelan@lupo.thebarn.com) Received: (from cattelan@localhost) by lupo.thebarn.com (8.9.1/8.9.1) id LAA29080; Sun, 8 Nov 1998 11:58:39 -0600 (CST) Message-Id: <199811081758.LAA29080@lupo.thebarn.com> From: Russell Cattelan MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Sun, 8 Nov 1998 11:58:38 -0600 (CST) To: current@FreeBSD.ORG Cc: "Eivind Eklund John Fieber" , "David E. Cross" Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Subject: Re: The infamous dying daemons bug In-Reply-To: <19981108140935.06929@follo.net> References: <13891.30546.555159.254752@lupo.thebarn.com> <19981108140935.06929@follo.net> X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs Lucid Message-ID: <13893.55958.378493.589621@lupo.thebarn.com> FCC: ~/Mail/sent I went back and checked my outgoing mail archives, I didn't find any messages from january but I did find this one from April. Is this similar to what other people are seeing? From: Russell Cattelan To: freebsd-current@freebsd.org Subject: VM scrambling pages if space is exceeded Date: Sun, 19 Apr 98 12:36:09 CDT This problem has been cropping up with "current" for about the last month. The basic problem shows up with daemon processes coreing sortly after swap space is maxed out. The most common programs: cron, sendmail and recently socks5. The main process keep running but every time it goes to fork a copy, the copy dies. I am running -current (as of Apr 14), built from scratch. Apr 19 12:03:43 lupo /kernel: swap_pager: suggest more swap space: 124 MB Apr 19 12:10:03 lupo /kernel: pid 15046 (cron), uid 0: exited on signal 11 Apr 19 12:14:27 lupo /kernel: pid 15057 (socks5), uid 0: exited on signal 11 Apr 19 12:14:50 lupo /kernel: pid 15059 (socks5), uid 0: exited on signal 11 I wasn't sure if I had bad bits at some point but I have two systems exhibiting the same behavior. -- Russell Cattelan Eivind Eklund writes: > On Fri, Nov 06, 1998 at 04:28:07PM -0600, Russell Cattelan wrote: > > I've been complaining about this problem since January. > > The 'dying daemons bug', not the 'inetd crashes/gives "junk pointer" > messages' bug? > > Let's list the facts we have here: > * Problem first spotted in January > > * Problem occur on my PPro box with the combinations > - 64MB RAM/128MB swap > - 64MB RAM/256 MB swap (much less frequently than with 128MB swap) > - 80MB RAM/256 MB swap (seems more frequent than with > 64MB/256MB, but I have not recorded how it behaves, so I > can't really say) > I've got a PPro 200 (Origin = "GenuineIntel" Id = 0x617 Stepping=7) > * Problem occur on a P200MMX (Origin = "GenuineIntel" Id = 0x543 Stepping=3) > with 96MB RAM and 200MB swap > * Problem has never occured on a P133 (unknown stepping) with 24MB RAM > and 64MB swap. > * Problem occur with both IDE and a bunch of different SCSI cards > (thus it seems we can eliminate the disk system) > > I know of three pieces of hardware I have that I've not yet gotten > confirmation that other doesn't have: > (1) AWE64 soundboard > (2) TV card > (3) Network card - a PCI ed (though I changed to this later than first > seeing the bugs, I think...) > > Eivind. > -- Russell Cattelan cattelan@thebarn.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message