From owner-freebsd-current Mon May 6 16:17:16 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id QAA16726 for current-outgoing; Mon, 6 May 1996 16:17:16 -0700 (PDT) Received: from sierra.zyzzyva.com (ppp0.zyzzyva.com [198.183.2.50]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id QAA16708 Mon, 6 May 1996 16:17:11 -0700 (PDT) Received: (from mail@localhost) by sierra.zyzzyva.com (8.7.5/8.6.11) id SAA09096; Mon, 6 May 1996 18:16:51 -0500 (CDT) Message-Id: <199605062316.SAA09096@sierra.zyzzyva.com> X-Authentication-Warning: sierra.zyzzyva.com: mail set sender to using -f Received: from localhost(127.0.0.1) by sierra via smap (V1.3) id sma009087; Mon May 6 18:16:26 1996 To: davidg@root.com Subject: Re: MBUFs leaking? In-reply-to: davidg's message of Mon, 06 May 1996 13:23:10 -0700. <199605062023.NAA07341@Root.COM> Cc: current@freebsd.org, stable@freebsd.org X-uri: http://www.zyzzyva.com/ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 06 May 1996 18:16:24 -0500 From: Randy Terbush Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >> Alan Cox writes: > >> > >> > I've had a similar problem. However, I hadn't noticed any correlation > >> > with NMBCLUSTERS. Try rolling back your pmap.c to the -RELEASE version. > >> > That "solved" the problem for me. > >> > >> Are you running with the associated new ld.so ? If not, try installing it > >> instead of backing out pmap.c, > >> > >> michael > > > >I assume this hasn't been commited to -stable? > > > >I'm beginning to think that I need to run -current to get around > >much of the instability I am running into.... > > This is the first I've heard about instability problems with the -stable > pmap.c. You're going to need to provide a lot more information before I'll > be able to diagnose the problem. Also make sure you are running with the most > recent version (1.58.4.3). > I can't be sure it is pmap.c. See my followup debug trace on a system crash from another -stable system. As I mentioned in earlier mail, I have a Tyan 133mhz Pentium with 64 MB RAM that will crash if attempting to run X on a kernel with the NMBCLUSTERS=4096 define. This sounds suspiciously like the problem that ... Date: Mon, 6 May 1996 16:16:37 -0400 (EDT) From: Charles Owens To: stable@freebsd.org, freebsd-hackers@freebsd.org Subject: Followup: signal 11 problem solved by yanking DDB! just posted about signal 11's with DDB compiled in. I have also compiled a -g /bin/sh on my system because I was getting coredumps from /bin/sh for no apparent reason compiling some of the ports. Since rebuilding with -g...no dumps... My comment about -current stability stems from my relative newness to the group, and my reading of many Emails from people say "this dongle patch from -current" fixed up my problem...