From owner-freebsd-current Fri Feb 5 17:01:47 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA03223 for freebsd-current-outgoing; Fri, 5 Feb 1999 17:01:47 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from uni-sb.de (uni-sb.de [134.96.252.33]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA03217 for ; Fri, 5 Feb 1999 17:01:45 -0800 (PST) (envelope-from rock@wurzelausix.CS.Uni-SB.DE) Received: from cs.uni-sb.de (cs.uni-sb.de [134.96.252.31]) by uni-sb.de (8.9.2/1999010400) with ESMTP id CAA01262; Sat, 6 Feb 1999 02:01:38 +0100 (CET) Received: from wurzelausix (wurzelausix.cs.uni-sb.de [134.96.247.1]) by cs.uni-sb.de (8.9.2/1999010400) with ESMTP id CAA18835; Sat, 6 Feb 1999 02:01:37 +0100 (CET) Received: from vodix.aremorika (vodix [134.96.247.43]) by wurzelausix (8.9.1/wjp/19980821) with ESMTP id CAA20903; Sat, 6 Feb 1999 02:01:37 +0100 (CET) From: "D. Rock" Received: by vodix.aremorika; Sat, 6 Feb 1999 02:01:36 +0100 (MET) Date: Sat, 6 Feb 1999 02:01:36 +0100 (MET) Message-Id: <199902060101.CAA04062@vodix.aremorika> To: current@FreeBSD.ORG, dillon@apollo.backplane.com Subject: Re: Seeing NFS saturation 'loop' when installworld'ing to NFS / and /usr Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Matt wrote: > This is very odd. This is the approximate backtrace that I get > when I throw my test machine into DDB: [..] > What is happening is that I am doing a 'make installworld' on my > test machine with / and /usr NFS V3 mounted R+W. I also have come to the conclusion that NFSv3 is badly broken. I switched all my R/W mounts back to NFSv2 and I am happy again. I can do a full "make relase" with an NFS mounted chroot directory with no problems, while I have seen many problems with v3 mounts. One thing to lock up the system (probably only the filesystem code, I can still switch virtual screens with Alt+Fx, but getty cannot start /usr/bin/login) if I mount NFSv3 with the -l flag. I tried to track down some of the problems doing a network snoop and noticed something interesting: NFSv3 seems to produce more than twice the packets during file write than NFSv2 Is this true. There are many more getattr() calls with NFSv3 than with NFSv2. Since my NFS server exports one filesystem exclusively to one FreeBSD machine (which is a little short on disk space), I also tried some tricks for speedup. I just mounted one big file, vnconfig'd, newfs'd it and mounted it via UFS. But unfortunately the machine panic'd really fast during filesystem activity. My tests on this are 3-4 months old though, I will give it another try. Daniel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message