From owner-freebsd-current Wed Nov 26 14:07:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id OAA23938 for current-outgoing; Wed, 26 Nov 1997 14:07:40 -0800 (PST) (envelope-from owner-freebsd-current) Received: from home.dragondata.com (toasty@home.dragondata.com [204.137.237.2]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id OAA23932 for ; Wed, 26 Nov 1997 14:07:36 -0800 (PST) (envelope-from toasty@home.dragondata.com) Received: (from toasty@localhost) by home.dragondata.com (8.8.5/8.8.5) id QAA18403 for freebsd-current@freebsd.org; Wed, 26 Nov 1997 16:07:30 -0600 (CST) From: Kevin Day Message-Id: <199711262207.QAA18403@home.dragondata.com> Subject: NFS under 3.0 To: freebsd-current@freebsd.org Date: Wed, 26 Nov 1997 16:07:28 -0600 (CST) X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I mentioned before that I had a 3.0 machine locking up/panic'ing a lot. I wa sgetting lots of 'nfsd send error 55's, after some digging, that seems to be 'no buffer space available'. I don't know if NMBCLUSTERS= even has anything to do with buffering for nfs writes, but bumping that way up fixed my problem. I still get the very rare error 55, but the machine has stayed up 5 days now. One problem was that during heavy nfs traffic, the machine would keep running, but any attempt to read/write to a nfs file/directory would hang the process. I'm not sure what to look at when that happens, but while it was going on, i did a netstat -m. 1028 mbufs in use: 711 mbufs allocated to data 304 mbufs allocated to packet headers 11 mbufs allocated to protocol control blocks 2 mbufs allocated to socket names and addresses 554/1392 mbuf clusters in use 2912 Kbytes allocated to network (42% in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines I'm using 3c509B's in all my machines. I only get nfsd send error 55's on both the machines when the one machine running 3.0 is active. Kevin