From owner-freebsd-hackers Fri May 9 01:13:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id BAA19992 for hackers-outgoing; Fri, 9 May 1997 01:13:46 -0700 (PDT) Received: from minor.stranger.com (stranger.vip.best.com [204.156.129.250]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id BAA19984 for ; Fri, 9 May 1997 01:13:42 -0700 (PDT) Received: from dog.farm.org (dog.farm.org [207.111.140.47]) by minor.stranger.com (8.6.12/8.6.12) with ESMTP id BAA03206; Fri, 9 May 1997 01:22:54 -0700 Received: (from dk@localhost) by dog.farm.org (8.7.5/dk#3) id BAA13828; Fri, 9 May 1997 01:14:03 -0700 (PDT) Date: Fri, 9 May 1997 01:14:03 -0700 (PDT) From: Dmitry Kohmanyuk Message-Id: <199705090814.BAA13828@dog.farm.org> To: taob@nbc.netcom.ca (Brian Tao) Cc: freebsd-hackers@freebsd.org Subject: Re: NFS service mysteriously stopping on 2.2-RELENG? Newsgroups: cs-monolit.gated.lists.freebsd.hackers Organization: FARM Computing Association Reply-To: dk+@ua.net X-Newsreader: TIN [version 1.2 PL2] Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article you wrote: > The 2.2-970420-RELENG NFS server I put together for a client is > humming along quite nicely, now that I've done fiddling with it. > However, there is an intermittent problem with the three BSD/OS > clients attached to it. One two occasions in the past week, all three > clients will "lose" the NFS server. > One client is running BSD/OS 2.1 with NFSv2 UDP mounts, and the > other two are BSD/OS 3.0 with NFSv3 TCP mounts. In addition to the > "nfs server not responding" syslog messages, the TCP clients > occasionally log this: > May 7 04:14:26 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.4 > May 7 05:38:45 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.2 > May 7 05:38:45 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.3 > May 7 05:38:45 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.1 > May 7 17:48:24 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.1 > May 7 17:48:24 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.2 > May 7 17:48:24 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.3 > May 7 17:48:24 shell1 kernel: short receive (0/4) from nfs server nfs:/user/.4 hmmm! I repeatedly get the same error on FreeBSD 2,2 clients (some 2.2-GAMMA, some 2.2.1-RELEASE) which are connected to NetApp server (running 4.0.3). (For those now aware of NetApp, it is dedicated NFS server software running on Intel and Alpha hardware, pricey, but well-performing). So, this problem is probably client-specific. (your clients are BSD/OS, so it is probably 4.4BSD specific.) It occurs with regularity, rarely, but constantly. I use NFSv2 mounts. >From looking at the source, it looks like an error while receiving RPC from the server... specifically, in /sys/nfs/nfs_socket.c here is a sample from my logs: # grep nfs /var/log/messages May 8 14:36:55 zen /kernel: nfs send error 32 for server filer:/ May 8 14:36:55 zen /kernel: nfs send error 32 for server filer:/ May 8 16:29:30 zen /kernel: short receive (0/4) from nfs server filer:/ May 8 17:30:55 zen /kernel: short receive (0/4) from nfs server filer:/ May 8 18:46:50 zen /kernel: short receive (0/4) from nfs server filer:/ May 8 19:10:20 zen /kernel: short receive (0/4) from nfs server filer:/ May 8 19:39:03 zen /kernel: short receive (0/4) from nfs server filer:/ May 8 20:05:56 zen /kernel: short receive (0/4) from nfs server filer:/ May 8 20:23:32 zen /kernel: short receive (0/4) from nfs server filer:/ #