From owner-freebsd-current  Mon Nov 16 09:16:21 1998
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id JAA03077
          for freebsd-current-outgoing; Mon, 16 Nov 1998 09:16:21 -0800 (PST)
          (envelope-from owner-freebsd-current@FreeBSD.ORG)
Received: from resnet.uoregon.edu (resnet.uoregon.edu [128.223.144.32])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id JAA03071
          for <current@FreeBSD.ORG>; Mon, 16 Nov 1998 09:16:19 -0800 (PST)
          (envelope-from dwhite@resnet.uoregon.edu)
Received: from localhost (dwhite@localhost)
          by resnet.uoregon.edu (8.8.8/8.8.8) with ESMTP id JAA16489;
          Mon, 16 Nov 1998 09:15:53 -0800 (PST)
          (envelope-from dwhite@resnet.uoregon.edu)
Date: Mon, 16 Nov 1998 09:15:52 -0800 (PST)
From: Doug White <dwhite@resnet.uoregon.edu>
To: "Eugene M. Kim" <astralblue@usa.net>
cc: FreeBSD-current Mailing List <current@FreeBSD.ORG>
Subject: Re: NFS memory leak
In-Reply-To: <Pine.BSF.4.02A.9811150727510.9796-100000@gw1.pl.cp>
Message-ID: <Pine.BSF.4.03.9811160902090.4440-100000@resnet.uoregon.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-current@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

On Sun, 15 Nov 1998, Eugene M. Kim wrote:

> I was trying to installworld yesterday when the bug popped out.
> 
> First, suddenly the NFS server stopped to respond, but didn't panic.
> After some work I found that the server could not send any IP packets;
> sendto() (which ping uses) returned an error with ENOBUF.  Then I
> rebooted the server, and ran ``systat -vm 1'' and ``ping CLIENT''. The
> ``free'' (between ``cache'' and ``daefr'') remained stable for a while,
> but as soon as I started to installworld on the client, the figure
> suddenly began to erode, and when it finally almost hit zero (I think it
> was 40 or something) -- ping stopped working.
> 
> Is this a known bug?  I don't remember this reported/discussed on
> -current. Throw me a stone if it's been. :-)

ENOBUF implies a couple of things:

1) bad route to the client so that packets are getting backed up waiting
for somewhere to go.
2) Incorrect interrupt configuration on your network card.

NFS has problems, but my gut feeling is that it's not the sole culprit
here.

Doug White                               
Internet:  dwhite@resnet.uoregon.edu    | FreeBSD: The Power to Serve
http://gladstone.uoregon.edu/~dwhite    | www.freebsd.org


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message