From owner-freebsd-current  Mon May  6 06:48:45 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id GAA13334
          for current-outgoing; Mon, 6 May 1996 06:48:45 -0700 (PDT)
Received: from Root.COM (implode.Root.COM [198.145.90.17])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id GAA13315
          Mon, 6 May 1996 06:48:39 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by Root.COM (8.7.5/8.6.5) with SMTP id GAA06775; Mon, 6 May 1996 06:48:14 -0700 (PDT)
Message-Id: <199605061348.GAA06775@Root.COM>
X-Authentication-Warning: implode.Root.COM: Host localhost [127.0.0.1] didn't use HELO protocol
To: Charles Owens <owensc@enc.edu>
cc: "Matthew N. Dodd" <winter@jurai.net>, freebsd-current@FreeBSD.ORG,
        freebsd-stable@FreeBSD.ORG
Subject: Re: MBUFs leaking? 
In-reply-to: Your message of "Mon, 06 May 1996 09:38:01 EDT."
             <Pine.FBS.3.93.960506091022.1525C-100000@dingo.enc.edu> 
From: David Greenman <davidg@Root.COM>
Reply-To: davidg@Root.COM
Date: Mon, 06 May 1996 06:48:14 -0700
Sender: owner-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>On Mon, 6 May 1996, Matthew N. Dodd wrote:
>
>> On Sun, 5 May 1996, David Greenman wrote:
>> > >to obey the options NMBCLUSTERS and only allocates about 180k to
>> > >mbufs.  I've played with maxusers, NMBCLUSTERS, and others and no effect.
>> > 
>> >    I can't explain this. There must be something interacting with this, like a
>> > user process limit or something. There haven't been any kernel changes that
>> > would affect this.
>> 
>
>I am seeing the same behavior as well!  I also have NMBCLUSTERS set to
>4096 but according to 'netstat -m', only 156k is allocated to mbufs!!  I
>was wondering if I was interpretting it correctly.  Any theories? 
>
>Here's the netstat output:
>
>36 mbufs in use:
>        11 mbufs allocated to data
>        9 mbufs allocated to packet headers
>        12 mbufs allocated to protocol control blocks
>        4 mbufs allocated to socket names and addresses
>8/76 mbuf clusters in use
>156 Kbytes allocated to network (13% in use)

   I didn't mention this before, but the space for the buffers is only
allocated when it is needed. NMBCLUSTERS only sets the _maximum_ amount. In
my comments above, I was assuming that the machines were doing the same
thing in both cases. If you don't see an error on the console about running
out of mb_map space, then I don't think you have a problem.

-DG

David Greenman
Core-team/Principal Architect, The FreeBSD Project