Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 23 Mar 2000 00:20:32 +0100
From:      Brad Knowles <blk@skynet.be>
To:        Matt Heckaman <matt@ARPA.MAIL.NET>, Christian Jachmann <jachmann@gigabell.net>
Cc:        "stable@freebsd.org" <stable@FreeBSD.ORG>
Subject:   Re: 3.4-Stable crashes..(heavy diskio+networking)
Message-ID:  <v04220804b4ff0562dfc6@[194.78.238.239]>
In-Reply-To: <Pine.BSF.4.21.0003221729320.7534-100000@epsilon.lucida.qc.ca>
References:  <Pine.BSF.4.21.0003221729320.7534-100000@epsilon.lucida.qc.ca>

next in thread | previous in thread | raw e-mail | index | archive | help
At 5:32 PM -0500 2000/3/22, Matt Heckaman wrote:

>  This may sound silly, but have you looked at the value for your
>  NMBCLUSTERS? FreeBSD panics when it runs out of them, as can happen in a
>  high network load situation. I scale mine based on the bandwidth and
>  assume a worst case scenario of a denial of service attack. I've found
>  16,384 good for T1+, moving to double that for 10Mbit, as for beyond that,
>  I've never worked with anything that high.

	At the bottom of his kernel config, he's got:

		options         NMBCLUSTERS=26112

	This would seem to beat your 16384 (which also happens to be the 
value I use in our news peering server running 3.2-RELEASE, although 
we're running Diablo and not INN and we're not doing the volumes of 
traffic he is).

--
   These are my opinions -- not to be taken as official Skynet policy
======================================================================
Brad Knowles, <blk@skynet.be>                || Belgacom Skynet SA/NV
Systems Architect, Mail/News/FTP/Proxy Admin || Rue Colonel Bourg, 124
Phone/Fax: +32-2-706.13.11/12.49             || B-1140 Brussels
http://www.skynet.be                         || Belgium


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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?v04220804b4ff0562dfc6>