From owner-freebsd-questions Thu Jun 15 06:45:16 1995 Return-Path: questions-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA16559 for questions-outgoing; Thu, 15 Jun 1995 06:45:16 -0700 Received: from mercury.unt.edu (mercury.unt.edu [129.120.1.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id GAA16549 for ; Thu, 15 Jun 1995 06:45:14 -0700 Received: from gab.unt.edu by mercury.unt.edu with SMTP id AA13579 (5.65c/IDA-1.4.4 for ); Thu, 15 Jun 1995 08:45:01 -0500 Received: from GAB/SpoolDir by gab.unt.edu (Mercury 1.13); Thu, 15 Jun 95 8:45:01 CST6CDT Received: from SpoolDir by GAB (Mercury 1.13); Thu, 15 Jun 95 8:44:59 CST6CDT From: "John Booth" Organization: University of North Texas To: wjw@IAEhv.nl (Willem Jan Withagen) Date: Thu, 15 Jun 1995 08:44:59 CST6CDT Subject: Re: Crashed Cc: questions@freebsd.org Priority: normal X-Mailer: Pegasus Mail v3.22 Message-Id: <86D5001FA9@gab.unt.edu> Sender: questions-owner@freebsd.org Precedence: bulk You need to add the option options "NMBCLUSTERS=1024" to your configuration file and remake/install your kernel. The mb_map doesn't deal with the disks, they are network buffers. On busy nets this option needs to be increased. We have about 20 unix hosts, 4 novell servers, and miscellaneous pcs and 1024 wasn't enough for our net--2048 worked adequately. Each buffer takes up 1k I believe, I think the default configuration will use 512 buffers. You can use netstat -m to see how many buffers are currently in use and the number of requests that were denied (along with other information). If there are any denied requests, you should increase the number of NMBCLUSTERS. Your hard disk is probably fine. If you run out of the buffers it will cause a variety of problems. ---------------------------------------------------------------------- College of Arts & Sciences Computing Services John A. Booth, john@gab.unt.edu