From owner-freebsd-questions Wed Aug 12 08:52:19 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id IAA15389 for freebsd-questions-outgoing; Wed, 12 Aug 1998 08:52:19 -0700 (PDT) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id IAA15384 for ; Wed, 12 Aug 1998 08:52:17 -0700 (PDT) (envelope-from root@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with ESMTP id IAA26465; Wed, 12 Aug 1998 08:52:13 -0700 (PDT) Message-Id: <199808121552.IAA26465@implode.root.com> To: josh cc: freebsd-questions@FreeBSD.ORG Subject: Re: kmem_map too small In-reply-to: Your message of "Wed, 12 Aug 1998 09:33:43 CDT." <35D1A7C7.5A44E4E@lucratec.com> From: David Greenman Reply-To: dg@root.com Date: Wed, 12 Aug 1998 08:52:13 -0700 Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG >http://www.freebsd.org/FAQ/FAQ83.html#83 > >On this url in order to avoid this PANIC addressed in my subject, I >upped my NMBCLUSTERS to '25000' > >Yes, 25,000. On a machine with 1 gig of ram. > >Using netstat -m I see this: > >4747 mbufs in use: > 3174 mbufs allocated to data > 1563 mbufs allocated to packet headers > 7 mbufs allocated to protocol control blocks > 3 mbufs allocated to socket names and addresses >3083/3640 mbuf clusters in use >7873 Kbytes allocated to network (85% in use) >0 requests for memory denied >0 requests for memory delayed >0 calls to protocol drain routines > >Usually, 99% is in use, and then my system panics and reboots itself. > >What the hell do I need to set my nmbclusters to have enough mbuf >clusters? > >100,000 nmb clusters???? "kmem_map too small" panic is not an indication of running out of mbuf clusters. It's an indication of running out of malloc space. You need to increase VM_KMEM_SIZE to something larger than the default of 32MB. I'd try doubling it. As to other things you've said above: An in-use percent of 99% does not indicate that you are out of buffers. It's a mostly bogus number that tells you how much is in-use of the peak since the system was started. The peak is the second number in the line above that. In your netstat output above, your system has peaked usage of 3640 mbuf clusters. If this doesn't increase significantly over time, then you will only need to configure perhaps 5000- 6000 mbuf clusters to keep your system happy (always plan for at least 1.5-2*peak). -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message