From owner-freebsd-bugs Mon Jan 1 18:03:00 1996 Return-Path: owner-bugs Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id SAA21507 for bugs-outgoing; Mon, 1 Jan 1996 18:03:00 -0800 (PST) Received: from Root.COM (implode.Root.COM [198.145.90.17]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id SAA21502 for ; Mon, 1 Jan 1996 18:02:57 -0800 (PST) Received: from corbin.Root.COM (corbin [198.145.90.50]) by Root.COM (8.6.12/8.6.5) with ESMTP id SAA00463; Mon, 1 Jan 1996 18:02:56 -0800 Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.7.3/8.6.5) with SMTP id SAA00219; Mon, 1 Jan 1996 18:02:57 -0800 (PST) Message-Id: <199601020202.SAA00219@corbin.Root.COM> To: Heikki Suonsivu cc: freebsd-bugs@freefall.freebsd.org Subject: Re: kern/896: Current quite dead 16th dec sup In-reply-to: Your message of "Tue, 02 Jan 96 03:18:43 +0200." <199601020118.DAA16430@katiska.clinet.fi> From: David Greenman Reply-To: davidg@Root.COM Date: Mon, 01 Jan 1996 18:02:56 -0800 Sender: owner-bugs@FreeBSD.ORG Precedence: bulk > real memory = 16777216 (16384..) > kmem_suballoc: bad status returned 3. > panic: kmem_suballoc > > after this normal rebooting in 15 seconds message. > > >This seems to have been caused by setting > >options "NMBCLUSTERS=8192" > >I don't know why it did not happen before. 8192 did work with mid >november kernels. This can only happen if, because of the various options in your kernel, consumed the entire 256MB of virtual address space. That would be quite incredible if true. It appears that you must have a tremendous amount of junk in your kernel and have a variety of kernel config parameters highly mistuned. In any case, your machine CAN NOT WORK correctly with NMBCLUSTERS=8192. A setting of 8192 can consume up to 16MB of memory just for the network buffers. Since that's all the memory your machine has, well, I think you see the problem. -DG