From owner-freebsd-current@FreeBSD.ORG Sat Oct 6 11:01:51 2007 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 441B716A420 for ; Sat, 6 Oct 2007 11:01:51 +0000 (UTC) (envelope-from askbill@conducive.net) Received: from conducive.net (lindfield.ch [203.194.153.81]) by mx1.freebsd.org (Postfix) with ESMTP id 0EFFC13C45D for ; Sat, 6 Oct 2007 11:01:51 +0000 (UTC) (envelope-from askbill@conducive.net) Received: from cm218-253-81-177.hkcable.com.hk ([218.253.81.177]:64985 helo=pb.local) by conducive.net with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.63 (FreeBSD)) (envelope-from ) id 1Ie7Pq-000A94-0L for freebsd-current@freebsd.org; Sat, 06 Oct 2007 11:01:50 +0000 Message-ID: <47076B1D.3040600@conducive.net> Date: Sat, 06 Oct 2007 07:01:49 -0400 From: =?UTF-8?B?6Z+T5a625qiZIEJpbGwgSGFja2Vy?= User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1.2) Gecko/20070221 SeaMonkey/1.1.1 MIME-Version: 1.0 To: freebsd-current@freebsd.org References: <20071005000046.GC92272@garage.freebsd.pl> In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: ZFS kmem_map too small. X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 06 Oct 2007 11:01:51 -0000 Ivan Voras wrote: > Pawel Jakub Dawidek wrote: > >> What I did was to rsync 200 FreeBSD src trees from one directory to >> another on the same ZFS file system. It worked fine. > > It looks like most problems (including mine I sent you before) are when > rsync (and possibly NFS?) are run over the network. How much kernel > memory does a heavily loaded network stack (multiple parallel > connections & TCP streams) consume? Just for the TCP send & receive > buffers I'd guess at least something like 128K*number_of_connections. > > Is this then hinting that ZFS testing is pointing to problems in the new stack? rsync perhaps exonerated if local disk-to-disk doesn't show it. But what of ssh? Does it happen with other-than-ZFS rsync use at comparable load? Or on the older stack - with zfs or other load? Not 'suggesting' - just questioning. Bill