From owner-freebsd-net@FreeBSD.ORG Fri Dec 11 10:29:33 2009 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6C359106566B for ; Fri, 11 Dec 2009 10:29:33 +0000 (UTC) (envelope-from john@traktor.dnepro.net) Received: from traktor.dnepro.net (roof1.dnepro.net [212.3.111.66]) by mx1.freebsd.org (Postfix) with ESMTP id E0A2E8FC12 for ; Fri, 11 Dec 2009 10:29:32 +0000 (UTC) Received: from traktor.dnepro.net (localhost [127.0.0.1]) by traktor.dnepro.net (8.14.3/8.14.3) with ESMTP id nBBATSBb050761 for ; Fri, 11 Dec 2009 12:29:30 +0200 (EET) (envelope-from john@traktor.dnepro.net) Received: (from john@localhost) by traktor.dnepro.net (8.14.3/8.14.3/Submit) id nBBATSYt050758 for freebsd-net@freebsd.org; Fri, 11 Dec 2009 12:29:28 +0200 (EET) (envelope-from john) Date: Fri, 11 Dec 2009 12:29:28 +0200 From: Eugene Perevyazko To: freebsd-net@freebsd.org Message-ID: <20091211102928.GA40831@traktor.dnepro.net> Mail-Followup-To: freebsd-net@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.3i Subject: How can I find the reason network writes fail with ENOMEM on 7.x? X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Dec 2009 10:29:33 -0000 I'm getting network write failures on a host running 7.2-PRERELEASE (I know, I should update it to STABLE, but I've heard of similar reports on 7-S too) Failures are expressed for example in BIND named[72084]: /usr/src/lib/bind/isc/../../../contrib/bind9/lib/isc/unix/socket.c:1567: unexpected error: named[72084]: internal_send: 192.168.71.91#1049: Cannot allocate memory named[72084]: client 192.168.71.91#1049: error sending response: out of memory and on ssh session spontaneously breaking with "Write failed: Cannot allocate memory" Frequency of those failures clearly correlates with network load for the host, which is mainly doing dummynet and ng_nat. How can I find what to tune in this case? -- Eugene Perevyazko