From owner-freebsd-stable@FreeBSD.ORG Tue Aug 5 12:24:44 2008 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E182C1065676 for ; Tue, 5 Aug 2008 12:24:44 +0000 (UTC) (envelope-from jdc@parodius.com) Received: from mx01.sc1.parodius.com (mx01.sc1.parodius.com [72.20.106.3]) by mx1.freebsd.org (Postfix) with ESMTP id CFB508FC0C for ; Tue, 5 Aug 2008 12:24:44 +0000 (UTC) (envelope-from jdc@parodius.com) Received: by mx01.sc1.parodius.com (Postfix, from userid 1000) id B9EE21CC0B1; Tue, 5 Aug 2008 05:24:44 -0700 (PDT) Date: Tue, 5 Aug 2008 05:24:44 -0700 From: Jeremy Chadwick To: "Igor V. Ruzanov" Message-ID: <20080805122444.GB88406@eos.sc1.parodius.com> References: <15CEC87F00BB7B4CA0E904C5FCF05646243E8601@exchangenj1> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.18 (2008-05-17) Cc: Vinny Abello , "freebsd-stable@freebsd.org" , "bind-users@isc.org" Subject: Re: Query timeouts on FreeBSD 7 over network X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 05 Aug 2008 12:24:45 -0000 On Tue, Aug 05, 2008 at 03:04:50PM +0400, Igor V. Ruzanov wrote: >> I've tried with the ULE scheduler and 4BSD and tried with and with >> out PREEMPTION turned on. Nothing makes a difference. > First of all you could try to connect only two machines via cross-over > cable, no any switches between the machines, no any VLANs and so on. > FreeBSD-7.0 works better with ULE-scheduler and kernel should be > preemtive (options PREEMPTION in kernel config). > - what is your kernel config? > >> I'm pretty sure this is related to the OS or the em driver in some way, because if I disable all ICMP rate limiting and run an extended ping from the local firewall, I experience a very low amount of random packet loss in no pattern, unlike if you have the ICMP rate limiting enabled. I believe -stable just got added to this thread, so I'm not sure if these details were provided prior. My apologies if this stuff has already been dealt with. 1) Are there any messages from the kernel about watchdog timeouts or other anomalies pertaining to the network? Look in dmesg. 2) pciconf -lv (only include the Ethernet entries please), vmstat -i and netstat -in output. 3) Try disabling MSI/MSI-X via /boot/loader.conf variables (you'll need to reboot after this): hw.pci.enable_msi="0" hw.pci.enable_msix="0" 4) Disabling TSO on the interface, and in the OS: ifconfig emXX -tso sysctl net.inet.tcp.tso=0 -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB |