From owner-freebsd-net@FreeBSD.ORG Sat May 30 02:27:03 2015 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 9FDE6D92 for ; Sat, 30 May 2015 02:27:03 +0000 (UTC) (envelope-from gondim@bsdinfo.com.br) Received: from mail.bsdinfo.com.br (mail.bsdinfo.com.br [191.243.120.163]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 51B531F66 for ; Sat, 30 May 2015 02:27:03 +0000 (UTC) (envelope-from gondim@bsdinfo.com.br) Received: from mail.bsdinfo.com.br (mail.bsdinfo.com.br [127.0.0.1]) by mail.bsdinfo.com.br (Postfix) with ESMTP id 8C9A430FC30 for ; Fri, 29 May 2015 23:26:20 -0300 (BRT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=bsdinfo.com.br; h=content-transfer-encoding:content-type:content-type :in-reply-to:references:subject:subject:to:mime-version :user-agent:from:from:date:date:message-id; s=dkim; t= 1432952779; x=1433816780; bh=RRrbIBenaJ0dmBOf3aWajP/H+tSS+BC2oPW IPQPpX/0=; b=cV9nTzTd+Stiglc1eKluKOUWEhT1Wl8394SLZPwMgwivlZJiMBI 87UdvO8ILK749MyLy2re6eCik1/OdQrfBIT+Q/WArkbt5v5j8yoO3dsuSg8CG1l+ B5hXmkYpgasSaZ9S91DJ5Dop5dOTuJlfTDggJgFi6b+D8WBDlxr/XxL4= X-Virus-Scanned: amavisd-new at mail.bsdinfo.com.br Received: from mail.bsdinfo.com.br ([127.0.0.1]) by mail.bsdinfo.com.br (mail.bsdinfo.com.br [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tv7EFld_ypQG for ; Fri, 29 May 2015 23:26:19 -0300 (BRT) Received: from [192.168.10.208] (unknown [186.193.54.69]) by mail.bsdinfo.com.br (Postfix) with ESMTPSA id 6CFBA30FC08 for ; Fri, 29 May 2015 23:26:19 -0300 (BRT) Message-ID: <55691FB6.5090307@bsdinfo.com.br> Date: Fri, 29 May 2015 23:25:58 -0300 From: Marcelo Gondim User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.7.0 MIME-Version: 1.0 To: freebsd-net@freebsd.org Subject: Re: FreeBSD 9.3 - Intel X520-SR2 stops passing packets References: <5560C395.8020807@farrokhi.net> <5564852D.8040008@bsdinfo.com.br> <1FB15FA4-6185-4206-9517-AE9667A1A57C@gmail.com> In-Reply-To: <1FB15FA4-6185-4206-9517-AE9667A1A57C@gmail.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 30 May 2015 02:27:03 -0000 On 28-05-2015 11:14, Guy Helmer wrote: >> On May 26, 2015, at 9:37 AM, Marcelo Gondim wrote: >> >> On 23-05-2015 15:14, Babak Farrokhi wrote: >>> Look at the interrupts per queue. 500,000 is the maximum and it is the >>> reason your interface is not accepting new packets. >>> >>>> Guy Helmer >>>> May 21, 2015 at 6:03 PM >>>> I’ve noticed that there have been reports of problems with Intel >>>> X520-SR2 network interfaces stopping working. I think I’m seeing a >>>> similar issue where the 10Gb interfaces stop receiving traffic >>>> (they’re being used in promiscuous mode to sniff traffic from a tap). >>>> ifconfig shows the interfaces are still active and the links are OK. >>>> ifconfig down/up restores activity. I’ve changed >>>> hw.intr_storm_threshold=8000 but I couldn’t tell if the interrupt >>>> storm threshold had been triggered at the time the interfaces stopped >>>> passing traffic. >>>> >>>> Output from sysctl: >>>> >>>> . . . >> Hi, >> >> I had this problem and one day updated the system 10.1- RELEASE to 10.1- STABLE and the problem stopped. I was one years with this problem and a script running and testing the interface when the interface stopped working I was doing exactly what you did. Today I no longer have that problem anymore. >> >> I'm using 10.1-STABLE r281235 > > Thanks for the indication of success with 10.1-STABLE. I am locked into using FreeBSD 9.x until I can go through the whole integration and acceptance testing cycle for 10.x, so I’m trying to find a solution that works on in 9.x. I have reviewed the diffs between 9.3 and 10.1-STABLE for ixgbe driver and haven’t noticed anything that stands out. > Guy, I may be wrong but I think the change was not the ixgbe driver.