From owner-freebsd-questions@FreeBSD.ORG Fri Sep 2 01:02:13 2005 Return-Path: X-Original-To: freebsd-questions@freebsd.org Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 82B8916A41F for ; Fri, 2 Sep 2005 01:02:13 +0000 (GMT) (envelope-from bunnyhero@gmail.com) Received: from wproxy.gmail.com (wproxy.gmail.com [64.233.184.195]) by mx1.FreeBSD.org (Postfix) with ESMTP id B635543D49 for ; Fri, 2 Sep 2005 01:02:12 +0000 (GMT) (envelope-from bunnyhero@gmail.com) Received: by wproxy.gmail.com with SMTP id i4so449957wra for ; Thu, 01 Sep 2005 18:02:11 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=dyFNI6Mko10MHeKAbCqX4X7HhcSuxbeK7kqN3xoG2eRYlaHW4+X6OlXbaXRUAF/pDVaZWoMZDj9xHC/AFYyZI5YLObxAckFlU8vl1yc6T2dhQvUE1myKL/W2LSzfzzkfSVSFJWzwvL7Fy+dGp2wUfVbjBe+UMN3ihBdcTWCGUxA= Received: by 10.54.21.70 with SMTP id 70mr2174376wru; Thu, 01 Sep 2005 18:02:11 -0700 (PDT) Received: by 10.54.14.43 with HTTP; Thu, 1 Sep 2005 18:02:11 -0700 (PDT) Message-ID: <318d047105090118026a72b154@mail.gmail.com> Date: Thu, 1 Sep 2005 21:02:11 -0400 From: bunny hero To: freebsd-questions@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Subject: No buffer space available? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 02 Sep 2005 01:02:13 -0000 Recently my FreeBSD 4.10 server has been experiencing various connection problems under heavy loads-- HTTP requests don't return, connections to the MySQL server fail, and /var/log/messages contains errors like: Aug 31 19:19:16 main sendmail[77493]: j810JGQD077493: SYSERR(www): makeconnection: cannot create socket: No buffer space available Aug 31 19:25:57 main sshd[77700]: error: PAM: failed create sockets: No buffer space available Aug 31 19:53:02 main named[81]: socket(SOCK_RAW): No buffer space available =20 I did some web searching, and read that I should do a netstat -m to see if I am running out of mbufs. But if I understand it correctly, the output of netstat seems to indicate that I am not running out of mbufs? 258/2320/26624 mbufs in use (current/peak/max): 156 mbufs allocated to data 102 mbufs allocated to packet headers 131/938/6656 mbuf clusters in use (current/peak/max) 2456 Kbytes allocated to network (12% of mb_map in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines Meanwhile, output of 'top' and 'vmstat' shows that my load average is usually under 1.0 (even under heavy loads) and that my system is hardly paging anything to disk (pi and po are usually 0)... although I admit I'm not 100% certain how to interpret the rest of vmstat's output or how it might apply. Also, when I do netstat -s -p tcp during heavy traffic, "listen queue overflows" increases at around 50-70 per second. I tried increasing net.inet.ip.intr_queue_maxlen (from 50 to 1024 to 2048) but that doesn't seem to make any difference. I'm not sure what else to look for, or where to go from here. Should I increase NMBCLUSTERS and/or NMBUFS anyway? Do I need more memory? Or perhaps the hardware simply can't handle the number of connections it's handling now? My system is: FreeBSD 4.10-SECURITY FreeBSD 4.10-SECURITY #0: Wed Jun 29 20:49:39 GMT 2005 Hardware: AMD Sempron 2600, 1G RAM, 80GB Any tips, pointers would be greatly appreciated. Thank you very much! wayne a. lee