From owner-freebsd-questions Thu Mar 19 21:39:09 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA26894 for freebsd-questions-outgoing; Thu, 19 Mar 1998 21:39:09 -0800 (PST) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from worldint.com (wiretap.worldint.com [209.69.11.19]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA26851 for ; Thu, 19 Mar 1998 21:38:41 -0800 (PST) (envelope-from becca@worldint.com) Received: (from becca@localhost) by worldint.com (8.7.5/8.7.3) id AAA15441; Fri, 20 Mar 1998 00:38:08 -0500 (EST) Date: Fri, 20 Mar 1998 00:38:08 -0500 (EST) From: Becca Anderson To: David Greenman cc: freebsd-questions@FreeBSD.ORG Subject: Re: TCP/IP Problems In-Reply-To: <199803182056.MAA29364@implode.root.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wed, 18 Mar 1998, David Greenman wrote: > >I'm have a machine that is setting on my isp's backbone > >(PP200/128MB/FreeBSD2.2.5/Intel EtherExpress Pro/100B) running apache. > >Connections from users with a fast connection (128K or better) seem fine, > >however, slower users have difficulty getting files > 16k via ftp or > >http. I have tcp_extensions turned off. It looks like the machine sends > >around 10k of data then the connection hangs until the other side times out. > >DNS is working correctly. I'm lost. > > Can you provide the output of a 'netstat -i' on "deepthroat" as well as > the equivilent on the router it is connected to? Also, how many mbuf > clusters have you configured the kernel for? What does 'netstat -m' show? > > -DG > > David Greenman > Core-team/Principal Architect, The FreeBSD Project > Actually, I was wrong about the mbufs in the kernel. I have it configured for 4096. My isp won't give me the netstat from the router. They claimed that they did tests and didn't see any dropped packets to our server. Tonight I did some more test and dumped the backets and got this message (on the receiving machine) when the file transfer stopped at 10k. 23:28:07.039841 O 60 03 0021: tsf1.lv.gle.verio.net > devel.worldint.com: icmp : source quench (DF) (ttl 252, id 53327) Here's the netstat's from the server tonight: Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll fxp0 1500 00.a0.c9.9a.c0.d9 180483 0 186126 0 62241 fxp0 1500 209.69.145.12 209.69.145.130 180483 0 186126 0 62241 fxp0 1500 209.69.145.13 209.69.145.131 180483 0 186126 0 62241 fxp0 1500 209.69.145.13 209.69.145.132 180483 0 186126 0 62241 fxp0 1500 209.69.145.15 209.69.145.157 180483 0 186126 0 62241 lp0* 1500 0 0 0 0 0 lo0 16384 26075 0 26075 0 0 lo0 16384 127 127.0.0.1 26075 0 26075 0 0 77 mbufs in use: 65 mbufs allocated to data 5 mbufs allocated to packet headers 6 mbufs allocated to protocol control blocks 1 mbufs allocated to socket names and addresses 64/128 mbuf clusters in use 265 Kbytes allocated to network (51% in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines Our isp did swap out our network cable and moved us to a different port on the hub to see if that would clear it up, but it didn't change anything. ------------------------------------------------------------------------------ Becca Anderson (becca@worldint.com) "Oh, that feels neat!" :) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message