From owner-freebsd-net Sun Jul 4 9: 0:59 1999 Delivered-To: freebsd-net@freebsd.org Received: from picalon.gun.de (picalon.gun.de [194.77.0.18]) by hub.freebsd.org (Postfix) with ESMTP id DEE3A1539F; Sun, 4 Jul 1999 09:00:42 -0700 (PDT) (envelope-from andreas@klemm.gtn.com) Received: from klemm.gtn.com (pppak04.gtn.com [194.231.123.169]) by picalon.gun.de (8.8.6/8.8.6) with ESMTP id SAA02568; Sun, 4 Jul 1999 18:00:26 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.9.3/8.9.3) id RAA13434; Sun, 4 Jul 1999 17:26:15 +0200 (CEST) (envelope-from andreas) Date: Sun, 4 Jul 1999 17:26:15 +0200 From: Andreas Klemm To: freebsd-net@FreeBSD.ORG Cc: phk@FreeBSD.ORG Subject: NTP configuration problems between FreeBSD-3.2-STABLE and cisco 25xx IOS 11.3(10)T Message-ID: <19990704172613.A12940@titan.klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i X-Operating-System: FreeBSD 3.2-STABLE SMP X-Disclaimer: A free society is one where it is safe to be unpopular Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi ! I don't get a freebsd box running as a NTP server for a cisco router. FreeBSD 3.2-STABLE of yesterday Cisco 2516, IOS 11.3(10)T Does some kind soul have an idea, what's missing ? What does "failed validity tests 20" mean, see debugging output below. Best regards Andreas /// %%%---- Topology cisco | | e0 172.16.2.2 | |--------+--------+----------| 172.16.2.0/24 | | xl0 172.16.2.1 | FBSD-3.2-STABLE | | ed0 172.16.1.1 | |--------+-------------------| 172.16.1.0/24 %%%---- FreeBSD configuration /etc/ntp.conf: server 172.16.2.1 broadcast 172.16.1.255 broadcast 172.16.2.255 enable monitor stats disable auth driftfile /etc/ntp.drift %%%---- cisco configuration ntp server 172.16.2.1 %%%---- cisco ntp status commands / debug output cisco#sh ntp ass address ref clock st when poll reach delay offset disp ~172.16.2.1 0.0.0.0 16 49 64 0 0.0 0.00 16000. * master (synced), # master (unsynced), + selected, - candidate, ~ configured cisco#sh ntp stat Clock is unsynchronized, stratum 16, no reference clock nominal freq is 250.0000 Hz, actual freq is 250.0000 Hz, precision is 2**19 reference time is 00000000.00000000 (02:00:00.000 CEST Mon Jan 1 1900) clock offset is 0.0000 msec, root delay is 0.00 msec root dispersion is 0.00 msec, peer dispersion is 0.00 msec cisco# *Mar 1 02:16:38: NTP: xmit packet to 172.16.2.1: *Mar 1 02:16:38: leap 3, mode 3, version 3, stratum 0, ppoll 64 *Mar 1 02:16:38: rtdel 0000 (0.000), rtdsp 10001 (1000.015), refid 00000000 (0.0.0.0) *Mar 1 02:16:38: ref 00000000.00000000 (02:00:00.000 CEST Mon Jan 1 1900) *Mar 1 02:16:38: org BB29F9D8.0457D000 (17:18:48.016 CEST Sun Jul 4 1999) *Mar 1 02:16:38: rec AF3BD426.D2FCCB94 (02:15:34.824 CEST Mon Mar 1 1993) *Mar 1 02:16:38: xmt AF3BD466.CDF0878F (02:16:38.804 CEST Mon Mar 1 1993) *Mar 1 02:16:38: NTP: rcv packet from 172.16.2.1: *Mar 1 02:16:38: leap 3, mode 4, version 3, stratum 0, ppoll 64 *Mar 1 02:16:38: rtdel 0000 (0.000), rtdsp 13C5B (1235.764), refid 00000000 (0.0.0.0) *Mar 1 02:16:38: ref 00000000.00000000 (02:00:00.000 CEST Mon Jan 1 1900) *Mar 1 02:16:38: org AF3BD466.CDF0878F (02:16:38.804 CEST Mon Mar 1 1993) *Mar 1 02:16:38: rec BB29FA18.04129000 (17:19:52.015 CEST Sun Jul 4 1999) *Mar 1 02:16:38: xmt BB29FA18.041B0000 (17:19:52.016 CEST Sun Jul 4 1999) *Mar 1 02:16:38: inp AF3BD466.D2FD4AE8 (02:16:38.824 CEST Mon Mar 1 1993) *Mar 1 02:16:38: NTP: packet from 172.16.2.1 failed validity tests 20 -- Andreas Klemm http://www.FreeBSD.ORG/~andreas http://www.freebsd.org/~fsmp/SMP/SMP.html powered by Symmetric MultiProcessor FreeBSD Latest song from our band: http://www.freebsd.org/~andreas/mp3/schaukel.mp3 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Sun Jul 4 9: 6:53 1999 Delivered-To: freebsd-net@freebsd.org Received: from critter.freebsd.dk (critter.freebsd.dk [212.242.40.131]) by hub.freebsd.org (Postfix) with ESMTP id 8EBA914BDA for ; Sun, 4 Jul 1999 09:06:51 -0700 (PDT) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (localhost [127.0.0.1]) by critter.freebsd.dk (8.9.3/8.9.2) with ESMTP id SAA12437; Sun, 4 Jul 1999 18:06:15 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: Andreas Klemm Cc: freebsd-net@FreeBSD.ORG Subject: Re: NTP configuration problems between FreeBSD-3.2-STABLE and cisco 25xx IOS 11.3(10)T In-reply-to: Your message of "Sun, 04 Jul 1999 17:26:15 +0200." <19990704172613.A12940@titan.klemm.gtn.com> Date: Sun, 04 Jul 1999 18:06:15 +0200 Message-ID: <12435.931104375@critter.freebsd.dk> From: Poul-Henning Kamp Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <19990704172613.A12940@titan.klemm.gtn.com>, Andreas Klemm writes: >Hi ! > >I don't get a freebsd box running as a NTP server for a cisco router. > >%%%---- FreeBSD configuration > >/etc/ntp.conf: >server 172.16.2.1 >broadcast 172.16.1.255 >broadcast 172.16.2.255 >enable monitor stats >disable auth >driftfile /etc/ntp.drift > >%%%---- cisco configuration You are asking the machine to be it's own server. you need to remove the server 172.16.2.1 line and put something better there. Check the "clock.txt" file on www.ntp.org and find a server near you and then put a server line with it's IP number in your conf. -- Poul-Henning Kamp FreeBSD coreteam member phk@FreeBSD.ORG "Real hackers run -current on their laptop." FreeBSD -- It will take a long time before progress goes too far! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Sun Jul 4 14:17:18 1999 Delivered-To: freebsd-net@freebsd.org Received: from picalon.gun.de (picalon.gun.de [194.77.0.18]) by hub.freebsd.org (Postfix) with ESMTP id 867D415245 for ; Sun, 4 Jul 1999 14:16:46 -0700 (PDT) (envelope-from andreas@klemm.gtn.com) Received: from klemm.gtn.com (pppak04.gtn.com [194.231.123.169]) by picalon.gun.de (8.8.6/8.8.6) with ESMTP id XAA10576; Sun, 4 Jul 1999 23:16:43 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.9.3/8.9.3) id XAA23160; Sun, 4 Jul 1999 23:12:25 +0200 (CEST) (envelope-from andreas) Date: Sun, 4 Jul 1999 23:12:25 +0200 From: Andreas Klemm To: Poul-Henning Kamp Cc: freebsd-net@FreeBSD.ORG Subject: Re: NTP configuration problems between FreeBSD-3.2-STABLE and cisco 25xx IOS 11.3(10)T Message-ID: <19990704231225.A22713@titan.klemm.gtn.com> References: <19990704172613.A12940@titan.klemm.gtn.com> <12435.931104375@critter.freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i In-Reply-To: <12435.931104375@critter.freebsd.dk>; from Poul-Henning Kamp on Sun, Jul 04, 1999 at 06:06:15PM +0200 X-Operating-System: FreeBSD 3.2-STABLE SMP X-Disclaimer: A free society is one where it is safe to be unpopular Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Jul 04, 1999 at 06:06:15PM +0200, Poul-Henning Kamp wrote: > > You are asking the machine to be it's own server. you need to remove > the > server 172.16.2.1 > line and put something better there. Check the "clock.txt" file on > www.ntp.org and find a server near you and then put a server line > with it's IP number in your conf. The machine should be it's own server, since I only have an ISDN dialup uplink to an ISP. Figure out, what it would cost, to synchronize my clock this way. Only on startup I run ntpdate and after that xntp should be used, to have one time in the network, even if it is not 100% accurate. 98% would be enough. I know, that it worked this way, when I installed some ntp client software on a NT 4.0 machine (my wifes computer). Why it doesn't work with the cisco router ? Does the cisco perhaps want to have something more accurate ? Can this be tweaked somehow ? -- Andreas Klemm http://www.FreeBSD.ORG/~andreas http://www.freebsd.org/~fsmp/SMP/SMP.html powered by Symmetric MultiProcessor FreeBSD Latest song from our band: http://www.freebsd.org/~andreas/mp3/schaukel.mp3 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Sun Jul 4 15: 3:48 1999 Delivered-To: freebsd-net@freebsd.org Received: from frmug.org (frmug-gw.frmug.org [193.56.58.252]) by hub.freebsd.org (Postfix) with ESMTP id 9BBB414E54 for ; Sun, 4 Jul 1999 15:03:18 -0700 (PDT) (envelope-from roberto@keltia.freenix.fr) Received: (from uucp@localhost) by frmug.org (8.9.1/frmug-2.3/nospam) with UUCP id AAA22943 for freebsd-net@FreeBSD.ORG; Mon, 5 Jul 1999 00:03:06 +0200 (CEST) (envelope-from roberto@keltia.freenix.fr) Received: by keltia.freenix.fr (Postfix, from userid 101) id 35EFD885F; Sun, 4 Jul 1999 23:48:13 +0200 (CEST) (envelope-from roberto) Date: Sun, 4 Jul 1999 23:48:13 +0200 From: Ollivier Robert To: freebsd-net@FreeBSD.ORG Subject: Re: NTP configuration problems between FreeBSD-3.2-STABLE and cisco 25xx IOS 11.3(10)T Message-ID: <19990704234813.A68454@keltia.freenix.fr> Mail-Followup-To: freebsd-net@FreeBSD.ORG References: <19990704172613.A12940@titan.klemm.gtn.com> <12435.931104375@critter.freebsd.dk> <19990704231225.A22713@titan.klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Mutt/0.95.5i In-Reply-To: <19990704231225.A22713@titan.klemm.gtn.com>; from Andreas Klemm on Sun, Jul 04, 1999 at 11:12:25PM +0200 X-Operating-System: FreeBSD 4.0-CURRENT/ELF ctm#5443 AMD-K6 MMX @ 200 MHz Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org According to Andreas Klemm: > an ISDN dialup uplink to an ISP. Figure out, what it would > cost, to synchronize my clock this way. If you want to synchronise your own clock, use the local clock device (127.127.*). -=-=- server 127.127.1.0 fudge 127.127.1.0 stratum 10 server tara.freenix.org prefer driftfile /var/log/ntp.drift multicastclient -=-=- remote refid st t when poll reach delay offset disp ============================================================================== 127.127.1.0 127.127.1.0 10 - 14 64 377 0.00 0.000 0.93 *193.56.58.67 .DCFa. 1 u 974 1024 377 0.75 -2.908 12.92 -- Ollivier ROBERT -=- FreeBSD: The Power to Serve! -=- roberto@keltia.freenix.fr FreeBSD keltia.freenix.fr 4.0-CURRENT #71: Sun May 9 20:16:32 CEST 1999 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Sun Jul 4 21:43:15 1999 Delivered-To: freebsd-net@freebsd.org Received: from critter.freebsd.dk (critter.freebsd.dk [212.242.40.131]) by hub.freebsd.org (Postfix) with ESMTP id 3AC8314ED5 for ; Sun, 4 Jul 1999 21:43:12 -0700 (PDT) (envelope-from phk@critter.freebsd.dk) Received: from critter.freebsd.dk (localhost [127.0.0.1]) by critter.freebsd.dk (8.9.3/8.9.2) with ESMTP id GAA15009; Mon, 5 Jul 1999 06:42:33 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: Andreas Klemm Cc: freebsd-net@FreeBSD.ORG Subject: Re: NTP configuration problems between FreeBSD-3.2-STABLE and cisco 25xx IOS 11.3(10)T In-reply-to: Your message of "Sun, 04 Jul 1999 23:12:25 +0200." <19990704231225.A22713@titan.klemm.gtn.com> Date: Mon, 05 Jul 1999 06:42:33 +0200 Message-ID: <15007.931149753@critter.freebsd.dk> From: Poul-Henning Kamp Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <19990704231225.A22713@titan.klemm.gtn.com>, Andreas Klemm writes: >On Sun, Jul 04, 1999 at 06:06:15PM +0200, Poul-Henning Kamp wrote: >> >> You are asking the machine to be it's own server. you need to remove >> the >> server 172.16.2.1 >> line and put something better there. Check the "clock.txt" file on >> www.ntp.org and find a server near you and then put a server line >> with it's IP number in your conf. > >The machine should be it's own server, since I only have >an ISDN dialup uplink to an ISP. Figure out, what it would >cost, to synchronize my clock this way. Then read the ntpd docs to see how to do that. >Why it doesn't work with the cisco router ? Does the cisco >perhaps want to have something more accurate ? Your server isn't synchronized, nobody, including itself will trust it. -- Poul-Henning Kamp FreeBSD coreteam member phk@FreeBSD.ORG "Real hackers run -current on their laptop." FreeBSD -- It will take a long time before progress goes too far! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Sun Jul 4 23: 7:13 1999 Delivered-To: freebsd-net@freebsd.org Received: from adelphi.physics.adelaide.edu.au (adelphi.physics.adelaide.edu.au [129.127.36.247]) by hub.freebsd.org (Postfix) with ESMTP id 54F9814D20 for ; Sun, 4 Jul 1999 23:07:09 -0700 (PDT) (envelope-from kkennawa@physics.adelaide.edu.au) Received: from bragg (bragg [129.127.36.34]) by adelphi.physics.adelaide.edu.au (8.8.8/8.8.8/UofA-1.5) with SMTP id PAA26339; Mon, 5 Jul 1999 15:37:08 +0930 (CST) Received: from localhost by bragg; (5.65/1.1.8.2/05Aug95-0227PM) id AA16914; Mon, 5 Jul 1999 15:36:47 +0930 Date: Mon, 5 Jul 1999 15:36:47 +0930 (CST) From: Kris Kennaway X-Sender: kkennawa@bragg To: Derek Jewett Cc: freebsd-net@freebsd.org Subject: Re: Secure FTP... In-Reply-To: <000e01bec4c9$153ecb60$5515a8c0@ws2983> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 2 Jul 1999, Derek Jewett wrote: > Using 3.1-R I need to provide a secure ftp login for an account. I want > the user to have write only access to their usr/home/user directory... I > perused the man page for chown and chgrp, and found how to change rights > on indevidual files but not directories... Could someone steer me to the > man page I need.. Thanks chown and chgrp work on directories as well as files. If you want the user to only be able to see files in their home directory (i.e. not even read anything else outside) then you probably want to look at the chroot capability of ftpd - see the chroot(1) manpage for a description of what it does, and the ftpd manpage for how to configure it for ftp logins. Kris ----- "Never criticize anybody until you have walked a mile in their shoes, because by that time you will be a mile away and have their shoes." -- Unknown To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 6:17:37 1999 Delivered-To: freebsd-net@freebsd.org Received: from silver.komanda.com.ua (silver.komanda.com.ua [195.5.38.138]) by hub.freebsd.org (Postfix) with ESMTP id E6A1914BED for ; Mon, 5 Jul 1999 06:17:30 -0700 (PDT) (envelope-from sysop@komanda.com.ua) Received: from silver.komanda.com.ua (silver.komanda.com.ua [195.5.38.138]) by silver.komanda.com.ua (8.8.8/8.8.8) with SMTP id QAA01066 for ; Mon, 5 Jul 1999 16:32:07 +0300 (EEST) (envelope-from sysop@komanda.com.ua) Date: Mon, 5 Jul 1999 16:32:07 +0300 (EEST) From: Alex Bulygin To: freebsd-net@freebsd.org Subject: FreeBSD & MacOS Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, folks! Im wondering is it possible to use Intel based PC with FreeBSD as file server for LAN segment based on Apple PowerPC? If so which ver. of FreeBSD is the best and how is performance comparatively to NT, Novell? Should it be connected through AppleTalk or TCP/IP? Is there some doc's about this one? Or someone has experience of maintaining such kind of things? Many thanks for any ideas or hints. TIA, Alex. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 8:58:18 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail.tellique.de (big-gw.tellique.de [195.126.133.179]) by hub.freebsd.org (Postfix) with ESMTP id B502014EAE for ; Mon, 5 Jul 1999 08:58:15 -0700 (PDT) (envelope-from ni@tellique.de) Received: from tellique.de (nolde.tellique.de [62.144.106.52]) by mail.tellique.de (8.8.7/8.8.8) with ESMTP id RAA06404; Mon, 5 Jul 1999 17:58:02 +0200 Message-ID: <3780D608.991AD760@tellique.de> Date: Mon, 05 Jul 1999 17:58:00 +0200 From: Juergen Nickelsen Organization: Tellique Kommunikationstechnik GmbH, Germany X-Mailer: Mozilla 4.6 [en] (WinNT; U) X-Accept-Language: de,en MIME-Version: 1.0 To: Tim Tsai Cc: freebsd-net@FreeBSD.ORG Subject: Re: fast transmit, slow receive? References: <19990616134731.A10074@futuresouth.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Tim Tsai wrote on freebsd-net: > We have a machine that is fast when we initiate an FTP transfer to > another machine but is VERY slow to receive it (a few KB per second > on). All the machines are running 2.2.8-STABLE (around February) with > Intel fxp0 cards on an Intel switch. I am not seeing any collisions > or any other problems> that I can tell. I had similar symptoms when there was a half-duplex vs. full-duplex mismatch between the switch and the NIC. You might want to check that. Greetings, -- Juergen Nickelsen Tellique Kommunikationstechnik GmbH Gustav-Meyer-Allee 25, 13355 Berlin, Germany Tel. +49 30 46307-552 / Fax +49 30 46307-579 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 9: 8:45 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail.tellique.de (big-gw.tellique.de [195.126.133.179]) by hub.freebsd.org (Postfix) with ESMTP id 44E3814F94 for ; Mon, 5 Jul 1999 09:08:34 -0700 (PDT) (envelope-from ni@tellique.de) Received: from tellique.de (nolde.tellique.de [62.144.106.52]) by mail.tellique.de (8.8.7/8.8.8) with ESMTP id SAA06804; Mon, 5 Jul 1999 18:08:20 +0200 Message-ID: <3780D872.D21648C2@tellique.de> Date: Mon, 05 Jul 1999 18:08:18 +0200 From: Juergen Nickelsen Organization: Tellique Kommunikationstechnik GmbH, Germany X-Mailer: Mozilla 4.6 [en] (WinNT; U) X-Accept-Language: de,en MIME-Version: 1.0 To: Dominic Mitchell Cc: Nick LoPresti , freebsd-net@FreeBSD.ORG Subject: Re: NIS Question References: <002001bebdf7$f8719e40$8c7361cf@dogwood.chromatix.com> <19990624085123.A98128@palmerharvey.co.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Dominic Mitchell wrote: > Making your box a NIS server is easy. Just do grep -i nis > /etc/defaults/rc.conf and plug the results into your /etc/rc.conf. > You probably want to enable nis_server and yppasswdd. Nick talked about a 2.2.6 machine, so it is all in /etc/rc.conf. /etc/defaults is a 3.x-ism. Greetings, -- Juergen Nickelsen Tellique Kommunikationstechnik GmbH Gustav-Meyer-Allee 25, 13355 Berlin, Germany Tel. +49 30 46307-552 / Fax +49 30 46307-579 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 9:33:29 1999 Delivered-To: freebsd-net@freebsd.org Received: from vortex.greycat.com (vortex.greycat.com [207.173.133.4]) by hub.freebsd.org (Postfix) with SMTP id C9F8514F8F for ; Mon, 5 Jul 1999 09:33:08 -0700 (PDT) (envelope-from dann@greycat.com) Received: (qmail 6701 invoked from network); 5 Jul 1999 16:33:06 -0000 Received: from bigphred.greycat.com (HELO greycat.com) (207.173.133.2) by vortex.greycat.com with SMTP; 5 Jul 1999 16:33:06 -0000 Message-ID: <3780DE74.2DD4F526@greycat.com> Date: Mon, 05 Jul 1999 09:33:56 -0700 From: Dann Lunsford Organization: You're kidding, right? X-Mailer: Mozilla 4.5 [en] (X11; U; FreeBSD 3.2-STABLE i386) X-Accept-Language: en MIME-Version: 1.0 To: Net@freebsd.org Subject: mbuf-less IP? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Just a question out of pure curiosity. I've been studying the Stevens set ("TCP/IP Illustrated", all three vols) and came across a comment that Van Jacobson had an experimental IP stack that didn't use the mbuf structures. The reasoning was that the current implementation was designed when memory was much more constrained, CPU speed was lower, and networks weren't as fast or fat. Reportedly, Jacobson found siginificant performance improvements and better resource utilization over current, mbuf based, implementations. So... Has anybody looked into this idea? I realize it would entail *MAJOR* rewrites and redesigns, and would not be undertaken lightly, but perhaps it would be worth considering for the future. The current design smacks of "It's always been done that way"; whenever I hear that phrase, a chill goes up my spine :-). Anyway, that's it. Just tossing it into the ring; please don't throw knives back :-). Dann Lunsford dann@greycat.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 10:10:15 1999 Delivered-To: freebsd-net@freebsd.org Received: from btm4r4.alcatel.be (btm4r4.alcatel.be [195.207.101.110]) by hub.freebsd.org (Postfix) with ESMTP id AD12D14C2F for ; Mon, 5 Jul 1999 10:09:47 -0700 (PDT) (envelope-from livensw@rc.bel.alcatel.be) Received: from btmq9s.rc.bel.alcatel.be (btmq9s.rc.bel.alcatel.be [138.203.65.182]) by btm4r4.alcatel.be (8.9.1a/8.9.1) with ESMTP id TAA09351; Mon, 5 Jul 1999 19:09:40 +0200 (MET DST) Received: from btmq9z.rc.bel.alcatel.be (btmq9z [138.203.65.192]) by btmq9s.rc.bel.alcatel.be (8.8.8+Sun/8.8.8) with ESMTP id TAA15564; Mon, 5 Jul 1999 19:11:40 +0200 (MET DST) Received: (from livensw@localhost) by btmq9z.rc.bel.alcatel.be (8.8.8+Sun/8.8.8) id TAA29339; Mon, 5 Jul 1999 19:09:34 +0200 (MET DST) Date: Mon, 5 Jul 1999 19:09:34 +0200 From: Wim Livens To: freebsd-net@freebsd.org Cc: wim.livens@alcatel.be Subject: de with null-cable gives errors and some packet-loss Message-ID: <19990705190934.A29039@rc.bel.alcatel.be> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm trying to connect a ZNYX 346 (= four 100baseTX board) via a null-cable to another (fxp or de) 100baseTX card. I'm using FreeBSD 2.2.8-RELEASE and already found out that I need to manually set the mediatype. Now a flood ping seems to work fine: 1421 packets transmitted, 1421 packets received, 0% packet loss round-trip min/avg/max/stddev = 0.096/0.100/0.189/0.009 ms BUT: a netperf over the same link gives a horrible throughput of 0.10 Mbits/sec ! So I assume there is a slight packet-loss at high load. netstat -I reports as many output error as there are packets transmitted. I've put a printf in de driver to see what causes the errors. It happens here: if (d_status & (TULIP_DSTS_TxNOCARR|TULIP_DSTS_TxCARRLOSS)) sc->tulip_dot3stats.dot3StatsCarrierSenseErrors++; After replacing the null-cable by a 100Mbps hub, everything worked fine (and the null-cable does work well with e.g. fxp interfaces). Please reply to me too because I'm not on the list. FYI: de0 rev 34 int a irq 11 on pci2:4:0 de0: ZNYX ZX34X 21140A [10-100Mb/s] pass 2.2 de0: address 00:c0:95:e0:e2:2c -- Wim Livens. Alcatel - Corporate Research Center wim.livens@alcatel.be Fr. Wellesplein 1 livensw@rc.bel.alcatel.be B-2018 Antwerpen Tel: +32 3 240 7570 Belgium. Fax: +32 3 240 9932 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 10:55:23 1999 Delivered-To: freebsd-net@freebsd.org Received: from khavrinen.lcs.mit.edu (khavrinen.lcs.mit.edu [18.24.4.193]) by hub.freebsd.org (Postfix) with ESMTP id E217514C2F for ; Mon, 5 Jul 1999 10:55:19 -0700 (PDT) (envelope-from wollman@khavrinen.lcs.mit.edu) Received: (from wollman@localhost) by khavrinen.lcs.mit.edu (8.9.1/8.9.1) id NAA23518; Mon, 5 Jul 1999 13:55:16 -0400 (EDT) (envelope-from wollman) Date: Mon, 5 Jul 1999 13:55:16 -0400 (EDT) From: Garrett Wollman Message-Id: <199907051755.NAA23518@khavrinen.lcs.mit.edu> To: Dag-Erling Smorgrav Cc: net@FreeBSD.ORG Subject: netstat(1) / sockstat(1) field width adjustments In-Reply-To: References: Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org < said: > The netstat patch has the unfortunate side effect of pushing the state > field so far to the right that most states (except LISTEN) spill over. > I consider that the lesser of two evils. I consider that the greater of two evils. -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same wollman@lcs.mit.edu | O Siem / The fires of freedom Opinions not those of| Dance in the burning flame MIT, LCS, CRS, or NSA| - Susan Aglukark and Chad Irschick To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 13:28:22 1999 Delivered-To: freebsd-net@freebsd.org Received: from slesse.cs.ubc.ca (slesse.cs.ubc.ca [142.103.10.68]) by hub.freebsd.org (Postfix) with ESMTP id 9CE2115409 for ; Mon, 5 Jul 1999 13:28:12 -0700 (PDT) (envelope-from mjmccut@slesse.cs.ubc.ca) Received: (from mjmccut@localhost) by slesse.cs.ubc.ca (8.8.5/8.8.5) id NAA18917; Mon, 5 Jul 1999 13:28:11 -0700 (PDT) Date: Mon, 5 Jul 1999 13:28:11 -0700 (PDT) Message-Id: <199907052028.NAA18917@slesse.cs.ubc.ca> To: From: Mark McCutcheon Subject: Re: FreeBSD & MacOS Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Alex Bulygin asks: > Im wondering is it possible to use Intel based PC with FreeBSD > as file server for LAN segment based on Apple PowerPC? If so > which ver. of FreeBSD is the best and how is performance > comparatively to NT, Novell? Should it be connected through > AppleTalk or TCP/IP? Is there some doc's about this one? Or > someone has experience of maintaining such kind of things? For AppleTalk support under UNIX, have a look at netatalk. It's in the FreeBSD ports collection (/usr/ports/net/netatalk). Homepage is: http://www.umich.edu/~rsug/netatalk/ While my Macintosh network support requirements aren't very demanding, netatalk has worked extremely well and I'm not aware of any adverse scaling properties. You need to make a custom kernel with AppleTalk support turned on (options NETATALK; see LINT). Ports installation is simple (beware, AppleTalk stack is broken on FreeBSD 2.2.5-RELEASE and requires a patch; all subsequent versions appear to be fine). netatalk allows both file and print services as well as AppleTalk routing. netatalk's services are transparent to Mac users, so it's a great deal easier to set up and administer than, for example, NetBIOS support for Macs (like Thursby Software Systems DAVE for WinNT). However, I don't administer WinNT and haven't administered a Novell network in over 10 years, so I can't help with any comparisons. Regards, Mark ---- Mark McCutcheon | UBC High-Speed Networks | To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 13:53:43 1999 Delivered-To: freebsd-net@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by hub.freebsd.org (Postfix) with ESMTP id 4D6A414ED3 for ; Mon, 5 Jul 1999 13:53:36 -0700 (PDT) (envelope-from root@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id NAA05826; Mon, 5 Jul 1999 13:53:02 -0700 (PDT) Message-Id: <199907052053.NAA05826@implode.root.com> To: Dann Lunsford Cc: Net@FreeBSD.ORG Subject: Re: mbuf-less IP? In-reply-to: Your message of "Mon, 05 Jul 1999 09:33:56 PDT." <3780DE74.2DD4F526@greycat.com> From: David Greenman Reply-To: dg@root.com Date: Mon, 05 Jul 1999 13:53:02 -0700 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >Just a question out of pure curiosity. I've been studying the Stevens >set ("TCP/IP Illustrated", all three vols) and came across a comment >that Van Jacobson had an experimental IP stack that didn't use the mbuf >structures. The reasoning was that the current implementation was >designed when memory was much more constrained, CPU speed was lower, and >networks weren't as fast or fat. Reportedly, Jacobson found >siginificant performance improvements and better resource utilization >over current, mbuf based, implementations. > >So... Has anybody looked into this idea? I realize it would entail >*MAJOR* rewrites and redesigns, and would not be undertaken lightly, but >perhaps it would be worth considering for the future. The current >design smacks of "It's always been done that way"; whenever I hear that >phrase, a chill goes up my spine :-). > >Anyway, that's it. Just tossing it into the ring; please don't throw >knives back :-). We've made substantial improvements in FreeBSD to the way that mbufs are allocated and used which have mitigated much of the old problems. It could still be a lot better, however. It's really the socket layer that needs the rewrite; I don't think the TCP/IP stack itself would be that difficult. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 14: 7:22 1999 Delivered-To: freebsd-net@freebsd.org Received: from picalon.gun.de (picalon.gun.de [194.77.0.18]) by hub.freebsd.org (Postfix) with ESMTP id 623A415079 for ; Mon, 5 Jul 1999 14:07:13 -0700 (PDT) (envelope-from andreas@klemm.gtn.com) Received: from klemm.gtn.com (pppak04.gtn.com [194.231.123.169]) by picalon.gun.de (8.8.6/8.8.6) with ESMTP id XAA22078 for ; Mon, 5 Jul 1999 23:07:01 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.9.3/8.9.3) id XAA33193 for freebsd-net@FreeBSD.ORG; Mon, 5 Jul 1999 23:03:39 +0200 (CEST) (envelope-from andreas) Date: Mon, 5 Jul 1999 23:03:38 +0200 From: Andreas Klemm To: freebsd-net@FreeBSD.ORG Subject: need a clock for NTP Message-ID: <19990705230337.A33121@titan.klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i X-Operating-System: FreeBSD 3.2-STABLE SMP X-Disclaimer: A free society is one where it is safe to be unpopular Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi ! I want to configure my FreeBSD server as NTP server in my local LAN. I heard of cheapo clocks that I can connect to a serial port. Where is such a beast available, what stratum does it have, what does it cost ? Andreas /// -- Andreas Klemm http://www.FreeBSD.ORG/~andreas http://www.freebsd.org/~fsmp/SMP/SMP.html powered by Symmetric MultiProcessor FreeBSD Latest song from our band: http://www.freebsd.org/~andreas/mp3/schaukel.mp3 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 15:35:45 1999 Delivered-To: freebsd-net@freebsd.org Received: from frmug.org (frmug-gw.frmug.org [193.56.58.252]) by hub.freebsd.org (Postfix) with ESMTP id 071D715206 for ; Mon, 5 Jul 1999 15:35:38 -0700 (PDT) (envelope-from roberto@keltia.freenix.fr) Received: (from uucp@localhost) by frmug.org (8.9.1/frmug-2.3/nospam) with UUCP id AAA24315 for freebsd-net@FreeBSD.ORG; Tue, 6 Jul 1999 00:35:36 +0200 (CEST) (envelope-from roberto@keltia.freenix.fr) Received: by keltia.freenix.fr (Postfix, from userid 101) id 14E048897; Tue, 6 Jul 1999 00:19:18 +0200 (CEST) (envelope-from roberto) Date: Tue, 6 Jul 1999 00:19:17 +0200 From: Ollivier Robert To: freebsd-net@FreeBSD.ORG Subject: Re: need a clock for NTP Message-ID: <19990706001917.A75957@keltia.freenix.fr> Mail-Followup-To: freebsd-net@FreeBSD.ORG References: <19990705230337.A33121@titan.klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Mutt/0.95.5i In-Reply-To: <19990705230337.A33121@titan.klemm.gtn.com>; from Andreas Klemm on Mon, Jul 05, 1999 at 11:03:38PM +0200 X-Operating-System: FreeBSD 4.0-CURRENT/ELF ctm#5443 AMD-K6 MMX @ 200 MHz Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org According to Andreas Klemm: > I want to configure my FreeBSD server as NTP server in my > local LAN. I heard of cheapo clocks that I can connect to > a serial port. You can build a cheapo DCF77 (you know, that German radio station :-)) for less than 30 US$ now. Pierre Beyssac (pb@freebsd.org) has drawing and part # for that. You'll be in stratum 1 of course: 200 [0:17] roberto@tara:~> ntpq -pn remote refid st t when poll reach delay offset disp ============================================================================== 127.127.1.0 127.127.1.0 10 - 30 64 377 0.00 0.000 0.93 *127.127.8.0 .DCFa. 0 - 102 64 176 0.00 2.186 0.01 202 [0:17] roberto@tara:~> ntpq -c rv status=02f7 leap_none, sync_lf_clock, 15 events, event_clock_excptn processor="i386", system="FreeBSD4.0-CURRENT", leap=00, stratum=1, precision=-27, rootdelay=0.00, rootdispersion=4.69, peer=25125, refid=DCFa, reftime=bb2bad0c.4a43cc07 Tue, Jul 6 1999 0:15:40.290, poll=6, clock=bb2bad8b.49908e58 Tue, Jul 6 1999 0:17:47.287, state=4, phase=2.186, frequency=50.951, jitter=2.770, stability=0.100 -- Ollivier ROBERT -=- FreeBSD: The Power to Serve! -=- roberto@keltia.freenix.fr FreeBSD keltia.freenix.fr 4.0-CURRENT #71: Sun May 9 20:16:32 CEST 1999 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Mon Jul 5 23:29:57 1999 Delivered-To: freebsd-net@freebsd.org Received: from mtk.comcor.ru (mtk.comcor.ru [212.45.0.156]) by hub.freebsd.org (Postfix) with SMTP id AA8491532F for ; Mon, 5 Jul 1999 23:29:46 -0700 (PDT) (envelope-from ryndin@mtk.comcor.ru) Received: by mtk.comcor.ru(Lotus SMTP MTA v4.6.3 (733.2 10-16-1998)) id C32567A6.0023A99A ; Tue, 6 Jul 1999 10:29:31 +0400 X-Lotus-FromDomain: COMCOR From: ryndin@mtk.comcor.ru To: net@freebsd.org Message-ID: Date: Tue, 6 Jul 1999 10:29:37 +0400 Subject: Firewall and Oracle SQLNet Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi everybody! Does anybody manage to install firewall before an Oracle SQL Server? We need to allow a few remote users to connect to our Oracle SQL Server. We install FreeBSD box with ipfw and discover next problem: we allow for remote users to connect to the Oracle box using port mentioned in SQLNet listener configuration (1521). But remote user try to connect twice: first time using port mentioned above and second time using some other port, which, as we suggest, Oracle server sent him during first connection. This port value start from 1030 (after Oracle restart) and increase after each connection and we don't manage to find it upper limit. As we suggest, Oracle uses second port to resolve sessions. We think that it is not a very good idea to allow users to connect to our server using such a wide port range. We look through all Oracle documentation and don't find any mention about the second connection. Oracle's people said that we need to use Oracle certified firewall, but it cost about 30000 backs and what hell of it!!!!! The question is does anybody managed to restrict Oracle in range of using second port values or have any idea about how to do it (there is no way to configure it in SQLNet configuration file). Thanks in advance, Alexey Ryndin. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 0:39:26 1999 Delivered-To: freebsd-net@freebsd.org Received: from enst.enst.fr (enst.enst.fr [137.194.2.16]) by hub.freebsd.org (Postfix) with ESMTP id 9FC4314BCC for ; Tue, 6 Jul 1999 00:39:23 -0700 (PDT) (envelope-from beyssac@enst.fr) Received: from bofh.enst.fr (bofh-2.enst.fr [137.194.2.37]) by enst.enst.fr (8.9.1a/8.9.1) with ESMTP id JAA20845; Tue, 6 Jul 1999 09:39:21 +0200 (MET DST) Received: by bofh.enst.fr (Postfix, from userid 12426) id 5556DD21A; Tue, 6 Jul 1999 09:39:16 +0200 (CEST) Message-ID: <19990706093916.A65244@enst.fr> Date: Tue, 6 Jul 1999 09:39:16 +0200 From: Pierre Beyssac To: ryndin@mtk.comcor.ru, net@FreeBSD.ORG Subject: Re: Firewall and Oracle SQLNet References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2i In-Reply-To: ; from ryndin@mtk.comcor.ru on Tue, Jul 06, 1999 at 10:29:37AM +0400 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Jul 06, 1999 at 10:29:37AM +0400, ryndin@mtk.comcor.ru wrote: > the Oracle box using port mentioned in SQLNet listener configuration (1521). But > remote user try to connect twice: first time using port mentioned above and > second time using some other port, which, as we suggest, Oracle server sent him > during first connection. This port value start from 1030 (after Oracle restart) > and increase after each connection and we don't manage to find it upper limit. As far as I know, dynamic ports can be disabled. If I remember correctly, dynamic ports are needed only when you use multiple servers for the same database. If you disable this, all connections should be on port 1521. -- Pierre Beyssac pb@enst.fr To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 1:35: 3 1999 Delivered-To: freebsd-net@freebsd.org Received: from silver.komanda.com.ua (silver.komanda.com.ua [195.5.38.138]) by hub.freebsd.org (Postfix) with ESMTP id 54B8D15048 for ; Tue, 6 Jul 1999 01:34:56 -0700 (PDT) (envelope-from sysop@komanda.com.ua) Received: from silver.komanda.com.ua (silver.komanda.com.ua [195.5.38.138]) by silver.komanda.com.ua (8.8.8/8.8.8) with SMTP id LAA03662 for ; Tue, 6 Jul 1999 11:49:27 +0300 (EEST) (envelope-from sysop@komanda.com.ua) Date: Tue, 6 Jul 1999 11:49:27 +0300 (EEST) From: Alex Bulygin To: freebsd-net@FreeBSD.ORG Subject: Re: FreeBSD & MacOS In-Reply-To: <199907052028.NAA18917@slesse.cs.ubc.ca> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > For AppleTalk support under UNIX, have a look at netatalk. It's > in the FreeBSD ports collection (/usr/ports/net/netatalk). Thanks, Ive downloaded it yet from 2.2.8-RELEASE /packages. So Im going to install it and test soonly. > While my Macintosh network support requirements aren't very > demanding, netatalk has worked extremely well and I'm not aware > of any adverse scaling properties. At my office, etidors headquarters, Macs are used for publishing production so performance level is vital. Every Quark file is about 2Mb without images and quality of protocol realization for I/O operation on sever is crucial point. > netatalk's services are transparent to Mac users, so it's a great > deal easier to set up and administer than, for example, NetBIOS > support for Macs (like Thursby Software Systems DAVE for WinNT). And what about TCP/IP for this deal? Will it be so transparent for Mac users as netatalk? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 2:38:10 1999 Delivered-To: freebsd-net@freebsd.org Received: from des.follo.net (des.follo.net [195.204.143.216]) by hub.freebsd.org (Postfix) with ESMTP id BCDAC153B1 for ; Tue, 6 Jul 1999 02:38:07 -0700 (PDT) (envelope-from des@des.follo.net) Received: (from des@localhost) by des.follo.net (8.9.3/8.9.3) id LAA26526; Tue, 6 Jul 1999 11:37:46 +0200 (CEST) (envelope-from des) To: Garrett Wollman Cc: Dag-Erling Smorgrav , net@FreeBSD.ORG Subject: Re: netstat(1) / sockstat(1) field width adjustments References: <199907051755.NAA23518@khavrinen.lcs.mit.edu> Organization: Yes Interactive Visit-Us-At: http://www.yes.no/ From: Dag-Erling Smorgrav Date: 06 Jul 1999 11:37:46 +0200 In-Reply-To: Garrett Wollman's message of "Mon, 5 Jul 1999 13:55:16 -0400 (EDT)" Message-ID: Lines: 13 X-Mailer: Gnus v5.5/Emacs 19.34 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Garrett Wollman writes: > < said: > > The netstat patch has the unfortunate side effect of pushing the state > > field so far to the right that most states (except LISTEN) spill over. > > I consider that the lesser of two evils. > I consider that the greater of two evils. So tell us why, instead of just pulling rank for no other apparent reason than satisfying your own ego. DES -- Dag-Erling Smorgrav - des@yes.no To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 6:46:23 1999 Delivered-To: freebsd-net@freebsd.org Received: from khavrinen.lcs.mit.edu (khavrinen.lcs.mit.edu [18.24.4.193]) by hub.freebsd.org (Postfix) with ESMTP id D419214F2B for ; Tue, 6 Jul 1999 06:46:21 -0700 (PDT) (envelope-from wollman@khavrinen.lcs.mit.edu) Received: (from wollman@localhost) by khavrinen.lcs.mit.edu (8.9.1/8.9.1) id JAA29874; Tue, 6 Jul 1999 09:46:19 -0400 (EDT) (envelope-from wollman) Date: Tue, 6 Jul 1999 09:46:19 -0400 (EDT) From: Garrett Wollman Message-Id: <199907061346.JAA29874@khavrinen.lcs.mit.edu> To: dg@root.com Cc: net@FreeBSD.ORG Subject: Re: mbuf-less IP? In-Reply-To: <199907052053.NAA05826@implode.root.com> References: <3780DE74.2DD4F526@greycat.com> <199907052053.NAA05826@implode.root.com> Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org < said: > We've made substantial improvements in FreeBSD to the way that mbufs are > allocated and used which have mitigated much of the old problems. It could > still be a lot better, however. It's really the socket layer that needs the > rewrite; I don't think the TCP/IP stack itself would be that difficult. Actually, the principal notion of Van's work is to completely eliminate the socket layer entirely, and write each protocol to directly interface to the user (well, to the back end of the system calls). One of the significant barriers to performance he identified was the presence of two separate flow-control models which operate essentially as `ships in the night' and often interfere with each other. (That's why the write size/transmit speed curve is discontinuous.) -GAWollman -- Garrett A. Wollman | O Siem / We are all family / O Siem / We're all the same wollman@lcs.mit.edu | O Siem / The fires of freedom Opinions not those of| Dance in the burning flame MIT, LCS, CRS, or NSA| - Susan Aglukark and Chad Irschick To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 11:19:33 1999 Delivered-To: freebsd-net@freebsd.org Received: from xylan.com (postal.xylan.com [208.8.0.248]) by hub.freebsd.org (Postfix) with ESMTP id 6D58314D3B for ; Tue, 6 Jul 1999 11:19:30 -0700 (PDT) (envelope-from wes@softweyr.com) Received: from mailhub.xylan.com by xylan.com (8.8.7/SMI-SVR4 (xylan-mgw 2.2 [OUT])) id LAA07618; Tue, 6 Jul 1999 11:18:52 -0700 (PDT) Received: from omni.xylan.com by mailhub.xylan.com (SMI-8.6/SMI-SVR4 (mailhub 2.1 [HUB])) id LAA23545; Tue, 6 Jul 1999 11:18:52 -0700 Received: from softweyr.com (dyn2.utah.xylan.com) by omni.xylan.com (4.1/SMI-4.1 (xylan engr [SPOOL])) id AA20006; Tue, 6 Jul 99 11:18:51 PDT Message-Id: <3782488A.6F205377@softweyr.com> Date: Tue, 06 Jul 1999 12:18:50 -0600 From: Wes Peters Organization: Softweyr LLC X-Mailer: Mozilla 4.5 [en] (X11; U; FreeBSD 3.1-RELEASE i386) X-Accept-Language: en Mime-Version: 1.0 To: Andreas Klemm Cc: freebsd-net@FreeBSD.ORG Subject: Re: need a clock for NTP References: <19990705230337.A33121@titan.klemm.gtn.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Andreas Klemm wrote: > > Hi ! > > I want to configure my FreeBSD server as NTP server in my > local LAN. I heard of cheapo clocks that I can connect to > a serial port. > > Where is such a beast available, what stratum does it have, > what does it cost ? GPS reference clocks are cheap these days. IIRC, the GPS clock is considered stratum 1, and your sever stratum 2. See for example http://www.truetime.com Inexpensive hand-held GPS receivers can be use for this, too, but are less accurate due to serial communications and you may have problems with signal strength unless you use a remote antenna. They do have the added advantage of being able to clip them onto your dash board, handlebar, or cabin roof for those times when you don't care if your network is getting the absolute correct time or not. ;^) -- "Where am I, and what am I doing in this handbasket?" Wes Peters Softweyr LLC http://softweyr.com/ wes@softweyr.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 11:57: 7 1999 Delivered-To: freebsd-net@freebsd.org Received: from picalon.gun.de (picalon.gun.de [194.77.0.18]) by hub.freebsd.org (Postfix) with ESMTP id 9732015003 for ; Tue, 6 Jul 1999 11:57:02 -0700 (PDT) (envelope-from andreas@klemm.gtn.com) Received: from klemm.gtn.com (pppak04.gtn.com [194.231.123.169]) by picalon.gun.de (8.8.6/8.8.6) with ESMTP id UAA28778; Tue, 6 Jul 1999 20:56:59 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.9.3/8.9.3) id UAA71444; Tue, 6 Jul 1999 20:38:54 +0200 (CEST) (envelope-from andreas) Date: Tue, 6 Jul 1999 20:38:53 +0200 From: Andreas Klemm To: Wes Peters Cc: freebsd-net@FreeBSD.ORG Subject: Re: need a clock for NTP Message-ID: <19990706203853.A71421@titan.klemm.gtn.com> References: <19990705230337.A33121@titan.klemm.gtn.com> <3782488A.6F205377@softweyr.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i In-Reply-To: <3782488A.6F205377@softweyr.com>; from Wes Peters on Tue, Jul 06, 1999 at 12:18:50PM -0600 X-Operating-System: FreeBSD 3.2-STABLE SMP X-Disclaimer: A free society is one where it is safe to be unpopular Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Thanks ! Will check it out. BTW, I read in some articels, that the receivers differ for example in the accuracy. Some poll only in minute intervalls, since other do it more often. What regarding that ? BTW, what do you think about SURE RPC-Modul DCF-77 from: http://www.linum.de/dcf77/funkuhrempfaenger.htm This device gets the power to work from the serial interface. I hate changing batteries ;-) It costs about $45. Andreas /// -- Andreas Klemm http://www.FreeBSD.ORG/~andreas http://www.freebsd.org/~fsmp/SMP/SMP.html powered by Symmetric MultiProcessor FreeBSD Latest song from our band: http://www.freebsd.org/~andreas/mp3/schaukel.mp3 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Tue Jul 6 18:11: 5 1999 Delivered-To: freebsd-net@freebsd.org Received: from xylan.com (postal.xylan.com [208.8.0.248]) by hub.freebsd.org (Postfix) with ESMTP id 2118814CB4 for ; Tue, 6 Jul 1999 18:11:00 -0700 (PDT) (envelope-from wes@softweyr.com) Received: from mailhub.xylan.com by xylan.com (8.8.7/SMI-SVR4 (xylan-mgw 2.2 [OUT])) id SAA13259; Tue, 6 Jul 1999 18:10:25 -0700 (PDT) Received: from omni.xylan.com by mailhub.xylan.com (SMI-8.6/SMI-SVR4 (mailhub 2.1 [HUB])) id SAA09350; Tue, 6 Jul 1999 18:10:25 -0700 Received: from softweyr.com (dyn2.utah.xylan.com) by omni.xylan.com (4.1/SMI-4.1 (xylan engr [SPOOL])) id AA11514; Tue, 6 Jul 99 18:10:23 PDT Message-Id: <3782A8FF.97708CFF@softweyr.com> Date: Tue, 06 Jul 1999 19:10:23 -0600 From: Wes Peters Organization: Softweyr LLC X-Mailer: Mozilla 4.5 [en] (X11; U; FreeBSD 3.1-RELEASE i386) X-Accept-Language: en Mime-Version: 1.0 To: Andreas Klemm Cc: freebsd-net@FreeBSD.ORG Subject: Re: need a clock for NTP References: <19990705230337.A33121@titan.klemm.gtn.com> <3782488A.6F205377@softweyr.com> <19990706203853.A71421@titan.klemm.gtn.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Andreas Klemm wrote: > > Thanks ! > > Will check it out. BTW, I read in some articels, that the > receivers differ for example in the accuracy. Some poll only > in minute intervalls, since other do it more often. > > What regarding that ? > > BTW, what do you think about SURE RPC-Modul DCF-77 from: > http://www.linum.de/dcf77/funkuhrempfaenger.htm > > This device gets the power to work from the serial interface. > I hate changing batteries ;-) It costs about $45. The prices are certainly great. I wish they had an english web page. Radio clocks in general work pretty well as long as you're not looking for a way to set clocks after power failures; they take a day or two to become really accurate. For plugging into a server and leaving for long, reliable run-times, they should work great. -- "Where am I, and what am I doing in this handbasket?" Wes Peters Softweyr LLC http://softweyr.com/ wes@softweyr.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 0:34:58 1999 Delivered-To: freebsd-net@freebsd.org Received: from NTTMSCCJ02.nttmsc.com.my (unknown [203.106.4.3]) by hub.freebsd.org (Postfix) with SMTP id B9B8314E66 for ; Wed, 7 Jul 1999 00:34:42 -0700 (PDT) (envelope-from ettikan@nttmsc.com.my) Received: by NTTMSCCJ02.nttmsc.com.my(Lotus SMTP MTA v4.6.3 (778.2 1-4-1999)) id C82567A7.00244F27 ; Wed, 7 Jul 1999 15:36:35 +0900 X-Lotus-FromDomain: NTTMSC From: "Ettikan Kandasamy" To: freebsd-net@FreeBSD.ORG Message-ID: Date: Wed, 7 Jul 1999 15:36:31 +0900 Subject: PCCARD Installation Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Disposition: inline Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello, I'm trying to install 3Com Fast Ethernet 10/100 Base-TX (model 3CCFE575BT-D) in to my Dell laptob-freeBSD 2.2.8. Could some one help me with the setup/installation procedure. I tried modifying the GENERIC file for the kernel and /etc/pccard* files. But did not succeed. 1) Does FreeBSD 2.2.8 has the driver for it or I need to download it. ( if no where can I download it) ? 2) Is there any page gives the step by step installation guide for this purpose ?? Let me know. 3) If not how/ where to begin ?? Thanks for the help. Ettikan. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 14:34:41 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 5F28E15347 for ; Wed, 7 Jul 1999 14:34:27 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id QAA20641 for freebsd-net@freebsd.org; Wed, 7 Jul 1999 16:34:26 -0500 (CDT) From: Mohit Aron Message-Id: <199907072134.QAA20641@cs.rice.edu> Subject: paper on improving webserver performance To: freebsd-net@freebsd.org Date: Wed, 7 Jul 1999 16:34:26 -0500 (CDT) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I apologize if you've seen this posting earlier - it was posted to freebsd-hackers list earlier because I didn't know about this list. I recently wrote a paper entitled "TCP Implementation Enhancements for Improving Webserver Performance". The abstract for the paper is attached below. The paper can be downloaded from: http://cs-tr.cs.rice.edu:80/Dienst/UI/2.0/Describe/ncstrl.rice_cs/TR99-335/ I thought the FreeBSD folks who manage the TCP/IP code might be interested in incorporating the proposals made in the paper in the FreeBSD code. - Mohit Aron aron@cs.rice.edu Abstract: This paper studies the performance of BSD-based TCP implementations in Web servers. We find that lack of scalability with respect to high TCP connection rates reduces the throughput of Web servers by up to 25% and imposes a memory overhead of up to 32 MB on the kernel. We also find that insufficient accuracy in TCP's timers results in overly conservative delays for retransmission timeouts, causing poor response time, low network utilization and throughput loss. The paper proposes enhancements to the TCP implementation that eliminate these problems, without requiring changes to the protocol or the API. We also find that conventional benchmark environments do not fully expose certain significant performance aspects of TCP implementations and propose techniques that allow these benchmarks to more accurately predict the performance of real servers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 16: 0:56 1999 Delivered-To: freebsd-net@freebsd.org Received: from implode.root.com (root.com [209.102.106.178]) by hub.freebsd.org (Postfix) with ESMTP id C1394154B9 for ; Wed, 7 Jul 1999 16:00:50 -0700 (PDT) (envelope-from root@implode.root.com) Received: from implode.root.com (localhost [127.0.0.1]) by implode.root.com (8.8.8/8.8.5) with ESMTP id QAA23652; Wed, 7 Jul 1999 16:00:10 -0700 (PDT) Message-Id: <199907072300.QAA23652@implode.root.com> To: Mohit Aron Cc: freebsd-net@FreeBSD.ORG Subject: Re: paper on improving webserver performance In-reply-to: Your message of "Wed, 07 Jul 1999 16:34:26 CDT." <199907072134.QAA20641@cs.rice.edu> From: David Greenman Reply-To: dg@root.com Date: Wed, 07 Jul 1999 16:00:10 -0700 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I apologize if you've seen this posting earlier - it was posted to >freebsd-hackers list earlier because I didn't know about this list. > >I recently wrote a paper entitled "TCP Implementation Enhancements for >Improving Webserver Performance". The abstract for the paper is attached >below. The paper can be downloaded from: > > http://cs-tr.cs.rice.edu:80/Dienst/UI/2.0/Describe/ncstrl.rice_cs/TR99-335/ > >I thought the FreeBSD folks who manage the TCP/IP code might be interested in >incorporating the proposals made in the paper in the FreeBSD code. I've quicky skimmed through the paper and it does appear to be well thought out. We've known about the scalability issues with the tcp timers and large numbers of connections for several years and I've been testing some experimental code from Garrett Wollman which completely fixes that. The other issue about the RTT estimator is a bit contraversial, however, as some believe that you have to have a 1/2 second minimum regardless due to delayed acks from the peer (and apparantly the TCP spec specifies 1/2 second rather than the 1/5 second that Berkeley derived stacks use). This doesn't mean that one shouldn't be accurate beyond that, however. -DG David Greenman Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org Creator of high-performance Internet servers - http://www.terasolutions.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 16:47:11 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 1D50415484 for ; Wed, 7 Jul 1999 16:47:08 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id SAA23986; Wed, 7 Jul 1999 18:47:05 -0500 (CDT) From: Mohit Aron Message-Id: <199907072347.SAA23986@cs.rice.edu> Subject: Re: paper on improving webserver performance To: dg@root.com Date: Wed, 7 Jul 1999 18:47:05 -0500 (CDT) Cc: freebsd-net@freebsd.org, druschel@cs.rice.edu (Peter Druschel) In-Reply-To: <199907072300.QAA23652@implode.root.com> from "David Greenman" at Jul 7, 99 04:00:10 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > I've quicky skimmed through the paper and it does appear to be well thought > out. We've known about the scalability issues with the tcp timers and large > numbers of connections for several years and I've been testing some > experimental code from Garrett Wollman which completely fixes that. The > other issue about the RTT estimator is a bit contraversial, however, as some > believe that you have to have a 1/2 second minimum regardless due to delayed > acks from the peer (and apparantly the TCP spec specifies 1/2 second rather > than the 1/5 second that Berkeley derived stacks use). This doesn't mean that > one shouldn't be accurate beyond that, however. > > -DG > > David Greenman > Co-founder/Principal Architect, The FreeBSD Project - http://www.freebsd.org > Creator of high-performance Internet servers - http://www.terasolutions.com Right. I do observe in a footnote that the minimum should be 200ms though I wasn't aware that the specs want this to be 500ms. In any case, having a minimum of 500ms would still be better than the 2-3s timeouts that happen sometimes. BTW, you can even get away with the 200ms (or 500ms) minimum. For example, if you have a large enough number of packets in the network (more than 2 should be enough), then you can argue that the ACKs won't be delayed as the receivers usually send an ACK every 2 or 3 packets. Then you don't have to put a minimum bound on the timeout. - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 19:16:46 1999 Delivered-To: freebsd-net@freebsd.org Received: from adelphi.physics.adelaide.edu.au (adelphi.physics.adelaide.edu.au [129.127.36.247]) by hub.freebsd.org (Postfix) with ESMTP id 3BA9C15493 for ; Wed, 7 Jul 1999 19:16:40 -0700 (PDT) (envelope-from kkennawa@physics.adelaide.edu.au) Received: from bragg (bragg [129.127.36.34]) by adelphi.physics.adelaide.edu.au (8.8.8/8.8.8/UofA-1.5) with SMTP id LAA25715; Thu, 8 Jul 1999 11:46:39 +0930 (CST) Received: from localhost by bragg; (5.65/1.1.8.2/05Aug95-0227PM) id AA16630; Thu, 8 Jul 1999 11:46:35 +0930 Date: Thu, 8 Jul 1999 11:46:35 +0930 (CST) From: Kris Kennaway X-Sender: kkennawa@bragg To: Mohit Aron Cc: freebsd-net@freebsd.org Subject: Re: paper on improving webserver performance In-Reply-To: <199907072134.QAA20641@cs.rice.edu> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 7 Jul 1999, Mohit Aron wrote: > I thought the FreeBSD folks who manage the TCP/IP code might be interested in > incorporating the proposals made in the paper in the FreeBSD code. I took a look through this paper, and the work seems (to me) to be quite useful. You've obviously written the necessary code to support this - have you considered making it available for evaluation and possible inclusion? Kris ----- "Never criticize anybody until you have walked a mile in their shoes, because by that time you will be a mile away and have their shoes." -- Unknown To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 19:34:14 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 2A89F15489 for ; Wed, 7 Jul 1999 19:34:08 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id VAA26105; Wed, 7 Jul 1999 21:33:56 -0500 (CDT) From: Mohit Aron Message-Id: <199907080233.VAA26105@cs.rice.edu> Subject: Re: paper on improving webserver performance To: kkennawa@physics.adelaide.edu.au (Kris Kennaway) Date: Wed, 7 Jul 1999 21:33:56 -0500 (CDT) Cc: freebsd-net@freebsd.org In-Reply-To: from "Kris Kennaway" at Jul 8, 99 11:46:35 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I took a look through this paper, and the work seems (to me) to be quite > useful. You've obviously written the necessary code to support this - have > you considered making it available for evaluation and possible inclusion? My code needs a bit of ironing before I release it. But I will consider releasing it at some point. Also, since the code was writting for FreeBSD-2.2.6 I'm wondering if I should rather port it over to FreeBSD-current before releasing it ? - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 20:13:57 1999 Delivered-To: freebsd-net@freebsd.org Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (Postfix) with ESMTP id 1F5B814BD0 for ; Wed, 7 Jul 1999 20:13:55 -0700 (PDT) (envelope-from julian@whistle.com) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.9.1a/8.9.1) with SMTP id UAA38163; Wed, 7 Jul 1999 20:13:39 -0700 (PDT) Date: Wed, 7 Jul 1999 20:13:38 -0700 (PDT) From: Julian Elischer To: Kris Kennaway Cc: Mohit Aron , freebsd-net@FreeBSD.ORG Subject: Re: paper on improving webserver performance In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org It's definitly interesting. The comment about having compared against timer wheels is especially interesting given that Garrett has been working on replacing the TCP timers with timer wheels... maybe he has some tricks up his sleave that win back some of the difference... julian On Thu, 8 Jul 1999, Kris Kennaway wrote: > On Wed, 7 Jul 1999, Mohit Aron wrote: > > > I thought the FreeBSD folks who manage the TCP/IP code might be interested in > > incorporating the proposals made in the paper in the FreeBSD code. > > I took a look through this paper, and the work seems (to me) to be quite > useful. You've obviously written the necessary code to support this - have you > considered making it available for evaluation and possible inclusion? > > Kris > > ----- > "Never criticize anybody until you have walked a mile in their shoes, > because by that time you will be a mile away and have their shoes." > -- Unknown > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-net" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 20:14:29 1999 Delivered-To: freebsd-net@freebsd.org Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (Postfix) with ESMTP id 3191E14BD0 for ; Wed, 7 Jul 1999 20:14:26 -0700 (PDT) (envelope-from julian@whistle.com) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.9.1a/8.9.1) with SMTP id UAA38214; Wed, 7 Jul 1999 20:14:19 -0700 (PDT) Date: Wed, 7 Jul 1999 20:14:18 -0700 (PDT) From: Julian Elischer To: Mohit Aron Cc: Kris Kennaway , freebsd-net@FreeBSD.ORG Subject: Re: paper on improving webserver performance In-Reply-To: <199907080233.VAA26105@cs.rice.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 7 Jul 1999, Mohit Aron wrote: > > > I took a look through this paper, and the work seems (to me) to be quite > > useful. You've obviously written the necessary code to support this - have > > you considered making it available for evaluation and possible inclusion? > > > My code needs a bit of ironing before I release it. But I will consider > releasing it at some point. Also, since the code was writting for FreeBSD-2.2.6 > I'm wondering if I should rather port it over to FreeBSD-current before > releasing it ? > definitly.. 2.2.x will never be allowed to have such a drastic change to it.. > > > > - Mohit > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-net" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 20:23:17 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 86C2314BD0 for ; Wed, 7 Jul 1999 20:23:10 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id WAA26783; Wed, 7 Jul 1999 22:23:08 -0500 (CDT) From: Mohit Aron Message-Id: <199907080323.WAA26783@cs.rice.edu> Subject: Re: paper on improving webserver performance To: julian@whistle.com (Julian Elischer) Date: Wed, 7 Jul 1999 22:23:07 -0500 (CDT) Cc: freebsd-net@freebsd.org In-Reply-To: from "Julian Elischer" at Jul 7, 99 08:13:38 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > It's definitly interesting. > The comment about having compared against timer wheels is especially > interesting given that Garrett has been working on replacing the TCP > timers with timer wheels... > > maybe he has some tricks up his sleave that win back some of the > difference... > Perhaps the following would help. My implementation of timing wheels was well optimized. To continue TCP to use integer variables, only a single event was inserted into the timing wheel corresponding to all the TCP timers. This event corresponded to the time that the earliest timer needed to fire. Further, this event wasn't cancelled even if the TCP timers were. Thus TCP would set and unset its timers by simply changing the integer variables. When the handler for the timing wheel event fired, it checked all the TCP timers and took corresponding action. After the action was taken for every timer, another event was inserted into the timing wheel that corrsponding to the next earliest TCP timer that needed to fire. To speed up the implementation further, the event inserted into the timing wheel was allocated with the tcpcb structure - so there aren't any malloc or free operations going on after the tcpcb structure is allocated. Inspite of the above, the timing wheel still proved to be a little more costly than the list based implementation. However, the timing wheel solution might be the way to go if the clock granularity were to be made finer than 500ms. - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 20:40:42 1999 Delivered-To: freebsd-net@freebsd.org Received: from sumatra.americantv.com (sumatra.americantv.com [208.139.222.227]) by hub.freebsd.org (Postfix) with ESMTP id 5AA0F14BE3 for ; Wed, 7 Jul 1999 20:40:33 -0700 (PDT) (envelope-from jlemon@americantv.com) Received: from right.PCS (right.PCS [148.105.10.31]) by sumatra.americantv.com (8.8.5/8.8.5) with ESMTP id WAA13669; Wed, 7 Jul 1999 22:40:32 -0500 (CDT) Received: from free.pcs (free.PCS [148.105.10.51]) by right.PCS (8.6.13/8.6.4) with ESMTP id WAA24179; Wed, 7 Jul 1999 22:40:31 -0500 Received: (from jlemon@localhost) by free.pcs (8.8.6/8.8.5) id WAA29430; Wed, 7 Jul 1999 22:40:30 -0500 (CDT) Date: Wed, 7 Jul 1999 22:40:30 -0500 (CDT) From: Jonathan Lemon Message-Id: <199907080340.WAA29430@free.pcs> To: aron@cs.rice.edu, net@freebsd.org Subject: Re: paper on improving webserver performance X-Newsgroups: local.mail.freebsd-net In-Reply-To: References: Organization: Architecture and Operating System Fanatics Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In article you write: > >> >> It's definitly interesting. >> The comment about having compared against timer wheels is especially >> interesting given that Garrett has been working on replacing the TCP >> timers with timer wheels... >> >> maybe he has some tricks up his sleave that win back some of the >> difference... Actually, if you recall my earlier posting, I had a hybrid approach that allowed using either the current approach, or a timing wheel approach. After a bunch of tests, it turns out that the timing wheel was faster, so I merged my code with Garrett's. >Perhaps the following would help. My implementation of timing wheels was >well optimized. To continue TCP to use integer variables, only a single event >was inserted into the timing wheel corresponding to all the TCP timers. This >event corresponded to the time that the earliest timer needed to fire. Further, >this event wasn't cancelled even if the TCP timers were. Thus TCP would set and >unset its timers by simply changing the integer variables. This seems to indicate that you still have the overhead where the timer fires, but no events are actually pending. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 20:46:26 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 0F5A814CA1 for ; Wed, 7 Jul 1999 20:46:22 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id WAA27087; Wed, 7 Jul 1999 22:46:19 -0500 (CDT) From: Mohit Aron Message-Id: <199907080346.WAA27087@cs.rice.edu> Subject: Re: paper on improving webserver performance To: jlemon@americantv.com (Jonathan Lemon) Date: Wed, 7 Jul 1999 22:46:19 -0500 (CDT) Cc: freebsd-net@freebsd.org In-Reply-To: <199907080340.WAA29430@free.pcs> from "Jonathan Lemon" at Jul 7, 99 10:40:30 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > This seems to indicate that you still have the overhead where the timer > fires, but no events are actually pending. True, but it avoids any pointer manipulations when timers are set and cancelled by TCP (which happens much more). Only if a timer is set by TCP that needs to fire earlier than the time that the timing wheel event handler needs to fire would you need to move around the timing wheel event. In any case, I'm very interested in your implementation since you appear to have seen faster performance results with it. - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 20:52:29 1999 Delivered-To: freebsd-net@freebsd.org Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (Postfix) with ESMTP id 929C414C14 for ; Wed, 7 Jul 1999 20:52:21 -0700 (PDT) (envelope-from julian@whistle.com) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.9.1a/8.9.1) with SMTP id UAA40833; Wed, 7 Jul 1999 20:52:16 -0700 (PDT) Date: Wed, 7 Jul 1999 20:52:14 -0700 (PDT) From: Julian Elischer To: Mohit Aron Cc: Jonathan Lemon , freebsd-net@FreeBSD.ORG Subject: Re: paper on improving webserver performance In-Reply-To: <199907080346.WAA27087@cs.rice.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 7 Jul 1999, Mohit Aron wrote: > > > > > This seems to indicate that you still have the overhead where the timer > > fires, but no events are actually pending. > > True, but it avoids any pointer manipulations when timers are set and cancelled > by TCP (which happens much more). Only if a timer is set by TCP that needs to > fire earlier than the time that the timing wheel event handler needs to fire > would you need to move around the timing wheel event. > > In any case, I'm very interested in your implementation since you appear to > have seen faster performance results with it. I might add that both have seen faster when compared with the existing, and both seem quite clever. I like the memory saving scheme too.. basically a "zombie tcp session" :-) > > > > - Mohit > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-net" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 21:13:12 1999 Delivered-To: freebsd-net@freebsd.org Received: from sumatra.americantv.com (sumatra.americantv.com [208.139.222.227]) by hub.freebsd.org (Postfix) with ESMTP id 50B5F14FDF for ; Wed, 7 Jul 1999 21:13:10 -0700 (PDT) (envelope-from jlemon@americantv.com) Received: from right.PCS (right.PCS [148.105.10.31]) by sumatra.americantv.com (8.8.5/8.8.5) with ESMTP id XAA13753; Wed, 7 Jul 1999 23:13:09 -0500 (CDT) Received: (from jlemon@localhost) by right.PCS (8.6.13/8.6.4) id XAA25254; Wed, 7 Jul 1999 23:13:08 -0500 Message-ID: <19990707231308.40142@right.PCS> Date: Wed, 7 Jul 1999 23:13:08 -0500 From: Jonathan Lemon To: Mohit Aron Cc: freebsd-net@freebsd.org Subject: Re: paper on improving webserver performance References: <199907080340.WAA29430@free.pcs> <199907080346.WAA27087@cs.rice.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: <199907080346.WAA27087@cs.rice.edu>; from Mohit Aron on Jul 07, 1999 at 10:46:19PM -0500 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Jul 07, 1999 at 10:46:19PM -0500, Mohit Aron wrote: > > > > This seems to indicate that you still have the overhead where the timer > > fires, but no events are actually pending. > > True, but it avoids any pointer manipulations when timers are set and cancelled > by TCP (which happens much more). Only if a timer is set by TCP that needs to > fire earlier than the time that the timing wheel event handler needs to fire > would you need to move around the timing wheel event. Yes, I thought the same thing too. Originally, my implmentation simply moved all connections in a TIME_WAIT states onto a separate timing wheel, and left the rest of the code alone. Turns out that the overhead of inserting/deleting events from the timing wheel is trivial as compared to the cost of checking the hash buckets and walking the chains checking for events. > In any case, I'm very interested in your implementation since you appear to > have seen faster performance results with it. I'm just about done; wrapping up some testing and making sure I didn't break anything. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 21:34:47 1999 Delivered-To: freebsd-net@freebsd.org Received: from sumatra.americantv.com (sumatra.americantv.com [208.139.222.227]) by hub.freebsd.org (Postfix) with ESMTP id E54E3154DF for ; Wed, 7 Jul 1999 21:34:41 -0700 (PDT) (envelope-from jlemon@americantv.com) Received: from right.PCS (right.PCS [148.105.10.31]) by sumatra.americantv.com (8.8.5/8.8.5) with ESMTP id XAA13871; Wed, 7 Jul 1999 23:34:38 -0500 (CDT) Received: (from jlemon@localhost) by right.PCS (8.6.13/8.6.4) id XAA25782; Wed, 7 Jul 1999 23:34:37 -0500 Message-ID: <19990707233437.30556@right.PCS> Date: Wed, 7 Jul 1999 23:34:37 -0500 From: Jonathan Lemon To: Julian Elischer Cc: Mohit Aron , freebsd-net@FreeBSD.ORG Subject: Re: paper on improving webserver performance References: <199907080346.WAA27087@cs.rice.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: ; from Julian Elischer on Jul 07, 1999 at 08:52:14PM -0700 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Jul 07, 1999 at 08:52:14PM -0700, Julian Elischer wrote: > I like the memory saving scheme too.. basically a "zombie tcp session" :-) Yeah, that seems nice. I was wondering earlier whether it was possible to relocate the TIME_WAIT pcbs onto their own pages, and allowing the kernel to page them out. But just making them 'zombies' appears to be more promising. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 22:11:54 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 58CA9154A4 for ; Wed, 7 Jul 1999 22:11:37 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id AAA28180; Thu, 8 Jul 1999 00:11:33 -0500 (CDT) From: Mohit Aron Message-Id: <199907080511.AAA28180@cs.rice.edu> Subject: Re: paper on improving webserver performance To: jlemon@americantv.com (Jonathan Lemon) Date: Thu, 8 Jul 1999 00:11:33 -0500 (CDT) Cc: freebsd-net@freebsd.org In-Reply-To: <19990707231308.40142@right.PCS> from "Jonathan Lemon" at Jul 7, 99 11:13:08 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Yes, I thought the same thing too. Originally, my implmentation simply > moved all connections in a TIME_WAIT states onto a separate timing wheel, > and left the rest of the code alone. > > Turns out that the overhead of inserting/deleting events from the timing > wheel is trivial as compared to the cost of checking the hash buckets and > walking the chains checking for events. > Possibly that might be the reason why my timing wheel is performing worse. In any case, a 5% difference isn't much to worry about - the earlier drafts of my paper also had a full description of the timing wheel. We threw that out because of space limitations and since the list implementation was simple and promising enough. In any case, I'd like to know how your timing wheel performs vs my list based implementation (the latter is really easy to implement). Perhaps you can compare the two. - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Wed Jul 7 22:26: 6 1999 Delivered-To: freebsd-net@freebsd.org Received: from sumatra.americantv.com (sumatra.americantv.com [208.139.222.227]) by hub.freebsd.org (Postfix) with ESMTP id E02F614DFC for ; Wed, 7 Jul 1999 22:26:02 -0700 (PDT) (envelope-from jlemon@americantv.com) Received: from right.PCS (right.PCS [148.105.10.31]) by sumatra.americantv.com (8.8.5/8.8.5) with ESMTP id AAA14036; Thu, 8 Jul 1999 00:26:02 -0500 (CDT) Received: (from jlemon@localhost) by right.PCS (8.6.13/8.6.4) id AAA28561; Thu, 8 Jul 1999 00:26:01 -0500 Message-ID: <19990708002601.33591@right.PCS> Date: Thu, 8 Jul 1999 00:26:01 -0500 From: Jonathan Lemon To: Mohit Aron Cc: freebsd-net@freebsd.org Subject: Re: paper on improving webserver performance References: <19990707231308.40142@right.PCS> <199907080511.AAA28180@cs.rice.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: <199907080511.AAA28180@cs.rice.edu>; from Mohit Aron on Jul 07, 1999 at 12:11:33AM -0500 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Jul 07, 1999 at 12:11:33AM -0500, Mohit Aron wrote: > > Possibly that might be the reason why my timing wheel is performing worse. > In any case, a 5% difference isn't much to worry about - the earlier drafts of > my paper also had a full description of the timing wheel. We threw that out > because of space limitations and since the list implementation was simple > and promising enough. In any case, I'd like to know how your timing wheel > performs vs my list based implementation (the latter is really easy to > implement). Perhaps you can compare the two. Sure, if you want to send me the code, I can see how they compare with the particular application that I'm using. (a web proxy server). -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 1:43:52 1999 Delivered-To: freebsd-net@freebsd.org Received: from enst.enst.fr (enst.enst.fr [137.194.2.16]) by hub.freebsd.org (Postfix) with ESMTP id E468D151ED for ; Thu, 8 Jul 1999 01:43:48 -0700 (PDT) (envelope-from beyssac@enst.fr) Received: from bofh.enst.fr (bofh-2.enst.fr [137.194.2.37]) by enst.enst.fr (8.9.1a/8.9.1) with ESMTP id KAA04236; Thu, 8 Jul 1999 10:43:46 +0200 (MET DST) Received: by bofh.enst.fr (Postfix, from userid 12426) id 8A79AD226; Thu, 8 Jul 1999 10:43:40 +0200 (CEST) Message-ID: <19990708104340.A6425@enst.fr> Date: Thu, 8 Jul 1999 10:43:40 +0200 From: Pierre Beyssac To: Wes Peters , Andreas Klemm Cc: freebsd-net@FreeBSD.ORG Subject: Re: need a clock for NTP References: <19990705230337.A33121@titan.klemm.gtn.com> <3782488A.6F205377@softweyr.com> <19990706203853.A71421@titan.klemm.gtn.com> <3782A8FF.97708CFF@softweyr.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2i In-Reply-To: <3782A8FF.97708CFF@softweyr.com>; from Wes Peters on Tue, Jul 06, 1999 at 07:10:23PM -0600 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Jul 06, 1999 at 07:10:23PM -0600, Wes Peters wrote: > The prices are certainly great. I wish they had an english web page. > Radio clocks in general work pretty well as long as you're not looking > for a way to set clocks after power failures; they take a day or two > to become really accurate. That doesn't include "raw" DCF77 clocks. These clocks just generate a series of pulses received through an AM antenna, and ntpd decodes this. It just takes 90 seconds average at startup to get in sync (0 to 60 seconds to sync with the beginning of a message, then 60 seconds to get a complete message). -- Pierre Beyssac pb@enst.fr To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 5:36:52 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 4D0D314E5B for ; Thu, 8 Jul 1999 05:36:48 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id HAA02862; Thu, 8 Jul 1999 07:36:46 -0500 (CDT) From: Mohit Aron Message-Id: <199907081236.HAA02862@cs.rice.edu> Subject: Re: paper on improving webserver performance To: jlemon@cs.wisc.edu (Jonathan Lemon) Date: Thu, 8 Jul 1999 07:36:46 -0500 (CDT) Cc: freebsd-net@freebsd.org In-Reply-To: <19990708002601.33591@right.PCS> from "Jonathan Lemon" at Jul 8, 99 00:26:01 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Sure, if you want to send me the code, I can see how they compare > with the particular application that I'm using. (a web proxy server). I can do that but my code is for FreeBSD-2.2.6 if that's not a problem. Lets take this discussion of the freebsd-net list. - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 7:47:54 1999 Delivered-To: freebsd-net@freebsd.org Received: from noop.colo.erols.net (noop.colo.erols.net [207.96.1.150]) by hub.freebsd.org (Postfix) with ESMTP id 07EA614F7A for ; Thu, 8 Jul 1999 07:47:52 -0700 (PDT) (envelope-from gjp@noop.colo.erols.net) Received: from localhost ([127.0.0.1] helo=noop.colo.erols.net) by noop.colo.erols.net with esmtp (Exim 2.12 #1) id 112FSi-0003mo-00; Thu, 8 Jul 1999 10:47:44 -0400 To: "Ettikan Kandasamy" Cc: freebsd-net@FreeBSD.ORG From: "Gary Palmer" Subject: Re: PCCARD Installation In-reply-to: Your message of "Wed, 07 Jul 1999 15:36:31 +0900." Date: Thu, 08 Jul 1999 10:47:43 -0400 Message-ID: <14557.931445263@noop.colo.erols.net> Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org "Ettikan Kandasamy" wrote in message ID : > > > Hello, > > I'm trying to install 3Com Fast Ethernet 10/100 Base-TX (model > 3CCFE575BT-D) in to my Dell laptob-freeBSD 2.2.8. Could some one help me with > the setup/installation procedure. That sounds suspiciously like the ethernet net card I got with my dell, and no, its not supported in FreeBSD at this time. The problem isn't the driver for the card, but support for CardBus, which is different enough from PCMCIA/PCCARD to need a lot of work. You really should ask follow up questions on freebsd-mobile. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 11:36: 2 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail-relay2.yahoo.com (mail-relay2.yahoo.com [206.251.17.77]) by hub.freebsd.org (Postfix) with ESMTP id 4725F15306 for ; Thu, 8 Jul 1999 11:35:56 -0700 (PDT) (envelope-from jayanth@yahoo-inc.com) Received: from borogove.yahoo.com (borogove.yahoo.com [205.216.162.65]) by mail-relay2.yahoo.com (8.9.3/8.9.3) with ESMTP id LAA18614; Thu, 8 Jul 1999 11:35:48 -0700 (PDT) Received: from yahoo-inc.com (milk.yahoo.com [206.132.89.117]) by borogove.yahoo.com (8.9.3/8.9.3) with ESMTP id LAA05820; Thu, 8 Jul 1999 11:35:47 -0700 (PDT) Message-ID: <3784EF83.8E7DD130@yahoo-inc.com> Date: Thu, 08 Jul 1999 11:35:47 -0700 From: Jayanth Vijayaraghavan X-Mailer: Mozilla 4.08 [en] (X11; I; FreeBSD 2.2.8-STABLE i386) MIME-Version: 1.0 To: Mohit Aron Cc: freebsd-net@FreeBSD.ORG Subject: Re: paper on improving webserver performance References: <199907072134.QAA20641@cs.rice.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Mohit, In your paper you discuss about having an application restrict the no. of idle connections to 10000 for http 1.1. I am sure there are a lot of servers that have much more than 10000 connections that are idle and have some sort of keep alive mechanism turned on (not just http 1.1), even if the time for which the connections are idle be of the order of minutes. Scanning the list now could be time consuming . Wouldnt a timewheel be more generic as it handles all the timers ? Jayanth To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 11:51:26 1999 Delivered-To: freebsd-net@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id E66461506A for ; Thu, 8 Jul 1999 11:51:23 -0700 (PDT) (envelope-from aron@cs.rice.edu) Received: (from aron@localhost) by cs.rice.edu (8.9.0/8.9.0) id NAA14204; Thu, 8 Jul 1999 13:51:18 -0500 (CDT) From: Mohit Aron Message-Id: <199907081851.NAA14204@cs.rice.edu> Subject: Re: paper on improving webserver performance To: jayanth@yahoo-inc.com (Jayanth Vijayaraghavan) Date: Thu, 8 Jul 1999 13:51:16 -0500 (CDT) Cc: freebsd-net@freebsd.org In-Reply-To: <3784EF83.8E7DD130@yahoo-inc.com> from "Jayanth Vijayaraghavan" at Jul 8, 99 11:35:47 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > In your paper you discuss about having an application restrict the no. of > idle connections to 10000 for http 1.1. I am sure there are a lot > of servers that have much more than 10000 connections that are idle > and have some sort of keep alive mechanism turned on (not just http 1.1), I consider the keepalive mechanism with HTTP/1.0 and the HTTP/1.1 styled persistent connections to be really synonymous. > even if the time for which the connections are idle be of the order of minutes. > Scanning the list now could be time consuming . Wouldnt a timewheel be more > generic as it handles all the timers ? > Absolutely - a timing wheel's performance degrades very little with idle connections. However, I was under the impression that most servers only allow idle connections upto a max of 15 secs. Second, I would've thought that servers would run into other problems with such a huge number of idle connections - e.g. running out of file descriptors. Though I don't know how real servers plan to control the idle connections, but I think they're probably going to impose some restriction on the maximum number of such connections and close them prematurely (which HTTP allows) after a timeout or after a max is reached. My results indicate that for 10000 idle connections (which seems rather large) the list implementation only has a degradation of 5%. The list implementation seemed both simple and scalable and that's the reason why I proposed it. However, the timing wheel performs within 5% and as far as performace goes, I don't really see compelling reasons for choosing one over the other. I've remarked this earlier on this list too. - Mohit To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 13:48:18 1999 Delivered-To: freebsd-net@freebsd.org Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by hub.freebsd.org (Postfix) with SMTP id 7E9F514FF9 for ; Thu, 8 Jul 1999 13:48:07 -0700 (PDT) (envelope-from luigi@labinfo.iet.unipi.it) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id UAA29438; Thu, 8 Jul 1999 20:18:25 +0200 From: Luigi Rizzo Message-Id: <199907081818.UAA29438@labinfo.iet.unipi.it> Subject: interesting connect(2) side effect... To: net@freebsd.org Date: Thu, 8 Jul 1999 20:18:25 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 2061 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, an interest side effect of connect(2) on BSD-derived systems (checked on the bible aka Stevens vol.2). It came out when implementing the sender side of PGM. As we all know, when doing a connect() on a socket, the faddr field of the pcb is set with the destination address -- a multicast one in the case of PGM, or for many uses of multicast UDP sockets. As a side effect,when demuxing input packets, faddr is also compared with the _source_ address of the incoming packets, unless it is INADDR_ANY. This happens in in_pcblookup() or variations. This is fine with TCP because a socket is only supposed to receive from its peer. Kind of works (because of the INADDR_ANY hack) with non-connected datagram sockets where transmission is done using sendto(). It is in my opinion totally broken with connected datagram sockets, where the connect has the side effect of in-kernel packet filtering based on source address (who said that you cannot do that!). Even if the above is a reasonable model of operation for some unicast UDP-based service, it is totally problematic with multicast where one might want to use the same socket to send multicast and collect feedback from the receivers -- in this case there is no way the source IP can match the multicast address. I suppose this problem came out when developing RTP/RTCP apps and is the reason why two different sockets are used for sending and receiving stuff! Would there be strong objections in making a slight modification to in_pcblookup() and friends to skip the check against faddr if this is a multicast address ? cheers luigi -----------------------------------+------------------------------------- Luigi RIZZO, luigi@iet.unipi.it . Dip. di Ing. dell'Informazione http://www.iet.unipi.it/~luigi/ . Universita` di Pisa TEL/FAX: +39-050-568.533/522 . via Diotisalvi 2, 56126 PISA (Italy) http://www.iet.unipi.it/~luigi/ngc99/ ==== First International Workshop on Networked Group Communication ==== -----------------------------------+------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 21:11:19 1999 Delivered-To: freebsd-net@freebsd.org Received: from xena.gsicomp.on.ca (gsi.enoreo.on.ca [209.82.52.145]) by hub.freebsd.org (Postfix) with ESMTP id D2A1114F1F for ; Thu, 8 Jul 1999 21:11:08 -0700 (PDT) (envelope-from matt@gsicomp.on.ca) Received: from matt (kwppp28.enoreo.on.ca [209.82.45.60]) by xena.gsicomp.on.ca (8.9.2/8.9.2) with SMTP id BAA27833 for ; Fri, 9 Jul 1999 01:09:40 -0400 (EDT) (envelope-from matt@gsicomp.on.ca) Message-ID: <000701bec9c1$41e43ae0$1200a8c0@gsicomp.on.ca> From: "Matthew Emmerton" To: Subject: Multiple NATD Date: Fri, 9 Jul 1999 00:12:25 -0400 Organization: GSI Computer Services MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I've been running a multi-homed (modem and ethernet) NAT box for some time. I ran into trouble attempting to get NAT to work with both interfacees at the same time (for example, when I bring up the modem and add static routes through my dial-up ISP to get around the horrible backend which my ethernet provider has to my ISP.) Thus, I've modified rc.network and rc.conf to handle multiple NAT processes (one per interface), each listening on a different port. Now, the trouble I was having was on 2.2.7-RELEASE. Is it possible to run one NAT processes over multiple interfaces, or is a one-to-one mapping needed still on 3.2-RELEASE? (Please respond with comments and suggestions via e-mail since my ISP doesn't have the freebsd newsgroups, nor do I subscribe to the mailing lists...yet) -- Matthew Emmerton || http://www.gsicomp.on.ca GSI Computer Services || P: +1 (800) 217-5409 (Canada) Technical Director || F: +1 (519) 335-6584 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 22:44:58 1999 Delivered-To: freebsd-net@freebsd.org Received: from hal.bsdguy.com (tnt132-175.nidlink.com [216.18.132.175]) by hub.freebsd.org (Postfix) with ESMTP id 812E114CEC for ; Thu, 8 Jul 1999 22:44:45 -0700 (PDT) (envelope-from shawn@bsdguy.com) Received: from hal.nidlink.com (localhost.nidlink.com [127.0.0.1]) by hal.bsdguy.com (8.9.3/8.9.2) with ESMTP id WAA00588 for ; Thu, 8 Jul 1999 22:49:25 -0700 (PDT) (envelope-from shawn@bsdguy.com) Message-ID: X-Mailer: XFMail 1.3 [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Date: Thu, 08 Jul 1999 22:49:24 -0700 (PDT) Reply-To: shawn@bsdguy.com From: Shawn Workman To: net@freebsd.org Subject: Setting up LAN Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I am having some difficulty setting up my LAN here in my house. I have my main box (FreeBSD 3.2 Stable) that has 2 NIC's, vr0 is my outside interface and vr1 is my inside interface. I can get out on the net no proble, with this box. the IP address of vr0 is 216.18.166.162 the IP address of vr1 is 216.18.166.163 my netmask is 255.255.255.224. My P130's IP address is 216.18.166.161 I have another box (Linux RedHat 6.0), it's IP address is 216.18.166.164 The P130 is connected to vr0 vr1 is running into a ethernet switch, this is the main feed for my LAN. eth0 on the Linux box is also plugged into the switch. I have the gateway of the Linux box set to the IP address of vr1 on the BSD box The problem I am having is that I can not ping either box from the other.. The only time I have been able to ping the boxes from one another is when I run natd on vr0 and have the appropriate rules setup for ipfw. I would really like to use my real IP addresses on the LAN since the boxes I am putting up will be accessed the internet and I do not want to use port forwarding of natd. Any suggestions.. Any help is greatly appreciated.. ---------------------------------- E-Mail: Shawn Workman Date: 08-Jul-99 Time: 22:40:02 This message was sent by XFMail ---------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 23: 7:35 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail-out2.apple.com (mail-out2.apple.com [17.254.0.51]) by hub.freebsd.org (Postfix) with ESMTP id D0FF014DB5 for ; Thu, 8 Jul 1999 23:07:31 -0700 (PDT) (envelope-from justin@walker3.apple.com) Received: from mailgate2.apple.com ([17.129.100.225]) by mail-out2.apple.com (8.8.5/8.8.5) with ESMTP id XAA52362 for ; Thu, 8 Jul 1999 23:07:32 -0700 Received: from scv4.apple.com (scv4.apple.com) by mailgate2.apple.com (mailgate2.apple.com- SMTPRS 2.0.15) with ESMTP id ; Thu, 08 Jul 1999 23:07:29 -0700 Received: from walker3.apple.com (walker3.apple.com [17.219.24.201]) by scv4.apple.com (8.9.3/8.9.3) with ESMTP id XAA40782; Thu, 8 Jul 1999 23:07:28 -0700 Received: by walker3.apple.com (8.9.1/8.9.1) id XAA01050; Thu, 8 Jul 1999 23:07:29 -0700 (PDT) Message-Id: <199907090607.XAA01050@walker3.apple.com> To: shawn@bsdguy.com Subject: Re: Setting up LAN Cc: net@freebsd.org Date: Thu, 8 Jul 1999 23:07:27 -0700 From: "Justin C. Walker" Reply-To: justin@apple.com X-Mailer: by Apple MailViewer (2.105.dev) Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > From: Shawn Workman > Date: 1999-07-08 22:45:08 -0700 > To: net@FreeBSD.ORG > Subject: Setting up LAN > X-Priority: 3 (Normal) > Delivered-to: freebsd-net@freebsd.org > X-Mailer: XFMail 1.3 [p0] on FreeBSD > X-Loop: FreeBSD.org > > I am having some difficulty setting up my LAN here in my house. > I have my main box (FreeBSD 3.2 Stable) that has 2 NIC's, vr0 is my outside > interface and vr1 is my inside interface. I can get out on the net no proble, > with this box. > > the IP address of vr0 is 216.18.166.162 > the IP address of vr1 is 216.18.166.163 I think this is your problem. You can't have the same subnet on two distinct interfaces (it confuses the stack). With your mask, you are using the subnet 216.18.166.160 on both. They need to be different. Why do you need to have the two interfaces on the BSD box? Seems like you can have a single subnet with all systems connected to the "cable" that connects to your ISP's bridge/router (I'm making a few leaps of the imagination here, to guess your real configuration). Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Manager, CoreOS Networking | Men are from Earth. Apple Computer, Inc. | Women are from Earth. 2 Infinite Loop | Deal with it. Cupertino, CA 95014 | *-------------------------------------*-------------------------------* To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Thu Jul 8 23:18:11 1999 Delivered-To: freebsd-net@freebsd.org Received: from hal.bsdguy.com (tnt132-134.nidlink.com [216.18.132.134]) by hub.freebsd.org (Postfix) with ESMTP id 0473314DB5 for ; Thu, 8 Jul 1999 23:18:03 -0700 (PDT) (envelope-from shawn@bsdguy.com) Received: from hal.nidlink.com (localhost.nidlink.com [127.0.0.1]) by hal.bsdguy.com (8.9.3/8.9.2) with ESMTP id XAA00713; Thu, 8 Jul 1999 23:22:40 -0700 (PDT) (envelope-from shawn@bsdguy.com) Message-ID: X-Mailer: XFMail 1.3 [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199907090607.XAA01050@walker3.apple.com> Date: Thu, 08 Jul 1999 23:22:40 -0700 (PDT) Reply-To: shawn@bsdguy.com From: Shawn Workman To: "Justin C. Walker" Subject: Re: Setting up LAN Cc: net@freebsd.org Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I forgot to mention that the FreeBSD box is also my firewall.. could I divide my subnet? Maybe have the netmask on the router be a netmask of 16 IP's and the netmask on my vr1 interface be 16 IP's? I suppose I could get another 224 network and have 32 IPs on the vr1 interface.. Of course then I would be loosing 4 IP's either way I go.. I suppose the easiest way to do it would be to use natd and forward the requests from the real IP's to the translated IP's, but I was trying to avoid that.. On 09-Jul-99 Justin C. Walker wrote: >> From: Shawn Workman >> Date: 1999-07-08 22:45:08 -0700 >> To: net@FreeBSD.ORG >> Subject: Setting up LAN >> X-Priority: 3 (Normal) >> Delivered-to: freebsd-net@freebsd.org >> X-Mailer: XFMail 1.3 [p0] on FreeBSD >> X-Loop: FreeBSD.org >> >> I am having some difficulty setting up my LAN here in my house. >> I have my main box (FreeBSD 3.2 Stable) that has 2 NIC's, vr0 is > my outside >> interface and vr1 is my inside interface. I can get out on the > net no proble, >> with this box. >> >> the IP address of vr0 is 216.18.166.162 >> the IP address of vr1 is 216.18.166.163 > I think this is your problem. You can't have the same subnet on > two distinct interfaces (it confuses the stack). With your mask, you > are using the subnet 216.18.166.160 on both. They need to be > different. > > Why do you need to have the two interfaces on the BSD box? Seems > like you can have a single subnet with all systems connected to the > "cable" that connects to your ISP's bridge/router (I'm making a few > leaps of the imagination here, to guess your real configuration). > > Regards, > > Justin > > -- > Justin C. Walker, Curmudgeon-At-Large * > Institute for General Semantics | > Manager, CoreOS Networking | Men are from Earth. > Apple Computer, Inc. | Women are from Earth. > 2 Infinite Loop | Deal with it. > Cupertino, CA 95014 | > *-------------------------------------*-------------------------------* > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-net" in the body of the message ---------------------------------- E-Mail: Shawn Workman Date: 08-Jul-99 Time: 23:17:50 This message was sent by XFMail ---------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 6:16:58 1999 Delivered-To: freebsd-net@freebsd.org Received: from mh.acorn.co.uk (mh.acorn.co.uk [136.170.131.2]) by hub.freebsd.org (Postfix) with ESMTP id 34F04155B7 for ; Fri, 9 Jul 1999 06:16:39 -0700 (PDT) (envelope-from kbracey@e-14.com) Received: from kbracey.acorn.co.uk (kbracey [136.170.129.213]) by mh.acorn.co.uk (8.8.6/8.8.6) with SMTP id OAA18394 for ; Fri, 9 Jul 1999 14:16:36 +0100 (BST) Date: Fri, 09 Jul 1999 14:16:44 +0100 From: Kevin Bracey To: freebsd-net@freebsd.org Subject: Re: interesting connect(2) side effect... Message-ID: In-Reply-To: <199907081818.UAA29438@labinfo.iet.unipi.it> X-Organization: Acorn Computers Ltd, Cambridge, United Kingdom X-Mailer: Messenger v1.40f for RISC OS MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.61b Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <199907081818.UAA29438@labinfo.iet.unipi.it> Luigi Rizzo wrote: > Hi, > > an interest side effect of connect(2) on BSD-derived systems (checked > on the bible aka Stevens vol.2). It came out when implementing the > sender side of PGM. > > As we all know, when doing a connect() on a socket, the faddr field > of the pcb is set with the destination address -- a multicast one > in the case of PGM, or for many uses of multicast UDP sockets. > > As a side effect,when demuxing input packets, faddr is also compared > with the _source_ address of the incoming packets, unless it is > INADDR_ANY. This happens in in_pcblookup() or variations. > > This is fine with TCP because a socket is only supposed to receive > from its peer. Kind of works (because of the INADDR_ANY hack) with > non-connected datagram sockets where transmission is done using > sendto(). > > It is in my opinion totally broken with connected datagram sockets, > where the connect has the side effect of in-kernel packet filtering > based on source address (who said that you cannot do that!). It's not a side effect - it's a basic required feature! My TFTP client relies on it to filter incoming multicast TFTP channels, which are keyed by server address+port. > > Even if the above is a reasonable model of operation for some > unicast UDP-based service, it is totally problematic with multicast > where one might want to use the same socket to send multicast and > collect feedback from the receivers -- in this case there is no way the > source IP can match the multicast address. > I suppose this problem came out when developing RTP/RTCP apps and > is the reason why two different sockets are used for sending and > receiving stuff! > > Would there be strong objections in making a slight modification to > in_pcblookup() and friends to skip the check against faddr if this is a > multicast address ? > Hmmm. Any user code that relied on this wouldn't be very portable. Surely easier to just not connect the socket in the first place, and use sendto? I've been involved in writing a TFTP Multicast server where this sort of came up - although you need to unicast as well as multicast from the socket, so you still can't connect. If you really want to connect, could you not open two sockets with SO_REUSEPORT/ADDR (whichever is necessary): Local address Remote address machine.5555 239.192.4.0.tftp-mcast machine.5555 *.* Then the kernel would filter incoming packets to the second port. -- Kevin Bracey, Senior Software Engineer Pace Micro Technology plc Tel: +44 (0) 1223 725228 645 Newmarket Road Fax: +44 (0) 1223 725328 Cambridge, CB5 8PB, United Kingdom WWW: http://www.acorn.co.uk/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 7:16:53 1999 Delivered-To: freebsd-net@freebsd.org Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by hub.freebsd.org (Postfix) with SMTP id A587C155E2 for ; Fri, 9 Jul 1999 07:16:49 -0700 (PDT) (envelope-from luigi@labinfo.iet.unipi.it) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id NAA01166; Fri, 9 Jul 1999 13:47:10 +0200 From: Luigi Rizzo Message-Id: <199907091147.NAA01166@labinfo.iet.unipi.it> Subject: Re: interesting connect(2) side effect... To: kbracey@e-14.com (Kevin Bracey) Date: Fri, 9 Jul 1999 13:47:10 +0200 (MET DST) Cc: freebsd-net@FreeBSD.ORG In-Reply-To: from "Kevin Bracey" at Jul 9, 99 02:16:25 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 2123 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > It is in my opinion totally broken with connected datagram sockets, > > where the connect has the side effect of in-kernel packet filtering > > based on source address (who said that you cannot do that!). > > It's not a side effect - it's a basic required feature! My TFTP client > relies on it to filter incoming multicast TFTP channels, which are keyed > by server address+port. i was considering this a side effect because i can't find documentation mentioning this as a feature -- certainly not the connect(2) manpage. Not to mention that nothing prevents a second, unprivileged app on your server send packets on the same UDP port -- if you want to do a safe filtering you do need a better mechanism. > > Would there be strong objections in making a slight modification to > > in_pcblookup() and friends to skip the check against faddr if this is a > > multicast address ? > > Hmmm. Any user code that relied on this wouldn't be very portable. well, my case was for implementing Cisco's PGM, and specifically the problem arises because the sender needs to get unicast NAKs from receivers/routers. I can't use sendto because PGM is connection-oriented (kind-of), and i cannot use a second socket because the state machine on the sender is associated with the first one. Now it turns out that for PGM there is an easier solution since the multicast demux code already does what i need and thus it is easy to use it for unicast as well. But i was thinking if the current way of implementing socket-matching for multicast UDP isn't just the result of quick adaptation of existing code without thinking too much at the side effects. cheers luigi -----------------------------------+------------------------------------- Luigi RIZZO, luigi@iet.unipi.it . Dip. di Ing. dell'Informazione http://www.iet.unipi.it/~luigi/ . Universita` di Pisa TEL/FAX: +39-050-568.533/522 . via Diotisalvi 2, 56126 PISA (Italy) http://www.iet.unipi.it/~luigi/ngc99/ ==== First International Workshop on Networked Group Communication ==== -----------------------------------+------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 8:22: 8 1999 Delivered-To: freebsd-net@freebsd.org Received: from databus.databus.com (databus.databus.com [198.186.154.34]) by hub.freebsd.org (Postfix) with SMTP id 1D13014EA5 for ; Fri, 9 Jul 1999 08:22:06 -0700 (PDT) (envelope-from barney@databus.databus.com) From: Barney Wolff To: freebsd-net@FreeBSD.ORG Date: Fri, 9 Jul 1999 11:18 EDT Subject: Re: interesting connect(2) side effect... Content-Length: 637 Content-Type: text/plain Message-ID: <3786139b0.5440@databus.databus.com> Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org My ancient 2.1.5 connect(2) manpage does document the UDP effect. It also works on Irix 5.3-6.5 and is documented in the manpage, so really it's not unintentional. Barney Wolff > From: Luigi Rizzo > Date: Fri, 9 Jul 1999 13:47:10 +0200 (MET DST) > > i was considering this a side effect because i can't find documentation > mentioning this as a feature -- certainly not the connect(2) manpage. > Not to mention that nothing prevents a second, unprivileged app on > your server send packets on the same UDP port -- if you want to do > a safe filtering you do need a better mechanism. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 8:25:11 1999 Delivered-To: freebsd-net@freebsd.org Received: from mh.acorn.co.uk (mh.acorn.co.uk [136.170.131.2]) by hub.freebsd.org (Postfix) with ESMTP id 4AD9114DF4 for ; Fri, 9 Jul 1999 08:25:00 -0700 (PDT) (envelope-from kbracey@e-14.com) Received: from kbracey.acorn.co.uk (kbracey [136.170.129.213]) by mh.acorn.co.uk (8.8.6/8.8.6) with SMTP id QAA22728 for ; Fri, 9 Jul 1999 16:24:58 +0100 (BST) Date: Fri, 09 Jul 1999 16:24:50 +0100 From: Kevin Bracey To: freebsd-net@freebsd.org Subject: Re: interesting connect(2) side effect... Message-ID: <154f91e49%kbracey@kbracey.acorn.co.uk> In-Reply-To: <199907091147.NAA01166@labinfo.iet.unipi.it> X-Organization: Acorn Computers Ltd, Cambridge, United Kingdom X-Mailer: Messenger v1.40f for RISC OS MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Posting-Agent: RISC OS Newsbase 0.61b Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message <199907091147.NAA01166@labinfo.iet.unipi.it> Luigi Rizzo wrote: > > > It is in my opinion totally broken with connected datagram sockets, > > > where the connect has the side effect of in-kernel packet filtering > > > based on source address (who said that you cannot do that!). > > > > It's not a side effect - it's a basic required feature! My TFTP client > > relies on it to filter incoming multicast TFTP channels, which are keyed > > by server address+port. > > i was considering this a side effect because i can't find documentation > mentioning this as a feature -- certainly not the connect(2) manpage. > Not to mention that nothing prevents a second, unprivileged app on > your server send packets on the same UDP port -- if you want to do > a safe filtering you do need a better mechanism. It's not "safe filtering" - it's just the way the protocol works :) It reduces client size/overhead if the kernel filters out the unwanted multicasts for you. Connecting a UDP socket does exactly what I'd expect. It restricts you to sending to particular addr/port pair, and limits you to receiving from that addr/port pair. Further, it also starts picking up ICMP messages related to that addr/port to allow you to detect Destination Unreachable conditions. If it didn't actually filter on source addr/port, then that's half the of the meaning of connect() lost. Now, there might be an argument for saying that connect to a multicast address means you can receive from anything, but I'm sure an equally strong argument for connecting to a multicast address meaning you can receive nothing. -- Kevin Bracey, Senior Software Engineer Pace Micro Technology plc Tel: +44 (0) 1223 725228 645 Newmarket Road Fax: +44 (0) 1223 725328 Cambridge, CB5 8PB, United Kingdom WWW: http://www.acorn.co.uk/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 8:26:55 1999 Delivered-To: freebsd-net@freebsd.org Received: from the.oneinsane.net (the.oneinsane.net [207.113.133.228]) by hub.freebsd.org (Postfix) with ESMTP id F168A14CF9 for ; Fri, 9 Jul 1999 08:26:52 -0700 (PDT) (envelope-from insane@lunatic.oneinsane.net) Received: from lunatic.oneinsane.net (insane@lunatic.oneinsane.net [207.113.133.231]) by the.oneinsane.net (8.9.3/8.9.3) with ESMTP id IAA16914 for ; Fri, 9 Jul 1999 08:26:51 -0700 (PDT) Received: (from insane@localhost) by lunatic.oneinsane.net (8.9.3/8.9.3) id IAA63760 for freebsd-net@freebsd.org; Fri, 9 Jul 1999 08:26:51 -0700 (PDT) (envelope-from insane) Date: Fri, 9 Jul 1999 08:26:51 -0700 From: "Ron 'The InSaNe One' Rosson" To: freebsd-net@freebsd.org Subject: NATD/VPN Message-ID: <19990709082651.B62862@lunatic.oneinsane.net> Reply-To: Ron Rosson Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.5i X-Operating-System: FreeBSD lunatic.oneinsane.net 3.2-STABLE X-Opinion: What you read here is my IMHO X-Disclaimer: I am a firm believer in RTFM X-WWW: http://www.oneinsane.net X-PGP-KEY: http://www.oneinsane.net/~insane/insane-pgp5i.txt X-Uptime: 8:17AM up 8 days, 8:13, 3 users, load averages: 0.21, 0.10, 0.03 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Just wanted to drop a line to both lists letting everyone know that I got the problem resolved with this issue. There was nothing wrong with my implimentation of NATD/VPN using -pptpalias. It was my clients Windows N2 Workstation. He reapplied Service Pack 4 and everything started working. Just goes to show you Microsoft is not very smart ;-). The only issue that is still open is a PR I submitted bin/12433 I might have classified it wrong. If I did please let me know what it should have been. In Closing I want to thank all the people who helped be get to the bottom of this problem even though it wasn't FreeBSD's issue but a MS issue. Thanks again and it is mailing lists like these and the people in them that make this such a great OS to entrust networks to ;-). TTFN -- ------------------------------------------------------------------- Ron Rosson ... and a UNIX user said ... The InSaNe One rm -rf * insane@oneinsane.net and all was null and void ------------------------------------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 11:25:14 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail-out2.apple.com (mail-out2.apple.com [17.254.0.51]) by hub.freebsd.org (Postfix) with ESMTP id 76FD814E10 for ; Fri, 9 Jul 1999 11:25:12 -0700 (PDT) (envelope-from justin@rhapture.apple.com) Received: from mailgate1.apple.com (A17-128-100-225.apple.com [17.128.100.225]) by mail-out2.apple.com (8.8.5/8.8.5) with ESMTP id LAA56306 for ; Fri, 9 Jul 1999 11:25:12 -0700 Received: from scv2.apple.com (scv2.apple.com) by mailgate1.apple.com (mailgate1.apple.com- SMTPRS 2.0.15) with ESMTP id for ; Fri, 09 Jul 1999 11:25:09 -0700 Received: from rhapture.apple.com (rhapture.apple.com [17.202.40.59]) by scv2.apple.com (8.9.3/8.9.3) with ESMTP id LAA12756 for ; Fri, 9 Jul 1999 11:25:08 -0700 Received: by rhapture.apple.com (8.9.1/8.9.1) id LAA00699 for net@FreeBSD.ORG; Fri, 9 Jul 1999 11:25:08 -0700 (PDT) Message-Id: <199907091825.LAA00699@rhapture.apple.com> To: net@freebsd.org Subject: Re: Setting up LAN In-Reply-To: <199907090607.XAA01050@walker3.apple.com> Date: Fri, 9 Jul 1999 11:25:07 -0700 From: "Justin C. Walker" Reply-To: justin@apple.com X-Mailer: by Apple MailViewer (2.105.dev) Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > From: Shawn Workman > Date: 1999-07-08 23:18:18 -0700 > To: "Justin C. Walker" > Subject: Re: Setting up LAN > Cc: net@FreeBSD.ORG > In-reply-to: <199907090607.XAA01050@walker3.apple.com> > X-Mailer: XFMail 1.3 [p0] on FreeBSD > X-Priority: 3 (Normal) > Delivered-to: freebsd-net@freebsd.org > X-Loop: FreeBSD.org > > I forgot to mention that the FreeBSD box is also my firewall.. That helps define the problem, at least. > could I divide my subnet? A simple ascii diagram may shed light: ^ | ;; to the rest of the world --------------+--------------- | ;; inside your place X ;; your FreeBSD box and firewall | ;; your internal subnet Now, your box (X) has two interfaces. From the point of view of the rest of the world, they don't care about anything below the line. They just know that you have subnet 216.18.166.160, which includes 14 hosts and two broadcast addrs (ain't backwards compatibility a joy?). So I think that sub-dividing your subnet may help, although, as you've noted, the power-of-2 thing will cost you some address space (unless you're happy keeping some hosts on the outside of the firewall. The tricky part is that the outside world (represented by a router above the line) thinks of your subnet as just that. Unless it (the upper router) knows that your FreeBSD box is a router, it's going to try to deliver packets to your subnet using ARP, not by forwarding to "X". Proxy ARP may help, together with splitting your subnet. > Maybe have the netmask on the router be a netmask of 16 IP's and the netmask on > my vr1 interface be 16 IP's? I've not used/configured proxy ARP, so I can't be sure how effective it is, or whether it's really the solution. Those with more experience may know. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Manager, CoreOS Networking | When crypto is outlawed, Apple Computer, Inc. | Only outlaws will have crypto. 2 Infinite Loop | Cupertino, CA 95014 | *-------------------------------------*-------------------------------* To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 11:30:35 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail.snowcrest.net (mail.snowcrest.net [216.102.43.227]) by hub.freebsd.org (Postfix) with ESMTP id EFD5414C3F for ; Fri, 9 Jul 1999 11:30:33 -0700 (PDT) (envelope-from djewett@snowcrest.net) Received: from ws2983 (oakfrB009.snowcrest.net [216.102.5.9]) by mail.snowcrest.net (8.9.0/8.9.0) with SMTP id LAA19013 for ; Fri, 9 Jul 1999 11:30:32 -0700 (PDT) Message-ID: <000a01beca39$19f202b0$5515a8c0@co.shasta.ca.us> From: "Derek Jewett" To: Subject: FTP latency.... Date: Fri, 9 Jul 1999 11:30:19 -0700 Organization: Shasta County MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0007_01BEC9FE.6CD5A710" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2014.211 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This is a multi-part message in MIME format. ------=_NextPart_000_0007_01BEC9FE.6CD5A710 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Using 3.1-R on a typical ethernet network.... Running ftpd under inetd = (the super server).. Problem: When users hit the box there is a delay at = the ftp connect screen.. The delay is up to one minute then they get = User: prompt and all works great... Any idea what is causing the delay? = I had the same problem with another box running 3.0-R.. Maybe I should = just run ftpd as a dedicated process, and not through inetd..?=20 thanks ------=_NextPart_000_0007_01BEC9FE.6CD5A710 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Using 3.1-R on a typical ethernet = network....=20 Running ftpd under inetd (the super server).. Problem: When users hit = the box=20 there is a delay at the ftp connect screen.. The delay is up to one = minute then=20 they get User: prompt and all works great... Any idea what is causing = the delay?=20 I had the same problem with another box running 3.0-R.. Maybe I should = just run=20 ftpd as a dedicated process, and not through inetd..?
 
thanks
------=_NextPart_000_0007_01BEC9FE.6CD5A710-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message From owner-freebsd-net Fri Jul 9 11:42:17 1999 Delivered-To: freebsd-net@freebsd.org Received: from mail-out1.apple.com (mail-out1.apple.com [17.254.0.52]) by hub.freebsd.org (Postfix) with ESMTP id 551451567C for ; Fri, 9 Jul 1999 11:42:14 -0700 (PDT) (envelope-from justin@rhapture.apple.com) Received: from mailgate2.apple.com ([17.129.100.225]) by mail-out1.apple.com (8.8.5/8.8.5) with ESMTP id LAA14892 for ; Fri, 9 Jul 1999 11:42:13 -0700 Received: from scv3.apple.com (scv3.apple.com) by mailgate2.apple.com (mailgate2.apple.com- SMTPRS 2.0.15) with ESMTP id ; Fri, 09 Jul 1999 11:42:02 -0700 Received: from rhapture.apple.com (rhapture.apple.com [17.202.40.59]) by scv3.apple.com (8.9.3/8.9.3) with ESMTP id LAA18978; Fri, 9 Jul 1999 11:42:01 -0700 Received: (from justin@localhost) by rhapture.apple.com (8.9.1/8.9.1) id LAA00714; Fri, 9 Jul 1999 11:42:01 -0700 (PDT) Message-Id: <199907091842.LAA00714@rhapture.apple.com> To: Derek Jewett Subject: Re: FTP latency.... Cc: freebsd-net@freebsd.org Date: Fri, 9 Jul 1999 11:41:59 -0700 From: "Justin C. Walker" Reply-To: justin@apple.com X-Mailer: by Apple MailViewer (2.105.dev) Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > From: Derek Jewett > Date: 1999-07-09 11:30:38 -0700 > To: freebsd-net@FreeBSD.ORG > Subject: FTP latency.... > > Using 3.1-R on a typical ethernet network.... Running ftpd under inetd (the > super server).. Problem: When users hit the box there is a delay at the ftp connect > screen.. The delay is up to one minute then they get User: prompt and all works > great... Any idea what is causing the delay? I had the same problem with another > box running 3.0-R.. Maybe I should just run ftpd as a dedicated process, and not > through inetd..? That kind of delay is usually indicative of a "reverse name lookup", which the server does to try and validate the incoming connection. It usually means that your client has an unregistered IP address (e.g., a MacIP- or DHCP-assigned address, with no associated host name). Of course, it could be something completely different :-}. The only way to be sure is try to debug the server. One way to do this is with a dedicated server - get a debug version that doesn't fork on accept, and see where the time goes. Regards, Justin -- Justin C. Walker, Curmudgeon-At-Large * Institute for General Semantics | Manager, CoreOS Networking | Men are from Earth. Apple Computer, Inc. | Women are from Earth. 2 Infinite Loop | Deal with it. Cupertino, CA 95014 | *-------------------------------------*-------------------------------* To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message