From owner-freebsd-stable Fri Apr 26 23:12:12 2002 Delivered-To: freebsd-stable@freebsd.org Received: from pintail.mail.pas.earthlink.net (pintail.mail.pas.earthlink.net [207.217.120.122]) by hub.freebsd.org (Postfix) with ESMTP id 4441B37B419 for ; Fri, 26 Apr 2002 23:12:06 -0700 (PDT) Received: from 209-239-194-210.oak.jps.net ([209.239.194.210] helo=rover) by pintail.mail.pas.earthlink.net with smtp (Exim 3.33 #2) id 171LR2-0001SZ-00 for freebsd-stable@freebsd.org; Fri, 26 Apr 2002 23:11:59 -0700 From: "Sameer R. Manek" To: Subject: PPP under 4.5-stable Date: Fri, 26 Apr 2002 23:11:40 -0700 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Has anyone else seen excessively high latency across a modem based PPP link? I have I have noticed that pings to the upstream router goes into the 6000ms range. This is especially when I'm downloading something in another window, or viewing an image intensive webpage. I have tried altering the MTU of the tun0 interface, a long with adding this to ppp.conf: set mtu 1000 But it doesn't seem to have any noticeable effect on the latency shown below. I don't think this is an issue with line noise, as i don't notice the conditions associated with retraining modems. This also seems to have been introduced after i upgraded to 4.5. I also didn't notice this problem when this same system was running 4.4-stable --- 209.239.192.11 ping statistics --- 115 packets transmitted, 111 packets received, 3% packet loss round-trip min/avg/max/stddev = 1370.107/6053.621/8329.974/1492.298 ms I have also noticed that during periods of "high traffic", about 6000cps, the load average kicks up to close to 1, though it's certainly not using cpu much, so it's probably a reflection of fact that ppp sits in the run queue, an io-wait state. last pid: 92042; load averages: 0.81, 0.76, 0.67 up 5+05:36:20 23:06:48 45 processes: 2 running, 42 sleeping, 1 stopped CPU states: 1.6% user, 0.0% nice, 0.0% system, 0.0% interrupt, 98.4% idle Mem: 46M Active, 433M Inact, 77M Wired, 37M Cache, 73M Buf, 33M Free Swap: 1265M Total, 1265M Free PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND 92042 manek 28 0 1988K 1204K RUN 0:00 1.07% 0.63% top 86648 root 2 0 2540K 1684K select 2:58 0.10% 0.10% ppp Sameer ------------------------------------------------------------------------- Sameer R. Manek Email: manek@ghur.net "What one has not experienced, one will never understand in print." --Isadora Duncan ------------------------------------------------------------------------- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message