Date: Thu, 18 Feb 2010 11:00:25 -0800 From: Stephen Hurd <hurds@sasktel.net> To: Harald Schmalzbauer <h.schmalzbauer@omnilan.de> Cc: Patrick Mahan <mahan@mahan.org>, freebsd-stable@freebsd.org, Stephen Hurd <shurd@sasktel.net> Subject: Re: RELENG_8 ignoring TCP window size? [Was: Re: Help for TCP understanding wanted, ACK-MSS-Window [Was: Re: best practice to watch TCP parms of established sockets]] Message-ID: <5c00c4d917943.4b7d1dc9@sasktel.net>
next in thread | raw e-mail | index | archive | help
This is a multi-part message in MIME format. --Boundary_(ID_kDTlApuHY4uYbHwcgladwg) Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT Content-disposition: inline All those duplicate ACKs look like a problme to me... I would be inclined to capture from the other end or from a monitor port on the switch to make sure those are actually hitting the wire. The only explination I can think of is that the system sending the duplicate ACKs is assuming that it's lost a segment for some reason and is trying to trigger fast recovery early. --Boundary_(ID_kDTlApuHY4uYbHwcgladwg) Content-type: multipart/mixed; boundary="Boundary_(ID_K7VhZx6vpNouByAbQ57Thg)" --Boundary_(ID_K7VhZx6vpNouByAbQ57Thg) Content-type: text/plain; charset=ISO-8859-1; format=flowed Content-transfer-encoding: 7BIT Stephen Hurd schrieb am 18.02.2010 17:18 (localtime): ... > one retransmit per five packets). Is this what you're seeing? If you > have a capture you could share covering a few seconds, I could take a > look and provide a better opinion. Thanks for your help, here's a small snippet of the iperf session. There are hundreds od duplicate ACKs. I'm not sure how to understand that. Like mentioned, I have to refresh some TCP basics before I can do usefull tests. But perhaps you can confirm that this behaviour is intendend, or where the problem could be... Thanks, -Harry --Boundary_(ID_K7VhZx6vpNouByAbQ57Thg)-- --Boundary_(ID_kDTlApuHY4uYbHwcgladwg)--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5c00c4d917943.4b7d1dc9>