RE: specifying finite runtime

It turns out that I had a SIGINT also killing the client after 5 seconds. The server then sent data until it timed out the client. Without the SIGINT, the client and server close synchronously.

It is not necessary to specify the transfer time on both the server and the client. Sorry for the inconvenience.

Dave

From: owner-iperf-users --at-- dast.nlanr.net [mailto:owner-iperf-users --at-- dast.nlanr.net] On Behalf Of Jeff Schultz
Sent: Wednesday, April 28, 2004 4:44 PM
To: iperf-users --at-- dast.nlanr.net; iperf-users --at-- dast.nlanr.net
Subject: Re: specifying finite runtime

I'm pretty sure you have to specify the -t option for the server as well if doing bi-directional. The default transfer time is 10 seconds so it sounds like you're hitting that.

Jeff

At 03:59 PM 4/28/2004, Craig, Dave wrote:

I’m running dual mode tests and I noticed that the server does not run for the same length of time as the client. For instance,

server% iperf –uVsB 3ffe:502:ffff::2

client% iperf –c3ffe:502:ffff::2 –V –t5 –du –l128

The client transmits for 5 seconds, and the server transmits for 10 seconds. Am I overlooking something in the command-line arguments?

Thank you,
Dave Craig

QUALCOMM Incorporated