Hi all,
When viewing All Protocols > Throughput and sorting the Current column under Data Descending (Top Talkers at top) I often see throughputs WAY off from reality. The bytes sent/rcvd counters are accurate enough - so I'm curious why the Throughputs would be so far off? I thought perhaps it had to do with netflow export timeouts and the test xfers I was doing were too short. So, using Iperf I sent a 500Kb/s udp stream between two hosts for 5 minutes. Iperf counters and Cisco interface counters both agreed the stream was ~ 500Kb/s, yet nTop showed everything from 20Kb/s near the beginning of the test to 1.7Mb/s near the end of the test. It was dead on at 500Kb/s a couple times - but why such disparity? Could it have to do with the refresh interval of the browser and the netflow export timers? How is the "Throughput" actually calculated? I guess I could tear into the code - but figured I'd throw it out to you guys first. I use netflow almost exclusively and it would be nice if the throughput values were close. My timers are as follows: nTop -r = 120. I did this because I would be analyzing data on a page and it kept refreshing and going to the top of page. I'd scroll down to look at some data and it would do this again. VERY annoying! Netflow: Active flows = 60secs, Inactive = 10 secs. These are the lowest timeouts available. Thanks! Gary =========================================================================== "This email is intended to be reviewed by only the intended recipient and may contain information that is privileged and/or confidential. If you are not the intended recipient, you are hereby notified that any review, use, dissemination, disclosure or copying of this email and its attachments, if any, is strictly prohibited. If you have received this email in error, please immediately notify the sender by return email and delete this email from your system."
_______________________________________________ Ntop mailing list [email protected] http://listgateway.unipi.it/mailman/listinfo/ntop
