Skip to Content.
Sympa Menu

perfsonar-user - Re: [perfsonar-user] Different results from iperf3 vs. bwctl -T iperf3

Subject: perfSONAR User Q&A and Other Discussion

List archive

Re: [perfsonar-user] Different results from iperf3 vs. bwctl -T iperf3


Chronological Thread 
  • From: Brian Candler <>
  • To: Aaron Brown <>
  • Cc: "" <>
  • Subject: Re: [perfsonar-user] Different results from iperf3 vs. bwctl -T iperf3
  • Date: Thu, 11 Dec 2014 16:13:12 +0000
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=pobox.com; h=message-id:date :from:mime-version:to:cc:subject:references:in-reply-to :content-type:content-transfer-encoding; q=dns; s=sasl; b=Nen4SN g3Sal3n3XFggdBREKS5DudpSpPoFxp6kWxTpY/Mzj7H64hUyWI+Q0juZgnEaXD8D VEERYYWCyJny60vyNJ60tGFxLicsLz7tu1AqB1c6gZ0Qv2kO34ZXxsV98FDYt3l4 vu42OtqI5cH7EJ591RuPLfknBRGJ9toedtoKM=

On 11/12/2014 13:56, Aaron Brown wrote:
That’s just bizarre. These hosts weren’t running NTP, right? I can only guess
that bwctl is doing something funky with error estimates when NTP isn’t
running.
Correct. Well, they are running ntpd, but are unsynced due to port 123 blocking. All peers show state .INIT.

I did not try making them peers to each other.

> Are these boxes going to be running NTP in the field? If so, hopefully that may fix things...

Yes, they will have a proper unfiltered Internet connection, and as you say they may be fine. I did notice a posting from someone else a while back about performance tests only being logged in one direction, but that could be an unrelated problem. It's frustrating when I see a problem and can't pin it down; I was up to midnight last night on this :-)

Anyway, I'm just about to pack up the boxes for shipping. Right now they are telling me that the test will take 15 seconds, but it actually finishes in 6.5 seconds with no results returned. owping is showing a latency of -6/6ms, so clearly the clocks are pretty close to each other, and well within the default error estimate of 0.1s.

Regards,

Brian.




Archive powered by MHonArc 2.6.16.

Top of Page