perfSONAR User Q&A and Other Discussion

[Larry Blunk <>]

   We have ping tests configured between our hosts and
mostly just look at the RTT latency and use owamp just for loss.
The ping RTT latency times are more stable than the owamp
one-way times and are good enough for our purposes.   You can
use the Nagios check_delay script to get the ping RTT latency times
(instead of check_owdelay).  Here's a comparison of the output
between sites in Ann Arbor and Detroit (where check_owdelay
shows negative latency due to NTP clock sync issues).  The
range of the check was set to one day.


PS_CHECK_OWDELAY CRITICAL - Average minimum delay is -1.258ms | 
Count=1436;; Min=-2.5;; Max=-0.7;; Average=-1.25849582172704;; 
Standard_Deviation=0.216893414590156;;

PS_CHECK_DELAY OK - Average minimum delay is 1.452ms | Count=289;; 
Min=1.39;; Max=1.61;; Average=1.45242214532872;; 
Standard_Deviation=0.0236514401040235;;

 -Larry Blunk
  Merit

On 03/30/2016 04:22 PM, Michael Reece wrote:
> I know that this issue has been raised before, but perhaps someone has a
> solution to it. For some reason the latency times between several of the
> nodes in my mesh are negative. I've been told before that this is likely
> due to the close proximity of the nodes in the mesh (most are less than
> a mile apart). This wouldn't really be a problem except for the fact
> these negative times show up red in the MaDDash latency grid. The legend
> states that only time of 10ms and greater will have the color red, but
> clearly the negative values are showing up red as well. Here is a
> screenshot of my latency grid on MaDDash:
> Inline image 1
> Has anyone found a solution to this annoying problem? I would appreciate
> any help.
>
> Thanks,
> Mike Reece