advanced ping graphs - weird loss data

[drewshin99]

Hi all, running into a problem I can't seem to figure out. I'm using the Advanced Ping template/graph and have it setup to send 5 pings. See log output below:

08/03/2010 08:11:17 PM - SPINE: Poller[0] Host[1416] DS[4876] SS[0] SERVER: C:/Apache2/htdocs/cacti/scripts/ss_fping.php ss_fping x.x.x.x 5 ICMP , output: min:778.9700 avg:1124.3475 max:1653.1800 dev:337.6253 loss:20.0000

The loss shows correctly as 20% here. However, when I view the graph and the rrd file, it seems to split it up into two data points.

1280865900: 1.7066666667e+001 3.7786680000e+002 4.9033826667e+002 4.3551262667e+
001 4.3756378667e+002
1280866200: 2.0666666667e+000 2.7165160000e+002 3.7584416667e+002 3.9524725333e+
001 3.1003990000e+002

See above, the 20% loss got put into the RRD file as 17.0666% and 2.0666%. The graph also shows these split values as well.

Can anyone give me any direction on how to fix this? I'm thinking it must be something in the "Data Input Method" or "Data Template" that I have hosed up somehow, because the output from the script is ok.

Thanks!

[gandalf]

I assume that this is due to an off-boundary rrdtool update. Integers are preserved on rrdtool updates only, if they exactly match 300 sec boundaries
R.

[drewshin99]

thanks gandalf for your response. i'm looking at my advanced ping graphs now, and the rrd file has data that's about 3-5% off from what's actually measured by the ss_fping script, just in case anyone else is wondering why their advanced ping graphs are a little off. i saw in another post you or tobi? mentioned that you could fool rrdtool into thinking that each update was made at exactly the 300 second interval if needed, but i don't want to mess with this little thing to break what's already a perfectly fine product for me