[smokeping-users] Slave loss computation broken?

[Lars Hecking]

 Take a look at the attached graph from a smokeping slave. Although there
 clearly are more than two dozen periods of packet loss, the legend says
 0.00% for avg/max/min/now packet loss. I've not seen this for any master
 graphs.

 Out of 30 slave graphs, only 8 or so show the problem, but the probe in
 question is always AnotherDNS with a record type of MX. Not all probes
 of this type exhibit the problem. E.g. for one particular probe:

 - overview/3h/30h graph wrong
 - 10day/400day graph correct (non-zero loss)

 All graphs based on a plain DNS probe seem to be correct.

 Config:

+ AnotherDNS

offset = 50%
step = 300
pings = 5

+++ FOO_DNS_MX

probe = AnotherDNS
menu = foo_dns_mx
title = DNS MX on FOO
host = 1.2.3.4
lookup = bar.company.com
recordtype = MX
slaves = host.sub.company.com

 smokeping 2.4.1
 rrdtool 1.2.27 (1.3 doesn't build on RHEL3)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smokeping.png
Type: image/png
Size: 21892 bytes
Desc: not available
Url : http://lists.oetiker.ch/pipermail/smokeping-users/attachments/20080702/229eb1c8/attachment-0001.png