[smokeping-users] capacity issues with CiscoRTTMonEchoICMP probe

[Dez Cadena]

Hello all,
I've been using the CiscoRTTMonEchoICMP probe to produce COS graphs. I'm 
using a cisco3845 as my SAA probe
and a linux server (dual xeon2.8G, 2G RAM, fast SCSI disks) for 
smokeping. I've been monitoring 4 hosts x 4 classes
of service = 16 targets. No problem so far.

When I tried to increase the number of hosts from 4 to 20 (i.e. 80 
targets), smokeping couldnt complete *any* round
in 300 seconds:
"smokeping[14594]: CiscoRTTMonEchoICMP: WARNING: smokeping took 631 
seconds to complete 1 round of polling.
It should complete polling in 300 seconds. You may have unresponsive 
devices in your setup"
No graphs are produced of course...

Any ideas how to resolve the issue? I've thought about having 2 or 3 
probes with different offsets but still I dont think
it would scale up as my goal is to have 500 targets. I suppose I could 
reduce the number of pings from 25 to 15 but I'd
rather not do that.

Any input would be appreciated,

Here's a relevant sample of my smokeping config:
*** Probes ***
+ CiscoRTTMonEchoICMP
pings = 25
packetsize = 56
ioshost = community at saa-probe-router

+++ nodeA

menu = nodeA
title = nodeA

++++ nodeA-A
menu = classA
host = 1.2.3.4
tos = 100

++++ nodeA-B
menu = classB
host = 1.2.3.4
tos = 99

++++ nodeA-C
menu = classC
host = 1.2.3.4
tos = 98

++++ nodeA-D
menu = classD
host = 1.2.3.4

--
Unsubscribe mailto:smokeping-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:smokeping-users-request at list.ee.ethz.ch?subject=help
Archive     http://lists.ee.ethz.ch/smokeping-users
WebAdmin    http://lists.ee.ethz.ch/lsg2.cgi