So, I need to measure rtt levels between two devices and trap when it is too high. Sounds like a job for IP SLA right. I am trying to do this on a 7200 and a 6500. When the threshold goes above 30ms I want to send an SNMP trap with custom text, and another SLA for a threshold above 50ms with custom text. I have setup on SLA on the 7200 for a 30 to 40 ms rtt threshold which shows failures but does not send a trap. Here is my config :-
ip sla logging traps
ip sla 1
ip sla reaction-configuration 1 react rtt threshold-value 40 30 action-type trapOnly
ip sla schedule 1 life forever start-time now
snmp-server enable traps ipsla
I can see that the SLA is failing sometimes but does not send a trap as below :-
SLA1#sh ip sla statistics
Round Trip Time (RTT) for Index 1
Latest RTT: 72 milliseconds
Latest operation start time: *12:45:17.355 UTC Thu Oct 8 2009
Latest operation return code: Over threshold
Number of successes: 43
Number of failures: 75
Operation time to live: Forever
Coms to my SNMP trap collector are good as I can shut down an interface and get a trap.
Code version is 12.4(15) Advanced Enterprise.
Is my config broken or is there something more sinister at work here :-s.
I found the answer. As expected, it was broken config.
My re-action config line missed out the reaction type. It should have read :-
ip sla reaction-configuration 1 react rtt threshold-value 60 30 threshold-type immediate action-type trapOnly
Interesting. I haven't really used IP SLA before but it looks like I can use it for proactive troubleshooting before it becomes an issue.