Afternoon all,
I have been experiencing a problem when using an IP SLA udp-echo threshold value to trigger another IP SLA udp-echo. This logic worked perfectly fine on a 2811 running c2800nm-adventerprisek9_ivs-mz.124-22.T but moving to a 2921 with c2900-universalk9-mz.SSA.153-2.2.T I now see an entirely different logic in the software.
My design involves an sla performing a udp-echo across a tunnel. This tunnel can route via a high or low latency link, the echo will return a RTT which is used to determine whether certain configuration changes limiting telephony connections are implemented or removed.
The sla implementation uses a 'slow ticker' sla firing an echo every 10 seconds, a trigger is configured to detect 5 consecutive echos over a rising threshold of 400ms and then action an second 'fast ticker' sla, which lives for 15 seconds and fires out an echo every 1 second with a threshold of 400ms. A track is set to monitor this second sla, and on detection of a 400ms echo, the track goes down, delays for 10 seconds before reporting to the event manager to invoke the config changes.
On the reinstatement of the low latency link, the trigger on the first sla has a falling threshold of 200ms, which then triggers the second sla, the track now sees an up state, delays for 10 seconds before reporting to the event manager to make the changes back.
The issue, which has only presented itself on the 2921 with IOS 15.0 is that once the second sla has been run once, it never returns to the pending state with its 15 second time to live. As it never returns to the pending state, it is not available to be triggered when the link is reinstated and as such is only ever run once.
My question: Is this a known bug on the IOS 15.0? As there is a clear behavioral difference from the 2811 running 12.0.
Configuration below
!
ip sla 1
udp-echo 10.100.1.1 80 source-ip 10.100.1.2 source-port 80 control disable
threshold 1000
timeout 1000
frequency 10
!
ip sla schedule 1 life forever start-time now
!
ip sla 2
udp-echo 10.100.1.1 80 source-ip 10.100.1.2 source-port 81 control disable
threshold 400
timeout 1000
frequency 1
!
ip sla schedule 2 life 15 start-time pending
!
ip sla reaction-trigger 1 2
ip sla reaction-configuration 1 react rtt threshold-value 400 200 threshold-type consecutive 5 action-type triggerOnly
!
track 2 ip sla 2
default-state up
delay down 10 up 10
!
event manager applet HIGH_DELAY
event track 2 state down
action 1.0 snmp-trap strdata "HIGH_DELAY"
!
event manager applet LOW_DELAY
event track 2 state up
action 1.0 snmp-trap strdata "LOW_DELAY"
Any help would be greatly appreciated.