IP SLA Operation Failing Although Link latency is Less than timeout

[mfaisal02tl]

We are tracking icmp-echo of remote IP of high BW link under IP SLA and then tracked in an HSRP. Here is the config details of IP SLA:

ip sla 10

 icmp-echo x.x.x.x source-ip y.y.y.y

 timeout 1000

 threshold 500

 frequency 3

ip sla schedule 10 life forever start-time-now

track 20 ip sla 10 reachability 

We have measured the icmp latency on our WAN link and it's not reaching more than 6 msec and we observe no request time outs icmp packet on our WAN link. Furthermore, we have setup the timeout value to 1000 msec (1 sec). However, for some reason, the IP SLA operation is continuously flapping thus causing fail over to HSRP as well. Since, the interesting traffic is real time, we have setup the frequency and timeout parameters accordingly. 

Based on above, can you please guide us why IP SLA is flapping although timeout is not actually reaching and it's value is way above the actual latency (6 msec). 

Switching, Other Switches

Regards

[Deepak Kumar]

Hi,

Increase the threshold value. It is seems lower then link latency.

[mfaisal02tl]

Thanks Deepak for your feedback.

Link latency in our case is maximum 6 msec, whereas threshold value we have setup is 1000 msec (1 sec). Which is much more than link latency.

[Georg Pauwen]

Hello,

post the output of:

show ip sla statistics

Also, try the values below. A default PING timeout originated from the router is 2 seconds, the default timeout and threshold for the SLA is 5 seconds. If you set the timeout to 1 second, it is less than the default on the router, and that might cause the continuous flapping...

ip sla 10
icmp-echo x.x.x.x source-ip y.y.y.y
timeout 2500
threshold 1000
frequency 3

ip sla schedule 10 life forever start-time-now

track 20 ip sla 10 reachability

[mfaisal02tl]

Thanks Georg of your response.
Please find below output you asked for;

sh ip sla statistics
IPSLAs Latest Operation Statistics

IPSLA operation id: 1
Latest RTT: 3 milliseconds
Latest operation start time: 17:22:22 UTC Sun Nov 3 2019
Latest operation return code: OK
Number of successes: 504
Number of failures: 11
Operation time to live: Forever

I will change the timeout and threshold as per your advice and will observe if that makes any improvement. I will keep posting the outcome.