Does anyone out there know if it is possible to track an object's availability (as part of a routing policy) but to specify the TTL of the ping packet used to track that object's availability?
I am trying to prefer one route over another for specific traffic and want to monitor the IP address of a remote end of an attached L3 link BUT there is also another route to the remote end meaning that the ipEcho monitor will always get a response even if the preferred route is down making it useless in failing over to the route in the routing table in the event the preferred link is lost.
NB - Using a preferred egress interface as part of the policy cannot be used in this particular case.
Any help or alternative ways to achieve the same aim gratefully received. Cheers
In your SLA, can you specify a specific source interface for ICMP and then create an ACL that denies ICMP from only that specific IP?
How would specifying the TTL of SLA ping force the ping test on primary path?
Is it because the destination is less hops away on the primary path than on the secondary ?
You could do a local policy that sends the pings to the primary path next-hop or do a static host route for ping destination that goes over the primary path.
Don't forget to rate helpful posts.
yes that's right the primary path is significantly fewer hops that the alternate one. Your post has got me thinking in a different way (which is good) so thanks for the suggestion. I don't have to have definitive solution for a few weeks yet so will leave this post open in case someone else has alternate methods of achieving the same thing. Thanks again..