The objective is to monitor initially if the multicast(for specified application) is available on the site. We have an AMD on that sites.
The first idea is to monitoring the traffic to the Server IP(Application Multicast range).
Aha - Sorry I got that a bit backwards....
You should be fine doing that, however remember that depending on how you get the data into the AMD is crucial. If you like many others use SPAN or similar feature, that might result in some problems for you since not all port mirroring able switches and OS'es are able to handle multicast. That's one reason why I always try to get a passive TAP instead. It has other benefits as well such as you will see corrupt packets (SPAN discards them) and you will be 100% sure that the SPAN feature or the Switch isn't playing tricks on you such as dropping packets or worse.
Good points. I'm really using SPAN, despite the customer knows the drawbacks of this solution.
we're using "default software services" and my doubt is if we have another best way to monitoring.
OK - If you're using "default software services" then I guess you're ending up with quite a number of them?
Just keep a keen eye on them and remember to use "Accepted Server IP Range" and if you can, "Accepted Client IP Range"
I also have to admit that I've never actually configured a named application based on multicast, but somewhere in the back of my mind there is something about dynamic IP address assignment for multicast but I might be wrong.
Coming back to yuor SPAN config. Depending on your network config, you could use VACL instead and thereby get better granularity on the data you pass to the AMD. But if you are using "Default SS" then I assume you want to keep an eye on a given WAN port or something and then you'll be better off with SPAN (but best with a TAP).