Have a basic understanding but don’t have a lot of experience deploying multicast so hoping somebody can help clear up some confusion.
If IGMPv3 isn’t supported for whatever reason then PIM-SSM can’t be used and PIM-SM is the next best choice. Holy crap it is way more complex.
The part that isn’t clear to me is how to handle different sources using the same IGMP group addresses if having to use PIM-SM. These sources are in different subnets/vlans/svi but it seems to me PIM-SM is one giant multicast domain and you have to make sure no two sources are using the same igmp group or else a receiver will get traffic from both sources. I came to the conclusion due to receivers sending igmp joins to *.g in PIM-SM whereas in PIM-SSM the source is specified.
It seems to me there would be some type of way to create multiple PIM-SM instances to segment out multicast sources or somehow identify which source the receiver is looking for. This doesn’t seem to be possible with PIM-SM and the more I think about it wouldn’t make sense since there is no way to tell which source a receiver wants. Which starts to sound more like PIM-SSM.
Hopefully somebody can confirm or fill me in on something I’m missing.
No comments:
Post a Comment