IGMP join/leave messages cause switch/router cpu surge

[pengcisco]

We have multicast confiugred on campus network. Edge switches (layer2) have uplinks to distribution switches (layer3 cat6506). The distribution switch have ip pim configured under vlan interface and edge switch configured with igmp snooping. Recently, observed a large amount of IGMP join and leave message from edge switches. The source looks like from all edge switches and destination multicast IP is not fixed. The large amount of igmp messages cause our egde switches and distribution switches cpu surge to 99%. After take out the ip pim configuration from the vlan interface, the cpu will back to normal. The igmp rate limiting and access group doesn't work as the destiniation multicast IP is not fixed. Any suggestion to prevent this kind of traffic?

[sadbulali]

Find whether the ip pim accept-rp command was previously configured throughout the network by using the show running-config privileged EXEC command. If the ip pim accept-rp command is not configured on any device, this problem can be addressed later. In those routers or multilayer switches already configured with the ip pim accept-rp command, you must enter the command again to accept the newly advertised RP.