A patch addressing this issue has been included in kernel-2.6.9-89.32.1.

As steps in https://bugzilla.redhat.com/show_bug.cgi?id=637764#c20, I cannot reproduce and verify this bug for RHEL4.8.z.
With Flavio's help, I manage to do this in a new method.

The host is dell-pe2850-04.rhts.eng.bos.redhat.com. First make sure eth0 and eth1 both have cables connected.
1. Configure following nics:
# cat /etc/sysconfig/network-scripts/ifcfg-bond0
DEVICE=bond0                                                                    
ONBOOT=yes                                                                      
BOOTPROTO=dhcp

# cat /etc/sysconfig/network-scripts/ifcfg-bond0.10 
DEVICE=bond0.10
ONBOOT=yes                                                                      
REORDER_HDR=no                                                                  
VLAN=yes                                                                        
BOOTPROTO=static                                                                
IPADDR=192.168.55.63                                                            
NETMASK=255.255.255.0

# cat /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0                                                                     
ONBOOT=yes                                                                      
MASTER=bond0                                                                    
SLAVE=yes
HOTPLUG=no

# cat /etc/sysconfig/network-scripts/ifcfg-eth1
DEVICE=eth1
ONBOOT=yes                                                                      
MASTER=bond0                                                                    
SLAVE=yes
HOTPLUG=no

2. Then over console,
# service network restart

3. In another ssh1,
# tcpdump igmp -i eth0

4. In another ssh2,
# tcpdump igmp -i eth1

5. Then over console,
# ./multicasting-client.py

Then currently active slave nic(eth0 in my case) from tcpdump should see something like
11:50:08.361289 IP 192.168.55.63 > IGMP.MCAST.NET: igmp v3 report, 1 group record(s) 

6. In another ssh3,
# ifenslave -c bond0 eth1

Then another nic(eth1 in my case) from tcpdump should very soon see something like 
11:52:24.313758 IP 192.168.55.63 > IGMP.MCAST.NET: igmp v3 report, 1 group record(s)

This igmp v3 report should be the same with the one in step 5. And this denotes the bug is fixed.

Created attachment 463779 [details]
multicasting-client.py

The script is from fleitner.

An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2010-0936.html