After the implementation of CDMA2000 core network connections the CS part complain from CPU over load and duplicate IP address alarm inside CSOFT and the service is stopped and all the connections with the BSS, HLR and UMG is interrupted. this problem happen after the connection of the router to the LAN switches.
There is no alarm information in the routers
The handling of the above case can be implemented by remove the pass permission for VLAN 40 from the routers or from the switches so the VRRP message will not looped between the routers and switches more again and stopped since there is only one trunk allow them to be passes.
port trunk allow-pass vlan 20 110 200
After that the CSOFT IP duplicate alarm is cleared and the CPU overload is cleared also and the service is returned back to work normally.
First we checked the IP address duplicate we found that there is no identical IP address in the network similar to the IP address of the CSOFT and we found that the alarm is still exist, we checked the CPU load of the LAN switches that connect the signaling from the CSOFT to the BSS, HLR and UMG we found that the CPU is over loaded and the RAM is high usage, so we can say that there is some type of internal storm inside the network causes this problem.
We began to check the physical connection we found that there are two LAN switches and two NE40 routers and LAN SW1 is connected to Router1 and LAN SW2 is connected to Router2 using trunk configuration and this trunk according to the configuration checking we found is allow just the signaling VLAN (VLAN 40) to be pass:
port trunk allow pass vlan 40
Also the two LAN switches are connected each together using trunk and allow the signaling network (VLAN 40) to pass through this trunk.
port link-type trunk
undo port trunk permit vlan 1
port trunk permit vlan 40 50 to 51
Also we found that the routers are connected each together using trunk and this trunk allow all the network to be pass:
port trunk allow-pass vlan 1 to 4094
So the VRRP messages between VLAN interfaces 40 between the two routers and between the two switches are transferred through the trunk between the switches and through the trunk between the routers, and caused internal loop for this messages.
During the implementation of core network we should take care for the trunk configuration because if the physical connection is correct the wrong configuration can cause packet loop and effect to the network operation.