In the Customer network 0.2% of PING packets loss when PING between NE40E a Cisco router via Cisco ATM Switch (IGX). check the attachment for more information.
There is no any problem with normal ping between cisco router and Huawei NE40E the only problem happened when ping big size packet from cisco router to huawei NE40E one direction problem . the reason for that is because Cisco router ping speed is very fast (1ms) which generate high bandwidth more than 8Mbps. NE40E CPU has protoction policy (CP-CAR) which limite the bandwidth for PING packet to 4M to avoid PING attack . so when the PING packet exceed the limiteation the CPU will drop it .
[SANC-NE40E-01]cpu-defend policy 10
cpu-defend policy 10
[SANC-NE40E-01-cpu-defend-policy-10]car icmp cir 9000
[SANC-NE40E-01]display cpu-defend car icmp statistics slot 2
Slot : 2
Application switch : Open
Default Action : Min-to-cp
Protocol switch: Open
Passed packet(s) : 421311
Dropped packet(s) : 300
Configged CIR : 9000 kbps Actual CIR in NP : 9000 kbps
Configged CBS : 40000 bytes Actual CBS in NP : 40000 bytes
Priority : middle
Min-packet-length : 128bytes
we did the following operation to analysis the problem in the site where the customer bandwdith is 10M the PING packet still has packet loss when PING packet size 1000 :
a.Add optical power attenuator.
b.Change the ATM subcard.
c.CPU-CAR for protection propose drop bandwidth exceeded PING packets .check the attachment for more details.