Can we close the rate limitation for icmp packets which be sent to CPU?

17

it can not support.
for chassis switch,we can use commands
display cpu-defend configuration [ packet-typepacket-type ] { all | slotslot-id| mcu} to check the rate�?for box switch:we can use commands
display cpu-defend configuration [ packet-typepacket-type ] [ all | slotslot-id ] to check the rate.

Other related questions:
Why is the rate limit for ICMP packets configured on an interface card inconsistent with the actual rate limit
An AR router implements rate limit to Layer 3 unicast packets by converting the number of packets into the number of bytes. ICMP packets are converted based on 84 bytes per packet. For example, if the rate limit configured for ICMP packets is 10, then the actual rate limit is calculated as 84 x 10 = 840 byte/s. If the actual packet length is not equal to 84 bytes, the actual rate limit will deviate from the configured value.

Why is the actual rate limit different from the ICMP packet rate limit configured on the interface board
AR series routers convert the number of packets to a number of bytes when limiting the rate of Layer 3 unicast packets on the interface board. Each Internet Control Message Protocol (ICMP) packet is converted to 84 bytes. For example, if the ICMP packet rate limit is configured as 10 packets per second, the actual rate limit is 840 (84 x 10) bytes per second. If the actual length of ICMP packets is not 84 bytes, the actual rate limit is different from the configured value. -If the actual length of ICMP packets is greater than 84 bytes, the actual rate limit is less than the configured rate limit. For example, if the ICMP packet rate limit is configured to be 10 packets per second, and the actual packet length is 320 bytes, the actual rate limit is 840/320 (rounded down) packets per second, that is, 2 packets per second. -If the actual length of ICMP packets is less than 84 bytes, the actual rate limit is greater than the configured rate limit. For example, if the ICMP packet rate limit is configured to be 10 packets per second, and the actual packet length is 64 bytes, the actual rate limit is 840/64 (rounded down) packets per second, that is, 13 packets per second.

Do S series switches support rate limiting based on IP addresses
S series switches (except S1700 switches) do not support rate limiting based on IP addresses.

How to set an interval for sending LBDT packets on an S series switches and how to distinguish LBDT packets sent by different interfaces
S series (except the S1700) and E series switches send LBDT packets in the following modes: 1. For fixed switches in V100R005 - An LBDT packet is sent every 100 ms. You can run the loopback-detect interval interval-time command to set an interval for sending LBDT packets. In the command, the value of interval-time ranges from 5 to 300, in seconds. By default, the interval for sending LBDT packets is 30s. - LBDT packets sent by different interfaces are distinguished based on protocol IDs carried in the packets. By default, the protocol ID of interface 1 on the S3700 is 0x606 and the protocol IDs of the other interfaces are incremented. You can run the loopback-detect protocol protocol-id command to manually set a protocol ID. Note that the protocol ID cannot be the same as an existing protocol ID. For details, see the command reference. 2. For fixed switches in V100R006 and later versions, modular switches, and E series switches - An LBDT packet is sent every 5s. You can run the loopback-detect packet-interval packet-interval-time command to set an interval for sending LBDT packets. The value of packet-interval-time ranges from 1 to 300, in seconds. - LBDT packets sent by different interfaces are distinguished based on interface indexes carried in the packets. Note: LBDT packets are sent frequently. Therefore, the CPU usage will increase if the LBDT function is enabled on all interfaces.

If you have more questions, you can seek help from following ways:
To iKnow To Live Chat
Scroll to top