Description

This article provides information about the errors not generated because old message was icmp counter, which is generated in the output of show system statistics icmp .

Symptoms

In the output of show system statistics , the errors not generated because old message was icmp counter is generated under ICMP. This article describes the counter and provides an example of the counter increasing.

Solution

The output of show statistics icmp :
jun@EX2200> show system statistics icmp
  icmp:
    0 drops due to rate limit
    0 calls to icmp_error
    100 errors not generated because old message was icmp
    Output Histogram
      449787 echo reply
      230734 destination unreachable
    0 messages with bad code fields
    0 messages less than the minimum length
    0 messages with bad checksum
    0 messages with bad source address
    0 messages with bad length
    0 echo drops with broadcast or multicast destinaton address
    0 timestamp drops with broadcast or multicast destination address
    Input Histogram 
      14 echo reply
      1520 destination unreachable
      1 routing redirect
      449787 echo
      449787 message responses generated
ICMP message type :

 ICMP messages are grouped into two classes:

  • ICMP error messages: Destination Unreachable, Redirect, Source Quench, Time Exceeded, and Parameter Problem.
  • ICMP query messages: Echo, Information, Timestamp, and Address Mask.

When the counter increases :

  • This counter is related to only the ICMP error messages; not to ICMP query messages.

  • This counter is increased when a packet is received with some condition, which would normally have generated an ICMP error message, but the received packet itself is an ICMP error message.

  • The system does not generate ICMP error messages that are caused by other ICMP error messages.

  • For example, when the system receives an ICMP Time Exceeded message and is unaware of the destination, it does not generate the ICMP Destination Unreachable message; instead it will increase the counter.
Test topology :

alt


Configuration :

  • In R1, the loopback address is configured as 1.1.1.1/32 and the static route for 200/8/100/8 is configured to the 100.0.0.2(R2) next-hop.

  • In R2, the static route for 200/8 is configured to the 100.1.0.3(R3) next-hop and no route to 1.1.1.1 or default route is configured.

  • In R3, the static route is configured as 100/8 and 1.1.1.1/32 is configured to the 100.1.0.1(R2) next-hop.


Verification :

  • A ping is executed from R1 to 200.0.0.1 with TTL 2 and 1.1.1.1 source IP. The counter increases in R2:

    • The ICMP request was sent to R3 via R2.
    • As TTL was 2, R3 generated the ICMP error message (time exceeded) to R1 via R2.
    • When the ICMP time expired message arrived at R2, it could not find the next-hop to 1.1.1.1.
    • Without this rule, R2 should have generated another ICMP error message (destination unreachable).
    • But, due to this rule, the ICMP Destination Unreachable message did not occur and the counter increased.
  • You can see the counter increasing in the output of  show system statistics icmp .
  • You also can see which ICMP error messages increases the counter by using the monitor traffic command in R2.