Customer might report that their monitoring tool is reporting high utilization on logical-tunnel interface (lt-0/0/0)

Network monitoring tool reporting high utilization on lt-0/0/0 interface

The statistics on lt-0/0/0 interfaces are cumulative of all the sub interfaces of lt-0/0/0 (lt-0/0/0.0,lt-0/0/0.2,lt-0/0/0.3, lt-0/0/0.4, lt-0/0/0.5 etc). Customers will need to consider that the traffic passes 2 units of lt-0/0/0 that are in a pair (ingress/egress). Then the lt-0/0/0 sums the statistics of all the interface units and shows it as cumulative number.

Example: lt-0/0/0.14 lt-0/0/0.13

Configuration Snipplet:

interfaces {

      lt-0/0/0 {

        unit 13 {

          encapsulation ethernet;

          peer-unit 14;

 

 

 Logical interface lt-0/0/0.13 (Index 164) (SNMP ifIndex 773) (Generation 229)

  Flags: Up SNMP-Traps 0x4000 Encapsulation: ENET2

  MAC: 02:00:00:00:00:33

  Traffic statistics:

   Input bytes :    8789899431652

   Output bytes :   108694883361298

   Input packets:     40293784111

   Output packets:     97062695023

  Local statistics:

   Input bytes :        417060

   Output bytes :        456780

   Input packets:         9930

   Output packets:         9930

 

 

 Logical interface lt-0/0/0.14 (Index 165) (SNMP ifIndex 774) (Generation 230)

  Flags: Up SNMP-Traps Encapsulation: ENET2

  MAC: 02:00:00:00:00:34

  Traffic statistics:

   Input bytes :   108694883830384

   Output bytes :    8789899506952

   Input packets:     97062693411

   Output packets:     40293784319

  Local statistics:

   Input bytes :        417060

   Output bytes :        458926

   Input packets:         9930

   Output packets:         9957

One can easily identify from the above output that the ingress/egress traffic statistics of unit 13 are equal to egress/ingress of unit 14. If calculated the input/output bytes for all interfaces units summed up together will be equal to that to the statistics shown by lt-0/0/0(show interfaces lt-0/0/0 extensive no-forwarding)

If the above explanation doesn't stand true in a specific customer's situation then the normal troubleshooting related to high interface utilization needs to be followed but if the above explanation is inline with customer's issue the recommendation must be to increase the threshold on the monitoring tool to accommodate the actual traffic flowing through the lt-0/0/0 interface or change the way to monitor the interface usage or monitor the physical interfaces.