Configuring Port Bandwidth Monitoring
NetVisor OS version 7.0.0 enables you to configure alerts in the form of SNMP traps and syslog messages when the data throughput across a port over an interval exceeds a configured bandwidth threshold. This feature helps in detecting network congestion and monitoring port bandwidth usage.
When you enable this feature, NetVisor OS logs separate syslog messages when the threshold is exceeded by ingress or egress traffic and also when the bandwidth usage decreases from a value above the threshold to a value below the threshold on a port. This creates four possible combinations of log messages for when the bandwidth usage exceeds or falls below the threshold by ingress or egress traffic. You can also enable SNMP traps for bandwidth monitoring through the CLI.
For more information on system log messages, refer NetVisor OS Log Messages Guide.
Note:
- NetVisor OS calculates the throughput for a port as the total bandwidth used by all types of network traffic. In other words, the bandwidth usage calculation is not traffic-specific.
- If you issue the port-stats-clear command during a port bandwidth monitoring interval, the throughput calculation for that interval may not be accurate.
- You cannot configure this feature on trunk ports. But if you configure bandwidth usage monitoring on a port and add it to a trunk, the port continues to be monitored.
- You cannot configure this feature on the CPU port (port 0).
To configure port bandwidth monitoring, use the command:
CLI (network-admin@switch) > port-bw-threshold-alert-modify
|
port-bw-threshold-alert-modify |
Modify port bandwidth monitoring. |
|
port port-list |
Specify the list of ports that you want to monitor. |
|
bw-threshold bw-threshold-string |
Specify the bandwidth threshold as a percentage of the total bandwidth of the ports. The default value is 0, which means that bandwidth monitoring is disabled by default. |
For example, to configure the bandwidth threshold as 25 percent for ports 24 to 26, use the command:
CLI (network-admin@switch) > port-bw-threshold-alert-modify port 24-26 bw-threshold 25
Use the port-bw-threshold-alert-show command to display the configuration:
CLI (network-admin@switch) > port-bw-threshold-alert-show
switch port bw-threshold
------ ---- ------------
switch 24 25%
switch 25 25%
switch 26 25%
To configure a port bandwidth monitoring interval, use the command:
CLI (network-admin@switch) > port-bw-threshold-mon-int-modify
|
port-bw-threshold-mon-int-modify |
Modify port bandwidth monitoring interval. |
|
monitor-interval 10..300 |
Specify a port bandwidth monitoring interval. The allowed range of values is between 10 and 300 seconds. The default value is 30 seconds. |
For example, to set a port bandwidth monitoring interval of 45 seconds, issue the command:
CLI (network-admin@switch) > port-bw-threshold-mon-int-modify monitor-interval 45
Use the port-bw-threshold-mon-int-show command to display the monitoring interval configuration.
CLI (network-admin@switch) > port-bw-threshold-mon-int-show
monitor-interval: 45
You can view the syslog message for events related to port bandwidth monitoring by issuing the log-system-show command:
CLI (network-admin@switch) > log-system-show layout vertical count 2
category: system
time: 2021-09-27,06:57:54.743025-07:00
name: tx_port_bw_th_exceed
code: 11551
level: note
port: 9
message: Egress port bandwidth threshold exceeded at port=9
category: system
time: 2021-09-27,06:59:34.747463-07:00
name: tx_port_bw_th_not_exceed
code: 11552
level: note
port: 9
message: Egress port bandwidth threshold no longer exceeding at port=9
To receive SNMP alerts when the bandwidth usage exceeds the configured threshold, you must configure an SNMP trap for port bandwidth monitoring. Use the snmp-trap-enable-modify command to enable or disable this SNMP trap.
To enable the SNMP trap for port bandwidth monitoring, use the command:
CLI (network-admin@switch) > snmp-trap-enable-modify port-bw-threshold-exceed-event
To disable the SNMP trap for port bandwidth monitoring, use the command:
CLI (network-admin@switch) > snmp-trap-enable-modify no-port-bw-threshold-exceed-event
The SNMP trap message for the event where port bandwidth usage exceeds the threshold is:
DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (317194) 0:52:51.94
SNMPv2-MIB::snmpTrapOID.0 = OID: SNMPv2-SMI::enterprises.47269.4.7.3.1
SNMPv2-SMI::enterprises.47269.4.7.1.1.2.2 = STRING: "egressPortBwThresholdExceeded"
SNMPv2-SMI::enterprises.47269.4.7.1.1.3.2 = STRING: "/nvOS/log/system.log"
SNMPv2-SMI::enterprises.47269.4.7.1.1.4.2 = Gauge32: 5
SNMPv2-SMI::enterprises.47269.4.7.1.1.5.2 = STRING: "2021-09-27,06:57:54.743025-07:00 aries-ext-01 nvOSd(3775) system tx_port_bw_th_exceed(11551) : level=note : port=9 : Egress port bandwidth threshold exceeded at port=9
You can also view the event where the port bandwidth usage exceeds the configured threshold by using the port-show command. For example, in the case where the bandwidth threshold is exceeded by the ingress traffic on port 23, the output of the port-show command is:
CLI (network-admin@switch) > port-show port 23
switch port status config
------ ---- ------------------------------------------------------- ------
switch 23 defer-bringup-wait,rx-bw-threshold-exceeded fd,10g
You can also view the counters for the port bandwidth exceed event by issuing the port-stats-show command:
CLI (network-admin@switch) > port-stats-show format port,Rx-bw-th-exceed-count,Tx-bw-th-exceed-count,ibits,obits, port 41-42 show-interval 1
port Rx-bw-th-exceed-count Tx-bw-th-exceed-count ibits obits
---- --------------------- --------------------- ----- -----
41 0 0 131K 754K
42 0 0 164M 749K
port Rx-bw-th-exceed-count Tx-bw-th-exceed-count ibits obits
---- --------------------- --------------------- ----- -----
41 0 0 131K 754K
42 0 0 171M 749K
port Rx-bw-th-exceed-count Tx-bw-th-exceed-count ibits obits
---- --------------------- --------------------- ----- -----
41 0 0 132K 754K
42 1 0 178M 749K