Administrator Guide

Displaying Traffic Statistics on HiGig Ports
You can verify the buffer usage and queue counters for high-Gigabit Ethernet (HiGig) ports and link bundles (port channels). The buffer
counters supported for front-end ports are extended to HiGig backplane ports.
You can display the queue statistics and buffer counters for backplane line-card (leaf) and switch fabric module (SFM - spine) NPU port
queues on a switch using the show commands described in this section. Transmit, receive, and drop counters are displayed. Buffer
counters include the total number of cells currently used by all queues on all ports in a port pipe.
The f10-bp-stats.mib is used for gathering statistics about backplane HiGig ports. Line-card NPUs value is 0; SFM NPUs range from 0 to 1.
In an NPU unit, port numbering of HiGig ports starts from the last front-end I/O port number used.
Use the show hardware sfm hg-stats and show hardware linecard hg-stats commands to display traffic statistics
about the HiGig links on a line-card or SFM NPU.
Use the clear hardware sfm hg-stats and clear hardware linecard hg-stats commands to reset HiGig port
statistics.
Link Bundle Monitoring
Monitoring linked LAG bundles allows traffic distribution amounts in a link to be monitored for unfair distribution at any given time. A
threshold of 60% is defined as an acceptable amount of traffic on a member link.
Links are monitored in 15-second intervals for three consecutive instances. A syslog and an alarm will be activated when the average of
the bundle utilization is greater than the trigger threshold, and the delta deviation between the lowest and highest utilized links that part of
the ECMP or the Port-Channel is > 10% When the deviation clears, another Syslog sends and a clear alarm event generates.
The link bundle utilization is calculated as the total bandwidth of all links divided by the total bytes-per-second of all links. If you enable
monitoring, the utilization calculation is performed when the utilization of the link-bundle (not a link within a bundle) exceeds 60%.
To enable and view link bundle monitoring, use the following commands.
• Enable link bundle monitoring.
ecmp-group
• View all LAG link bundles being monitored.
show running-config ecmp-group
Link bundle monitoring can be also enable on port-channels, here it is the way it can be configured:
interface Port-channel 111
no ip address
switchport
no shutdown
link-bundle-monitor enable
To view the links that are being monitored, use the show link-bundle-distribution command.
Dell(conf-if-po-111)#do show link-bundle-distribution
Link-bundle trigger threshold - 22
LAG bundle - 111 Utilization[In Percent] - 25 Alarm State - Active
Interface Line Protocol Utilization[In Percent]
PeGi 255/1/36 Up 25
PeGi 255/1/37 Up 25
PeGi 255/2/38 Up 25
PeGi 255/2/39 Up 50
PeGi 255/3/40 Up 0
PeGi 255/3/45 Up 25
Interfaces
417