Monitoring the aggregate egress queue statistics per port provides in-profile, out-of-profile, and total statistics for both forwarded and dropped packets and octets on a specific port.
When enabled, all queues on the port are monitored, including SAP egress, network egress, subscriber egress, and egress queue group queues, as well as system queues which can be used, for example, to send port-related protocol packets (LACP, EFM, and so on).
This is enabled and disabled using the following command:
config port <port-id> [no] monitor-agg-egress-queue-stats
When enabled, the line card polls the related queues to derive the aggregates which provide the delta of the queue statistics since turning on the monitoring. This means that the reported statistics are not reduced by those from a deleted queue and so the aggregates correctly represent the forwarded/dropped statistics since the start of monitoring.
The aggregates can be shown with the following command:
show port [<port-id>] [statistics [egress-aggregate]] [detail]
As an example, the output below enables monitoring of aggregate egress queue statistics on port 2/1/1 and then shows the monitored statistics:
*A:PE# configure port 2/1/1 monitor-agg-egress-queue-stats
*A:PE# show port 2/1/1 statistics egress-aggregate
===============================================================================
Port 2/1/1 Egress Aggregate Statistics on Slot 2
===============================================================================
Forwarded Dropped Total
-------------------------------------------------------------------------------
PacketsIn 144 0 144
PacketsOut 0 0 0
OctetsIn 12353 0 12353
OctetsOut 0 0 0
===============================================================================
*A:PE#
To clear the aggregate statistics, the monitoring must be disabled and then re-enabled. The aggregate statistics are also cleared when the card is cleared (using a clear card slot-number command) or power-cycled (with the tools perform card slot-id command). Additionally, aggregate statistics related to MDA are cleared when the MDA is cleared (using the clear mda mda-id command) or the MDA is inserted into an IOM. The aggregate statistics are not cleared when a shutdown/no shutdown is performed on the card and, or MDA.
There is no specific limit on the number of queues that can be monitored, but the amount of each line card's CPU resources allocated to the monitoring is bounded; consequently, when more queues on a card’s ports are monitored, the aggregate statistics are updated the less frequently.
Monitoring of aggregate statistics is supported on PXC sub-ports but not on a PXC physical port. It is also not supported on satellite ports.