Sometimes, an increase in the 'Drops' counter is seen on an interface like the following:
admin@router> show interfaces ge-0/1/4 extensive
<snip>
Output errors:
Carrier transitions: 0, Errors: 0, Drops: 120091, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0
10 seconds later :
admin@router> show interfaces ge-0/1/4 extensive
Physical interface: ge-0/1/4, Enabled, Physical link is Up
<snip>
Output errors:
Carrier transitions: 0, Errors: 0, Drops: 149958, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0
Does this mean there is a hardware or software issue or it is this an expected behavior? In this article, one scenario will be explained for the subscriber-management environment.
The configuration of the device under test environment:
interfaces {
ge-2/0/0 {
hierarchical-scheduler maximum-hierarchy-levels 2;
vlan-tagging;
mtu 4484;
unit 400 {
encapsulation ppp-over-ether;
vlan-id 400;
pppoe-underlying-options {
max-sessions 32000;
service-name-table PPPoE-ST;
}
}
}
pp0 {
unit 1 {
ppp-options {
chap;
}
pppoe-options {
underlying-interface ge-2/0/0.400;
server;
}
family inet;
}
}
class-of-service {
traffic-control-profiles {
TCP-50M-IP-OUT {
scheduler-map SM-OUT;
shaping-rate 50m;
overhead-accounting frame-mode bytes -20;
guaranteed-rate 50m;
delay-buffer-rate 50m;
}
interfaces {
pp0 {
unit 1 {
output-traffic-control-profile TCP-50M-IP-OUT;
}
}
scheduler-maps {
INTERNET-OUT-1000K-TO-10G {
forwarding-class BE scheduler BE;
forwarding-class NC scheduler NC;
}
}
schedulers {
BE {
transmit-rate percent 95;
buffer-size percent 90;
priority low;
}
NC {
transmit-rate percent 5;
buffer-size percent 10;
priority low;
}
}
}
Physical interface output before the test:
admin@router> show interfaces ge-2/0/0 extensive | match "Output errors:|Carrier transitions:"
Output errors:
Carrier transitions: 1, Errors: 0, Drops: 0, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0
The box had one active subscriber:
admin@router> show subscribers summary port
Interface Count
ge-2/0/0 1
During the test, subscribers received 60 Mbps of traffic. After starting the traffic, the “Drops” counter started growing:
{MASTER}
admin@router show interfaces ge-2/0/0 extensive | match "Output errors:|Carrier transitions:"
Output errors:
Carrier transitions: 1, Errors: 0, Drops: 975223598, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0
{MASTER}
admin@router show interfaces ge-2/0/0 extensive | match "Output errors:|Carrier transitions:"
Output errors:
Carrier transitions: 1, Errors: 0, Drops: 976378447, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors:
Amount of dropped packets after stopping the test:
admin@router show interfaces queue pp0.1 | match "Tail-dropped packets" | except "0 0 pps"
Tail-dropped packets : 209479406 0 pps
Which is exactly equal to “Drop” counter of the physical interface:
admin@router show interfaces ge-2/0/0 extensive | match "Output errors:|Carrier transitions:"
Output errors:
Carrier transitions: 1, Errors: 0, Drops: 209479406, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0
In the configuration above where the committed information rate is equal to the peak information rate, the tail drops are expected, since the card doesn't allocate any memory for storing additional traffic.
2020-10-23: Article reviewed for accuracy, no changes required; article accurate and valid