09-03-2015 03:10 AM - edited 03-08-2019 01:37 AM
Hi, I have 2x 10Gbps link (setup as a port-channel 20Gbps) When the links hits about 50percent traffic - around 10Gbps i see discards on the interface. Looks like its being dropped by Nexus 7k default policy. What would be the recommended way to get this fixed?
Config
interface port-channel123
switchport
switchport access vlan 123
mtu 9216
interface Ethernet7/1
switchport
switchport access vlan 123
mtu 9216
channel-group 123 mode active
no shutdown
interface Ethernet1/1
switchport
switchport access vlan 123
mtu 9216
channel-group 123 mode active
no shutdown
Nexus7k# show int po123
port-channel123 is up
admin state is up
Hardware: Port-Channel, address: e02f.6da0.2abd (bia e02f.6da0.2abd)
MTU 9216 bytes, BW 20000000 Kbit, DLY 10 usec
reliability 255/255, txload 3/255, rxload 1/255
Encapsulation ARPA, medium is broadcast
Port mode is access
full-duplex, 10 Gb/s
Input flow-control is off, output flow-control is off
Auto-mdix is turned off
Switchport monitor is off
EtherType is 0x8100
Members in this channel: Eth1/1, Eth7/1
Last clearing of "show interface" counters never
2 interface resets
Load-Interval #1: 30 seconds
30 seconds input rate 59298680 bits/sec, 8040 packets/sec
30 seconds output rate 301193920 bits/sec, 19097 packets/sec
input rate 59.30 Mbps, 8.04 Kpps; output rate 301.19 Mbps, 19.10 Kpps
Load-Interval #2: 5 minute (300 seconds)
300 seconds input rate 341378536 bits/sec, 32048 packets/sec
300 seconds output rate 552364904 bits/sec, 41829 packets/sec
input rate 341.38 Mbps, 32.05 Kpps; output rate 552.36 Mbps, 41.83 Kpps
RX
11996078021 unicast packets 51496 multicast packets 70349 broadcast packets
11996173986 input packets 19137490181632 bytes
367490606 jumbo packets 0 storm suppression packets
0 runts 0 giants 0 CRC/FCS 0 no buffer
0 input error 0 short frame 0 overrun 0 underrun 0 ignored
0 watchdog 0 bad etype drop 0 bad proto drop 0 if down drop
0 input with dribble 1531155 input discard
0 Rx pause
TX
9399605287 unicast packets 2172924 multicast packets 6 broadcast packets
9401752337 output packets 14036625859282 bytes
216013568 jumbo packets
0 output error 0 collision 0 deferred 0 late collision
0 lost carrier 0 no carrier 0 babble 0 output discard
0 Tx pause
Nexus7k# show policy-map interface po123 INput type queuing
Global statistics status : enabled
port-channel123
Service-policy (queuing) input: default-4q-8e-in-policy
SNMP Policy Index: 301991781
Class-map (queuing): 2q4t-8e-in-q1 (match-any)
queue-limit percent 10
bandwidth percent 50
queue dropped pkts : 0
queue dropped bytes : 0
queue transmit pkts: 302723181 queue transmit bytes: 48848337084
Class-map (queuing): 2q4t-8e-in-q-default (match-any)
queue-limit percent 90
bandwidth percent 50
queue dropped pkts : 1531155
queue dropped bytes : 2710366321
queue transmit pkts: 362980779040 queue transmit bytes: 289420756491895
Nexus7k# show class-map type queuing 2q4t-8e-in-q-default
Type queuing class-maps
========================
class-map type queuing match-any 2q4t-8e-in-q-default
Description: Classifier for Ingress queue 2 of type 4q2t8e
match cos 0-4
match dscp 0-39
09-03-2015 05:44 AM
Hi Sam,
If this is a regular port channel and you have a single stream which has 10G+ traffic, the drops are expected. In a port-channel, the traffic is load balanced per flow based on the load balance algorithm.
You can check which port the destination packet takes by using this command - http://www.cisco.com/web/techdoc/dc/reference/cli/nxos/commands/l2/show_port-channel_load-balance.html
If you are using multiple flows and still facing the same issue, try changing your load balance algorithm and things should work normal.
PS: This analysis is purely done because you are hitting the issue exactly at 50% with two links. Hence I do not suspect a QoS issue here.
Regards,
Manoj
09-03-2015 06:39 AM
Hi Manoj,
Thanks for your reply. I had a look and we are using below load balancing
Module 1:
Non-IP: src-dst mac
IP: src-dst ip rotate 0
Module 7:
Non-IP: src-dst mac
IP: src-dst ip rotate 0
I believe the traffic was one flow between 2 devices. Do you think its worth creating a new policy map and change 'bandwidth percent 50' to somewhere around 80 percent and apply to port-channel?
Thanks
Sam
09-03-2015 06:53 AM
Hi Sam,
If it is a single flow - you will not able to cross 10G. One flow can only go via one link and not both in a PC.
You can check the specific interface which the port channel is using by giving the src-dest addresses in the same command. Refer to the examples on that link.
Also the command "sh port-channel traffic interface port-channel <pc-no>" will show you some interesting output.
Regards,
Manoj
09-04-2015 02:48 AM
Thanks Manoj. It actually looks like its not a single flow.
I see around 9-10Gbps of traffic on Port-channel, on both member ports I see traffic divided 55/45 percent
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide