Port channel loadbalacing issue on VSS

[y.mgouni]

Hello,

I have a C6807 in VSS mode, and I created a port channel with 3 interfaces connected to a router.

The links in module 1 take traffic but the link in module 2 take just the input and output no as you can see below:

sh int te 1/1/5 | i rate 

Queueing strategy: fifo 

30 second input rate 1607336000 bits/sec, 188996 packets/sec 

30 second output rate 23916000 bits/sec, 9539 packets/sec

sh int te 1/1/1 | i rate 

Queueing strategy: fifo 

30 second input rate 1652979000 bits/sec, 193997 packets/sec 

30 second output rate 25151000 bits/sec, 9900 packets/sec

sh int te2/1/1 | i rate 

Queueing strategy: fifo  30 second input rate 1555220000 bits/sec, 185880 packets/sec 

30 second output rate 10000 bits/sec, 11 packets/sec

if we shut the uplinks interfaces on switch 1 (Active switch) , the traffic will not go through sw2 uplinks : 

int te 1/1/1

VSS-CORE(config-if)#shut

VSS-CORE(config-if)#int te 1/1/5

VSS-CORE(config-if)#shut

VSS-CORE(config-if)#end

VSS-CORE#sh int te2/1/1 | i rate*

  30 second input rate 1954808000 bits/sec, 232900 packets/sec 

30 second output rate 8000 bits/sec, 9 packets/sec <<<<<<<

VSS-CORE#sh int te2/1/1 | i rate 

Queueing strategy: fifo 

30 second input rate 1954808000 bits/sec, 232900 packets/sec 

30 second output rate 8000 bits/sec, 9 packets/sec <<<<<<<<<<<<<

I suspect that it's the VSS issue, can you please assit.

Regards;

[Francesco Molino]

Hi

When you shut all links from chassis 1 in your PO, traffic is interrupted which means loss of traffic?

This PO is facing what kind of equipment? Another vss or else?

Can you elaborate please and share the config from both side + include the output of sh etherch summary from both side?

[y.mgouni]

Hello,

Yes when i shut all the links in Chassis 1 in PO traffic is interrupted.

The PO is facing an ASR 9K.

VSS side:

=========

sh run int PO3
Building configuration...

Current configuration : 147 bytes
!
interface Port-channel3
no switchport
ip address 192.168.130.9 255.255.255.252
load-interval 30
end

sh etherchannel summary
Flags: D - down P - bundled in port-channel
I - stand-alone s - suspended
H - Hot-standby (LACP only)
R - Layer3 S - Layer2
U - in use N - not in use, no aggregation
f - failed to allocate aggregator

M - not in use, no aggregation due to minimum links not met
m - not in use, port not aggregated due to minimum links not met
u - unsuitable for bundling
d - default port

w - waiting to be aggregated
Number of channel-groups in use: 8
Number of aggregators: 8

Group Port-channel Protocol Ports
------+-------------+-----------+-----------------------------------------------
1 Po1(RU) - Te1/3/1(P) Te1/3/2(P) Te1/4/1(P)
Te1/4/2(P)
2 Po2(RU) - Te2/3/1(P) Te2/3/2(P) Te2/4/1(P)
Te2/4/2(P)
3 Po3(RU) LACP Te1/1/1(P) Te1/1/5(P) Te2/1/1(P)
5 Po5(RU) - Te1/3/7(P) Te2/3/7(D)
6 Po6(RU) - Te1/3/8(D) Te2/3/8(P)
7 Po7(RU) - Te1/1/2(P) Te1/1/3(P)
8 Po8(RU) - Te2/1/2(P) Te2/1/3(P)
11 Po11(RU) - Te1/1/7(P) Te2/1/7(P)

ASR side:

==============

Bundle-Ether20
Status: Up
Local links <active/standby/configured>: 3 / 0 / 3
Local bandwidth <effective/available>: 30000000 (30000000) kbps
MAC address (source): b026.802b.e753 (Chassis pool)
Inter-chassis link: No
Minimum active links / bandwidth: 1 / 1 kbps
Maximum active links: 64
Wait while timer: 2000 ms
Load balancing:
Link order signaling: Not configured
Hash type: Default
Locality threshold: None
LACP: Operational
Flap suppression timer: Off
Cisco extensions: Disabled
Non-revertive: Disabled
mLACP: Not configured
IPv4 BFD: Not configured
IPv6 BFD: Not configured

Port Device State Port ID B/W, kbps
-------------------- --------------- ----------- -------------- ----------
Te0/4/0/2 Local Active 0x8000, 0x0003 10000000
Link is Active
Te0/4/0/3 Local Active 0x8000, 0x0002 10000000
Link is Active
Te0/4/1/1 Local Active 0x8000, 0x0001 10000000

sh run int Bundle-Ether20
Wed Mar 4 11:52:18.001 UTC
interface Bundle-Ether20
description LIEN CORE HALLES
ipv4 address 192.168.130.10 255.255.255.252
lacp mode active
load-interval 30
flow ipv4 monitor Netflow-Arbor sampler SAMPLER_MAP ingress
ipv4 flowspec disable
!

[Francesco Molino]

The link from vss chassis is facing asr interface Te0/4/1/1?
On the output, the other 2 interfaces on asr it says link active but not for this one.

Is it a copy paste typo?

[y.mgouni]

It's a copy past typo

Bundle-Ether20
Status: Up
Local links <active/standby/configured>: 3 / 0 / 3
Local bandwidth <effective/available>: 30000000 (30000000) kbps
MAC address (source): b026.802b.e753 (Chassis pool)
Inter-chassis link: No
Minimum active links / bandwidth: 1 / 1 kbps
Maximum active links: 64
Wait while timer: 2000 ms
Load balancing:
Link order signaling: Not configured
Hash type: Default
Locality threshold: None
LACP: Operational
Flap suppression timer: Off
Cisco extensions: Disabled
Non-revertive: Disabled
mLACP: Not configured
IPv4 BFD: Not configured
IPv6 BFD: Not configured

Port Device State Port ID B/W, kbps
-------------------- --------------- ----------- -------------- ----------
Te0/4/0/2 Local Active 0x8000, 0x0003 10000000
Link is Active
Te0/4/0/3 Local Active 0x8000, 0x0002 10000000
Link is Active
Te0/4/1/1 Local Active 0x8000, 0x0001 10000000
Link is Active

[philclemens1835]

What happens with the following troubleshooting steps:

On the VSS pair, move all LACP links to the same switch and see if the behavior changes.

Conversely,

On the ASR, move all LACP links to the same MPA and observe the behavior.

This should help you isolate the issue.

You could also use a third switch, if you have one, and set up a basic LACP port channel and connect it to each of these devices to see which device is causing the issue.

[Francesco Molino]

Are you able to create a 2nd PO with 2 links on both side (1 per chassis), and use 2 hosts to tests?
Based on outputs, everything looks fine, Even if the load-balancing method isn't set for example, you should be able to reach the other end host going through the PO using the link attached to 2nd chassis,