[c-nsp] ASR9K HSRP Configuration

[Chris Jones]

Hey Mohammad,

You seem to be trying to combine two different failover technologies.

ICCP/mLACP would be used to create a port channel that spans both ASR9ks.  In this scenario one would be active and one passive, and you'd configure the same interface details on both.  This setup works especially well in VPLS/PW-type scenarios, as it has the effect of holding the "backup" path down thus helping prevent loops, but it works for Layer 3 as well.  You don't need HSRP here (in fact, you can't use it between the ASR9K in this mode, as the port on the "standby" 9K is always down)

HSRP would be used where you wanted two interfaces, both online, to also share a 3rd "VIP".  In this case as Gert says - you don't need ICCP, or mLACP.  Just create two separate LAGs (one on each 9k) then run HSRP on the subinterfaces.   You can still use vPC on the N5K side if you need to - but that's completely independent of anything you do on the 9ks

Chris
________________________________
From: cisco-nsp <cisco-nsp-bounces at puck.nether.net> on behalf of Mohammad Khalil <eng_mssk at hotmail.com>
Sent: Saturday, 29 August 2020 1:05 AM
To: cisco-nsp at puck.nether.net <cisco-nsp at puck.nether.net>
Subject: [c-nsp] ASR9K HSRP Configuration

Greetings all
I am trying to bring up HSRP between ASR9Ks who are connected through bundle ether to N5K

RP/0/RSP0/CPU0:New9K1A#  sh run int Gi0/0/1/14
Fri Aug 28 01:38:25.742 CST
interface GigabitEthernet0/0/1/14
 description 5k1 - Eth1/1
 bundle id 405 mode active
 cdp
 speed 1000
 carrier-delay up 0 down 0
 transceiver permit pid all
!

RP/0/RSP0/CPU0:New9K1A#  sh run int Gi0/0/1/15
Fri Aug 28 01:38:27.157 CST
interface GigabitEthernet0/0/1/15
 description 5k2 - Eth1/1
 bundle id 405 mode active
 cdp
 speed 1000
 carrier-delay up 0 down 0
 transceiver permit pid all

I have established bundle ether 405 in between:

interface Bundle-Ether405
 description Bundle to NEW-NEXUS5k-1 & 2 - Eth1/1-2 (To use 10G later)
 lacp switchover suppress-flaps 300
 mlacp iccp-group 1
 mlacp switchover recovery-delay 60
 mac-address 5.5.5
 bundle wait-while 100
 bundle maximum-active links 4 hot-standby
 load-interval 30

RP/0/RSP0/CPU0:New9K1A#sh bundle bundle-ether 405
Fri Aug 28 01:39:16.413 CST

Bundle-Ether405
  Status:                                    Up

ICCP is up and running:

RP/0/RSP0/CPU0:New9K1A#sh iccp group 1
Fri Aug 28 01:39:45.368 CST
Redundancy Group 1
  member ip:10.255.255.250 (New9K1B), up (connected)
    monitor: route-watch (up)
  backbone interface Te0/0/2/0: up
  backbone interface Te0/0/2/1: up
  backbone interface Te0/0/2/2.15: down
  enabled applications: mLACP
  isolation recovery delay timer: 30 s, not running

I have established sub interface on the bundle ether:

interface Bundle-Ether405.3106
 vrf sme
 ipv4 address 10.95.10.125 255.255.255.224
 encapsulation dot1q 3106

On the 2nd ASR9K:
interface Bundle-Ether405.3106
 vrf sme
 ipv4 address 10.95.10.126 255.255.255.224
 encapsulation dot1q 3106

HSRP is not coming up:

BE405.3106     3359 110 P Active  local           unknown        10.95.10.97

BE405.3106     3359 100   Init    unknown         unknown        10.95.10.97

Am I missing something?

Thanks
_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/