Why does a layer 3 hop use vrrp secondary on nexus 6001

Question

I have a nexus N5K-C5672UP as core layer 3 for our VLAN's. VLAN 601 on core 2 is secondary vrrp:

core 02:

core02# show vrrp
  Interface  VR IpVersion Pri   Time Pre State   VR IP addr
---------------------------------------------------------------
    Vlan601   1   IPV4     90    1 s  Y  Backup   172.17.1.1

    interface Vlan601
  no shutdown
  ip address 172.19.1.3/24
  vrrp 1
    priority 90
    address 172.19.1.1
    no shutdown

core01:

core01# show vrrp
  Interface  VR IpVersion Pri   Time Pre State   VR IP addr
---------------------------------------------------------------
    Vlan601   1   IPV4    110    1 s  Y  Master   172.19.1.1

 interface Vlan601
  no shutdown
  ip address 172.19.1.2/24
  vrrp 1
    priority 110
    address 172.19.1.1
   no shutdown

I have a an esxi host - with linux trunked to core02. I would expect he would not use the layer 3 interface of vlan 601 from core02 - but that he would use the layer 3 interface of core01. However a traceroute shows he uses the secondary vrrp interface IP as his first hop:

using vrrp backup IP as first hop:

user@host:~$ ssh host "traceroute anotherhost"
traceroute to anotherhost (172.19.1.xx), 30 hops max, 60 byte packets
 1  172.19.1.3 (172.19.1.3)  0.648 ms  0.808 ms  0.951 ms # vrrp backup IP 
2  172.19.15.2 (172.19.15.2)  1.082 ms  1.180 ms  1.105 ms

Am I wrong here - but shouldn't he be using layer 3 here and hitting the primary vrrp even though he is trunked to core02 - that is not layer 3 - the trunk should have nothing to do with him using the layer 3 on core02.

Please help - I dont want him using the secondary vrrp.

Answer 1

In the case of vPC configurations, both HSRP/VRRP peers will actively route packets when they receive them. It is a special behavior for vPC.

Some information on vPC active/active routing