cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1190
Views
0
Helpful
1
Replies

HSRP/VRRP between NXOSv nodes on CML2 failing

ghenning.mil
Level 1
Level 1

When using CML 2.1, we are unable to get HSRP or VRRP to talk between two NXOSv nodes (default CML node definition for NX-OSv Titanium switch).  We have tried them with switched interfaces and routed interfaces, both directly connected and with an L2 switch between them.  In all cases, we are able to ping the base IP address of the opposite node, so we know we've got basic connectivity.  We have enabled the VRRP, HSRP and Interface-VLAN feature.  We've done the license grace-period, so we should have all features available.  Both the VRRP and HSRP configurations appear to load OK on the local node, we can ping the virtual IP from the local node, but it never syncs up with the remote node. When we put a L2 switch between the NXOS nodes, we can ping the "real" IP address of each NXOS node from the switch, but not the virtual IP for the HSRP/VRRP.  Looking at packet captures on the links, we don't see any VRRP/HSRP management traffic. The routers claim to be transmitting HSRP/VRRP packets, but show no received packets.

An inhouse engineer who had been running VIRL 1.6 said the same HSRP/VRRP configurations worked fine on that version, so we keep coming back to it being something within CML 2.x.

 

We've stripped the configs down to as minimal as we can:
interface Vlan400
no shutdown
ip address 192.168.0.3/24
hsrp 4
authentication text cisco
priority 150
ip 192.168.0.1
interface Ethernet2/1
switchport
switchport access vlan 400
no shutdown

 

Then this is what we get on the HSRP where it doesn't see the remote router:

Vlan400 - Group 4 (HSRP-V1) (IPv4)
Local state is Active, priority 150 (Cfged 150)
Forwarding threshold(for vPC), lower: 1 upper: 150
Hellotime 3 sec, holdtime 10 sec
Next hello sent in 0.835000 sec(s)
Virtual IP address is 192.168.0.1 (Cfged)
Active router is local
Standby router is unknown
Authentication text "cisco"
Virtual mac address is 0000.0c07.ac04 (Default MAC)
2 state changes, last state change 00:54:29
IP redundancy name is hsrp-Vlan400-4 (default)

 

The L2-switch positioned between the NXOS nodes does appear to be getting ARP replies for the VIP, but pings still fail:
L2-switch#show arp
Protocol Address Age (min) Hardware Addr Type Interface
Internet 192.168.0.1 0 0000.0c07.ac04 ARPA Vlan400
Internet 192.168.0.2 12 5254.0006.8dda ARPA Vlan400
Internet 192.168.0.3 12 5254.0007.a4cf ARPA Vlan400
Internet 192.168.0.5 - 5254.000e.8190 ARPA Vlan400

 

We're running out of ideas as to why it's failing.

1 Reply 1

Herlander Stock
Level 1
Level 1

I purchased the software yesterday, but I'm really disappointed. I saw you tube video of two renown CCIE instructor explaining and using it I thought it was a great product, but when started and deploy only 5 devices and two PC on my very strong Dell workstation with dual physical CPU a 2.6 and 128 GB of RAM. Gave CML the default 8 GB of RAM and 4 CPU, but was not enough. Very slow loading devices, Console one devices at a time, and if you move between devices there is not persistent state of your console, etc.