Configuring a Redundant Pair of VMA SAN Gateways - Process Guide

For bonding of the two Ethernet ports, eth1 and eth2, on both gateways, the link aggregation mode

‘balance-rr’ is used in this example process since for most switches explicit port trunking for the ports

bonded using ‘balance-rr’ is not required and simpler to deploy. If you intend to use one of the other

possible gateway port link aggregation modes, then you will also need to configure trunking of the

ports connected to eth1 and eth2 of both VMA SAN Gateways accordingly. When using two

Ethernet switches as recommended, if trunking of the switch ports connecting to the gateways is

required, then split trunking or distributed trunking will need to be supported on the Ethernet switches

used.

Configure VLAN for Inter-Gateway Communication

The same bonded Ethernet ports on each VMA SAN Gateway will also be used for inter-gateway

communication and coordination. To provide for this, create a tagged VLAN which includes the ports

connected to the eth1 and eth2 Ethernet ports on each gateway. A tagged VLAN configuration is

used so that user gateway management requests and inter-gateway communication can flow over the

same physical gateway ports.

In this example VMA SAN Gateway redundant pairing process VLAN14 is used, but any available

VLAN-ID can be used.

At this time, only the Ethernet ports eth1 and eth2 of the ‘primary’ gateway should be connected to

the appropriate switches; do not connect any Ethernet ports of the ‘secondary’ gateway to the

Ethernet switches. This will be done later in the process.

Configure Primary Gateway to expect Secondary Gateway

Set the number of ‘cluster expected-nodes’ on the ‘primary’ gateway to ‘2’ so that later it will pair with

the ‘secondary’ gateway and synchronize the configuration database. To do this, issue the following

commands on the ‘primary’ gateway:

> enable

# config term

# cluster expected-nodes 2

# show cluster global

hp-gateway1 [hp-redundant1: master] (config) # cluster expected-nodes 2

*hp-gateway1 [hp-redundant1: master] (config) # show cluster global

Cluster ID: 00600-0008-6090

Cluster name: hp-redundant1

Management IP: 0.0.0.0/0

Cluster master IF: eth1

Cluster node count: 1

Node Status:

Node ID: 1 <--- (local node)

Host ID: e4111da840b0

Hostname: hp-gateway1

Node Role: master

Node State: online

Node internal address: 10.61.184.238, port: 60102

Node external address: 10.61.184.238

Recv. Heartbeats from: -1

Send Heartbeats to: -1