HP NC320m - PCI Express Gigabit Server Adapter User Manual page 26

Active Path Failover
Active Path Failover allows a ProLiant server to predict and bypass failed network paths through use of a
user assigned echo node. An echo node is a device on the network with which connectivity is required.
Periodically, each team member transmits an Address Resolution Protocol (ARP) request packet to the echo
node. If a response is not received from the echo node within a set timeout period, the team member is
marked as failed.
If Active Path Failover is enabled, you must enter a valid IP address for the echo node. In addition, you
can specify the MAC address of the echo node, which allows the echo node packets to be unicast to the
echo node. If no Echo Node MAC address is provided, the echo node packets are broadcast.
Two types of echo node request probes are available:
Community Address ARP-(Recommended) Community Address ARP is a newer method that uses a
standard ARP to test connectivity. It works for all devices designated as the echo node including
routers, Hot Standby Routing Protocol (HSRP) devices, and Linux servers. If Community Address ARP
is selected, you must enter the IP address to be used for the community probe. You can also
configure the MAC address to be used, or use the default MAC address.
Directed ARP-Directed ARP tests connectivity with the echo node device using a modified ARP frame.
This method works for most switches or servers that are designated as the echo node. It is not
recommended for routers or Linux servers that are designated as the echo node.
Fast Path Failover
Fast Path Failover allows a ProLiant server to determine the fastest path to the core network (or the root
bridge) in addition to detecting full connectivity loss. Fast Path Failover examines the path cost information
contained in the Spanning Tree Bridge Protocol Data Unit (BPDU) frames and determines which switch is
the best one to use for the team's primary port.
When configuring Fast Path Failover, you must select which Spanning Tree Protocol (STP) type to use. The
two predominate STP types are as follows:
IEEE 802.1D—Because IEEE 802.1D is VLAN unaware, it is not necessary to configure Fast Path
Failover to listen to VLAN.
Cisco PVST+—Allows you to select the VLAN ID to be used for Fast Path Failover monitoring. (Refer
to "Configuring VLAN IDs.")
Router Path Failover
Router Redundancy Protocols (HSRP, VRRP) allow you to set up two or more routers into a single router
group, which acts as a backup for a virtual router. At any given time, only one router is designated as the
active router, which takes up the functionality of the router. If the active router fails, a second router or
standby router assumes the role of the active router. If the standby router fails or becomes the active
router, then another router is elected as the standby router.
The Router Path Failover ensures the team's connectivity to the active router. The router protocol, as
selected by the user, defines the method used to select the active router by periodically exchanging
multicast frames amongst routers in a router group. The NCU teaming software monitors these multicast
frames to validate an active path to the active router in the router group.
When configuring Router Path Failover, you must select which router protocol to use. The two protocols
available are as follows:
HSRP—Cisco Hot Standby Router Protocol
VRRP—Virtual Router Redundancy Protocol
Using advanced networking features 26