HP Auto Port Aggregation (APA) Release Notes for HP-UX 11i v1 and v2 May 2005

1 Release Notes for HP-UX 11i v1
This document provides information about the HP-UX 11i v1 (B.11.11.20) May 2005 release of HP Auto Port Aggregation
(APA) and LAN Monitor software.
Announcement
This release of HP APA now supports Proactive Failover Support (PFO) for LAN Monitor failover groups, TSO (TCP
Segmentation Offload), and 64-bit statistics.
Auto Port Aggregation
HP APA is a software product that creates link aggregates, often called "trunks," which provide a logical grouping of
two or more physical ports into a single "Fat-Pipe." Network traffic is load balanced across all of the links in the
aggregation, which allows you to build large bandwidth logical links into the server that are highly available and
completely transparent to the client and server applications.
Two additional features are automatic link failure detection and recovery, as well as optional support for load balancing
of network traffic across all of the links in the aggregation. This allows you to build large bandwidth "logical" links into
the server that are highly available and completely transparent to the client and server applications.
HP APA supports Serviceguard with certain conditions. See the HP Auto Port Aggregation (APA) Support Guide for more
information.
LAN Monitor
The LAN Monitor mode of HP APA provides a failover capability configuration tools similar to Serviceguard; however,
LAN Monitor does not support HP Serviceguard. In the event of link failure, LAN Monitor will automatically migrate the
data flow from the primary link to one of the standby links in the failover group.
What's in This Version
The B.11.11.20 version of HP APA:
• Supports the TSO (TCP Segmentation Offload), aka Large Send, feature if an aggregate is created with Ethernet
cards that all support TSO or if the active port in a LAN Monitor failover group supports TSO.
• Supports 64-bit MIB (RFC 2863) statistics if all the interfaces within a link aggregation or failover group support
64-bit statistics.
• Supports proactive failover for LAN Monitor failover groups. Proactive failover is a cost-based failover policy in
which each port in a failover group is assigned a cost. With this policy, the active port of a failover group is always
the link that is the most efficient at carrying traffic. Efficiency is measured in terms of normalized port cost, which
is the assigned port cost divided by the port's current link speed. The lower the normalized port cost, the higher
the efficiency.
• Supports the logging of status messages to the syslog.log file.
• Enables you to add and remove DOWN ports to and from a link aggregation, and to create and delete a failover
group that has a DOWN port.
• Enables you to choose the MAC address of the first LAN port or first LACP-enabled port as the LACP System ID.
See the HP Auto Port Aggregation (APA) Support Guide (HP Part Number J4240-90031) for more information on the
new features and how to use them.
Known Problems and Workarounds
This section describes the known problems and workarounds for this release.
Known Problems with Switches
• HP ProCurve Switches
When disabling Cisco's Fast Ether Channel protocol on the HP 9000 server, the HP ProCurve switches block that
particular port from further use.
Corrective Action: In order to use the network physical port on the HP 9000 server, physically move the connection
from the server to the switch to another non-Cisco Fast EtherChannel port on the switch.
• Cisco Catalyst 5000 Switches
• Turning off Cisco's Fast Ether Channel protocol on on one network physical port on the HP 9000 server might
cause the entire link aggregate (the port it is associated with) to be deconfigured.
Corrective Action: Reconfigure the switch; do not include the port in the link aggregate being configured.
See the appropriate switch documentation for information on how to reconfigure link aggregates.
Announcement 4