Vanguard 7300 Series, Installation Manual

Page: 132 / 227

4-22 Operating Your Vanguard 7300
Board Management
Board Management
Introduction
Board management is limited to board level actions, this includes resetting the
hardware, testing the hardware, downloading any necessary software to the board,
downloading any additional configuration information, activating and deactivating
the board. The board management function supports recovery of T1-E1 boards on
the 7300 platform.
The purpose of the recovery mechanism is to allow for a method of recovery of
faults which might occur in the T1/E1 card. This card is heavily populated by active
devices and supports a lot of channels. The recovery mechanism detects faults and
takes action to protect the node and its resource and allow other boards that are not
involved in the fault to continue operation.
Hot swap of the T1-E1 card is specifically excluded as a functional requirement in
this version of board recovery. The functional requirements exclude removing a
hardware faulty T1-E1 board and inserting a equivalent working board in its place
while the node is functioning and expecting the board to be activated.
Once recovery is triggered, the recovery function reloads the T1/E1 board and takes
whatever necessary steps needed to complete the process of bringing the failed card
on line.
The recovery preserves the topology of the operational (running) configuration.
Applications which use the failed resources are informed that the resource is
deactivated, but they are not necessarily deleted. These instances are available to
report the deactivate condition. Reactivation after recovery brings the system on line
as it was before the failure. The process does not do any booting actions such as
reading CMEM.
Configuration
Board management has configuration. In the CTP the configuration is controlled by
a menu entry in the Configuration Menu:
Configure->Board Management . The
same parameters are available for viewing in the Examine and List menus. The
Board Management configuration is specified in the following table:
Keepalive Timeout
Range 1 to 20
Default 5
Description This is the time between the sending of keepalive/heartbeat
messages. A shorter timeout allows a more responsive recovery to
occur, but with a higher processing overhead to the node. The
parameter is in tenth of seconds