Motorola 6806800C08B 사용자 설명서
Availability Service
NetPlane Core Services
NetPlane Core Services Overview User’s Guide (6806800C08B)
27
validation data for managed FRUs and the linkages between entities and AMF logical nodes.
All the processor environment entities in the hardware entity containment tree, which
correspond to AMF nodes, contain the node name of the associated node. The node name
provides the linkage between the hardware and the software system models.
All the processor environment entities in the hardware entity containment tree, which
correspond to AMF nodes, contain the node name of the associated node. The node name
provides the linkage between the hardware and the software system models.
Further functionality provided by AvSv includes:
z
Automatic and administrative means to instantiate, terminate and restart resources
z
Automatic and administrative means to manage or reflect Service Group, Service Unit,
Service Instance and Resource state
Service Instance and Resource state
z
Administrative means to perform switch-over
z
Administrative means to reset (but not power cycle) nodes
z
Heartbeat and event subscription schemes for fault detection, isolation and identification
z
Health-check services to probe and prevent system trauma that lead to faults
z
Fault recovery mechanisms to fail-over SIs which maintain service availability in case of
system trauma
system trauma
z
Fault repair mechanisms to restore failed components
z
Validation of hardware resources (managed FRUs) entering the system
The AvSv itself cannot be a single-point of failure. It provides its own internal scheme and
mechanisms to protect itself from its own failure.
mechanisms to protect itself from its own failure.
2.5.1.1
Architecture
The AvSv service is comprised of the following main software components:
z
Availability Director
z
Availability Manager
z
Availability Node Director
z
Availability Agent
2.5.1.1.1
Availability Director
The Availability Director maintains the most abstract portions of the Software System Model,
such as cluster membership of nodes, service groups, service instances and service nodes.
such as cluster membership of nodes, service groups, service instances and service nodes.
Its main tasks include fault detection, isolation and recovery procedures as defined in the SAF
AMF. Any problems and failures on a component that cannot be handled locally, are promoted
to the Availability Director which controls and triggers the isolation of the affected component
and, if possible, the activation of a stand-by component.
AMF. Any problems and failures on a component that cannot be handled locally, are promoted
to the Availability Director which controls and triggers the isolation of the affected component
and, if possible, the activation of a stand-by component.
2.5.1.1.2
Availability Manager
The Availability Manager maintains the hardware model of the system described above. It acts
as link between the underlying HPI and the AMF system modelling.
as link between the underlying HPI and the AMF system modelling.