System Management Glossary

This topic describes architectural terms and definitions and other common terms and definitions that are associated with System Management and the Warm Standby model.

Architectural Terms


Term	Definition
Term	Definition
Solution Set	A solution set is the group of all hardware and software necessary to provide all the needed functionality. In a Warm Standby system, this would contain 2 Core Instances and 1 or more peripheral and management (a special type of peripheral) instances.
Core Instance	Installation of core monitor and associated VHT components, including Queue Manager. Typical components included in a core instance are Queue Manager, Provider, OpMode Server, Report Writer, Readerboard Adapter, and CTI Connector.
Peripheral Instance	Installation of peripheral monitor and associated VHT components. Typical components included in a peripheral instance are a peripheral monitor, Message Bus, etc.
Management Instance	A special type of peripheral instance that contains system management components. Typical components included in a management instance are peripheral monitor, Message Bus, etc.
Core Monitor	The core monitor's primary responsibilities include: Starting components, including Queue Manager, and restarting them if they stop. Switching which core instance is handling traffic during failure conditions (failover). Carrying out commands from the System Management user interface, like switching which core instance is active. Other details: Monitors the status of a sibling core instance to determine its health. Makes necessary configuration changes when switching which core Instance is active. If restarts exceed a configurable number within a configurable time period, core monitor will initiate an automated failover. Example: Failover when Queue Manager restarts three times in 120 seconds.
Peripheral Monitor	The peripheral monitor's primary responsibilities include: Starting components (QWatchClient, Message Bus, Outbound IVR, Outbound Contact Client, etc.) and restarting them if they stop. Monitoring status of the core Instances to determine which Queue Manager is active. Reconfiguring components to point to the active Queue Manager when it changes. If restarts exceed a configurable number within a configurable time period, the peripheral monitor will restart itself. Example: Restart when any component restarts eight times in 120 seconds.
Watchdog (heartbeat)	Each core or peripheral monitor has a watchdog which will restart that monitor if it stops.
Management API	The Management API receives status messages from all instances and generates e-mail notifications to inform administrators of changes. Multiple Management APIs can be clustered to share the same information and provide redundancy. Other details: This service replaces SENSe, starting in Release 8.0. E-mail can be sent to any list of e-mail addresses. Management API can be configured to use SMTP authentication or no authentication.
System Management User Interface	The System Management user interface shows the current health of the system and allows the user to perform certain corrective actions, such as switching which core Instance is active.
Datastore	The datastore synchronizes callback data and user data between instances. Queue Manager loads existing callbacks into its memory when it becomes active. This enables one Queue Manager to pick up where another one left off.
Failover	When a serious problem occurs, the system automatically switches from the currently active core Instance to the backup.
Make Primary	For maintenance or for any other reason, an administrator can use the System Management user interface to manually switch from the currently active core Instance to the backup.
Erlang	Erlang is a general-purpose concurrent, garbage-collected programming language and runtime system. Much of the Callback solution is written using Erlang.
Erlang node	An Erlang node is collection of applications existing as a network entity in the Erlang mesh.