Dell Vizioncore Guía de inicio rápido descargar pdf (Pagina 33)

Introducing vFoglight 33

How vFoglight Works

When you define a service, a corresponding service level is automatically created. You

can also use Alarm Filters to decide what alarms are relevant to a given service and can

impact the availability of that service. See Services and availability below for more

information.

Using the Dynamic Managed Components feature, you can define a set of components

that will be automatically included in a service when certain data arrives.

The monitoring dashboards in Foglight are organized around the theme of services.

See the Foglight User Guide for more information about working with services.

Service Levels, Alarms, and State

In vFoglight, service levels are availability measurements on services. Default service

levels measure availability based on the state of individual nodes in a model instance.

A vFoglight rule evaluates a series of conditions designed to create an alarm if a

performance problem—an exception condition—has occurred. The alarm has one of

three severity levels: Warning, Critical, or Fatal. Rules are associated with nodes in the

model. When one of the conditions of a rule evaluates to true, an alarm is attached to the

node in the model instance with which the rule is associated.

A state is set for each node in a model instance. If there have been no exception

conditions, the node’s state is set to the default state Normal. However, if an alarm is

attached to a node, then the state of the node is set to the severity level of the alarm

(Warning, Critical, or Fatal). In general, the state of a node reflects the most severe

alarm attached to that node.

In addition to having an individual state, each node also has an aggregate state. By

default, the aggregate state of a node represents the state of itself and all its children.

Though there are some exceptions, in general, vFoglight calculates a node’s aggregate

state by evaluating the state of each of its child nodes and setting the aggregate state to

be equal to the most severe child node state. In this way, state is propagated up the

model instance.

Services and availability

The availability of a service is, by default, established based on a rollup of all the alarms

fired for the monitored components added to that service.

However, there can be cases in which you do not want this type of alarm rollup to be

performed. Often, only a subset of the alarms fired for a monitored component are

relevant to a given service.

1 2 ... 28 29 30 31 32 33 34 35 36 37 38 ... 63 64

Sin comentarios

Dell Vizioncore Guía de inicio rápido Pagina 33