AppDynamics proactively monitors the performance of your application and provides alerts regarding performance concerns prior to the occurrence of an actual outage.
AppDynamics defines a set of Health Rules where each rule contains certain conditions that maps to main performance indicators to effectively measure the health status of applications.
AppDynamics has enabled a few default Health Rules. If the “CPU utilization is too high”, for instance, the Health Rule triggers an event named “Health Rule Violation Started – Warning”, the CPU utilization should be on a node be greater than 75% there would be a “Health Rule Violation Started – Critical”.
Similarly AppDynamics has various template types for your own Health Rules. Each template type provides a few entities, like the Node, Tier, Business Transaction, or JMX Object, and “health rule” metrics that are related to those particular entities.
Pick the entity that the Health Rule is applicable for, after choosing the Health Rule template type . This is necessary as a Health Rule requires an object to evaluate against.
On selecting the entities to evaluate against, it is essential to build the conditions that reflect a warning status of your application. Every warning /critical status can have more than one defined conditions, wherein every condition compares an important performance metric with a static value or dynamic baseline.
For instance, a warning condition can be built to trigger when the Calls Per Minute is lower than 2 standard division of the default baseline and the Average Response Time of the node is greater than 10 ms.
An event can be triggered off according to your defined condition, once you have a Health Rule defined. Look for the event after an outage to know of its occurrence. AppDynamics alerts you of an event occurrence, when you specify Policies and associated Actions relating to your Health Rule event.
Appdynamics enables policy monitoring of the events logged. Once an event that matches, the policy’s triggering condition shows up, after which the policy triggers a set of Actions associated to it. Typically the policy triggering condition includes the matching events and matching entities so that an event is of a specific type. For instance the Health Rule Violation Started – Warning, and takes place on a specific entity, for instance Node 2.
On defining the triggering conditions for the Policy, specify the actions that AppDynamics needs to perform when the policy fires. Appdynamics offers multiple types of actions that range from simple alerting actions (via SMS or email), to diagnostics actions, to advanced custom actions that integrate with existing alerting systems (JIRA or PagerDuty).
With all Health Policies, Rules, and Actions in place, AppDynamics alerts you when an application begins to experience performance issues. Proactive alerting depends on defining a group of Health Rules and triggering Policies that sets off an alert whenever your application enters a state of concern.
Appdynamics helps you understand the performance of your production applications. Get AppDynamics today.