Skip to content

Terminology

Resources

In the monitoring system (referred to as MCT), various resources are designated as monitoring targets. These resources span different regions, have different monitoring types, and vary in size. To define the different levels of these monitoring targets, we utilize the following concepts, listed from small to large:

  • Check
  • Zone
  • Cluster
  • Datacenter

Check

A Check represents the smallest unit in MCT, representing a specific type of test based on a single endpoint. For instance, if you want to monitor HTTP ports 80 and 443, you would define two different checks in MCT.

Zone

A Zone is a grouping of checks that share the same testing plugin and testing interval. In some cases, checks within the same zone may also share certain parameters, reducing the duplication of effort required to set up each individual check.

Cluster

A Cluster refers to a group of interconnected computers or servers used for microservices that collaborate as a single system to perform tasks or provide services. In MCT, all checks within the same cluster are assigned to the same group of monitoring agents located in the same Datacenter. Depending on the size of the checks, you can use a single cluster or divide them into different clusters with appropriate size definitions.

Datacenter

In MCT, the concept of datacenter is mapped to monitoring agents located in the same region. For example, by default, all checks assigned to the SJC01_DATA datacenter will be executed by MCT agents located in the SJC region.

Service

A Service refers to a functionality provided to end customers. It may span multiple datacenters and environments (such as Fedramp or commercial). In MCT, we also offer a service-oriented view, allowing operations or DevOps teams to monitor their services across different datacenters.

Action

In MCT, operators have several actions they can take:

  • Add: Create new checks.
  • Edit: Modify existing checks.
  • Maintain: Put checks under maintenance to avoid alerts.
  • Recover: Move checks out of maintenance status.
  • Offline: Stop monitoring.

Maintain

When a failure or issue is detected, the tool captures and records the incident, generating an alert within the monitoring portal. However, the maintain feature in MCT ensures that immediate notifications to the responsible parties are not triggered. This feature creates a controlled environment where incidents can be monitored and managed without causing unnecessary interruptions or notifications.

Recovery

The recovery process in MCT focuses on returning the monitoring targets to a normal operational state after the maintenance phase. Once the necessary fixes or adjustments have been made and the targets are deemed stable and healthy again, the operations team can acknowledge the recovery. This acknowledgment marks the end of the maintenance period and allows the tool to resume its normal alerting and notification functions. It ensures that the responsible parties are promptly informed of any subsequent issues that may arise.

Automatic Test

In MCT, all monitoring processes are automated and executed at predefined intervals. The trigger for automatic execution is based on defined time intervals within a specific Zone. To enable automatic monitoring execution, simply right-click on the respective zone name and select the Set Timer option.

Manual Test

For quick verification purposes or when a one-time execution is desired, the Manual Test option can be used. By right-clicking on the check name, the monitoring process will be executed only once. This manual test provides a detailed output of the entire execution steps, regardless of whether they succeed or fail. Unlike automatic execution, where successful checks do not retain detailed execution records, the Manual Test captures and prints all the relevant information.