Skip to main content

Observability in ArcGIS Enterprise

Observability is a pillar of a well-architected ArcGIS system. By actively observing your ArcGIS Enterprise organization, you can stabilize system uptime, identify service performance issues or outages, and proactively adjust allocated resources across participating machines to run the underlying applications.

Observability solutions can provide active checks for commonly used endpoints and alert the appropriate contacts when responses fall outside of expected tolerances. In addition, you can use them to collect historical information that can be used to corroborate with system and software logs during root cause analysis or postmortem investigations.

Observability methods

ArcGIS Enterprise on Kubernetes supports a variety of methods to promote observability:

  • The Enterprise Manager Overview page provides an active view of your organization as a whole.

  • The ArcGIS Enterprise metrics API captures information about ArcGIS Enterprise services, users, and content items.

  • System logs are timestamped records of events within ArcGIS Enterprise.

  • Audit logs record user activity and any changes that occur to the system for integration with Security Information and Event Management (SIEM) tools.

  • Error reports provide information to help troubleshoot services that have crashed unexpectedly.

  • Health check reports validate overall organization health and provide insights into operational efficiency.

  • Portal usage reports provide usage statistics for items, users, and groups.

Cluster observability

To enforce appropriate security boundaries, ArcGIS Enterprise on Kubernetes does not have access to cluster-level privileges. This means it cannot provide cluster-level metrics, such as number of nodes or available node system resources. This information can be vital for troubleshooting issues and optimizing performance, so it is important to implement third-party solutions to capture, persist, query, and visualize cluster-level metrics.

In addition to security boundaries, organizational boundaries can also present challenges to effective observability. For example, the GIS administrator responsible for observing ArcGIS Enterprise may be different from the IT administrator responsible for observing the cluster. Because the performance of ArcGIS Enterprise on Kubernetes depends on both the state of the organization and the state of the underlying cluster infrastructure, these different administrators must closely collaborate to observe the system as a whole.