In the Prisma SD-WAN ecosystem, the centralized cloud controller provides a robust multi-layered visibility framework to ensure network reliability. To effectively detect critical events like site outages or application performance issues (SLA violations), the controller aggregates several types of operational data. Incidents are the primary mechanism for high-level alerting; they are automatically generated when the system detects significant state changes, such as an ION device going offline (site outage) or a path failing to meet the required performance metrics.
While Incidents provide the "what" and "where," Alerts offer granular notifications for specific events that may not yet have escalated to a full-scale incident. To provide deep context, the controller also utilizes Statistics, which include real-time and historical telemetry regarding bandwidth, latency, jitter, and packet loss. These statistics allow administrators to visualize the specific SLA violation as it occurs. Furthermore, Audit logs are essential for tracking configuration changes or administrative actions that might have preceded an outage, helping engineers correlate human intervention with network behavior.
By combining these four elements—Incidents for major events, Alerts for specific notifications, Statistics for performance validation, and Audit logs for change tracking—a network administrator gains a 360-degree view of the fabric. This comprehensive approach moves beyond simple "up/down" monitoring, allowing for "Day 2" operational excellence where performance degradation is identified and remediated before it impacts the end-user experience.
