Within the Juniper Mist AI ecosystem, telemetry data and network activities are organized into distinct categories to facilitate efficient troubleshooting and monitoring. While "Organization" (A) represents a management hierarchy and "Security" (C) is a functional domain (such as Rogue AP detection or WxLAN policies), Mist specifically filters operational events into three primary categories: Site, AP (Access Point), and Client. These categories are reflected across the Mist dashboard, particularly within the Insights and Monitor views.
Site Events (E) represent macro-level issues that affect the entire physical location or a significant portion of the network infrastructure. Examples include DHCP server unreachable errors, DNS latency spikes, or site-wide configuration changes. These events are crucial for identifying systemic problems that originate from the wired side of the network or the local environment rather than individual wireless devices.
AP Events (B) focus on the health and status of the hardware itself. This category includes events such as AP reboots, firmware upgrades, Ethernet link toggling (up/down), and radio failures. By filtering for AP-specific events, administrators can quickly determine if a connectivity issue is localized to a specific hardware unit or a PoE power supply problem.
Client Events (D) provide the most granular level of visibility, tracking the specific journey of an end-user device. This includes the entire 802.11 state machine: association, authentication, and DHCP handshake. Client events are further classified as "Good," "Bad," or "Neutral," allowing Marvis AI to correlate failures (like an incorrect PSK or a slow DHCP response) to a specific user session. This three-tiered event structure—Site, AP, and Client—is the foundation of Mist's AI-driven root cause analysis, enabling the system to pinpoint exactly where in the stack a failure occurred.
