Skip to content

Monitoring Signals

Azure provides a comprehensive monitoring solution to track the health and performance of your virtual machines. These signals help you proactively identify and resolve issues before they impact your users.

Signal Category Metric/Log Name Source Normal Range Alert Threshold Tool
Compute Percentage CPU Host < 70% > 85% for 15m Azure Monitor
Memory Available Memory Guest OS via AMA / VM Insights > 10% < 5% VM Insights
Disk IO Data Disk IOPS Host Within SKU limit > 90% of limit Azure Monitor
Latency Disk Read/Write Latency Host Varies by disk tier Sustained increase above baseline VM Insights
Network Network In/Out Host Varies by size > 90% of BW Azure Monitor
Health VM Availability Platform 1 (Available) < 1 Resource Health
Diagnosis Boot Diagnostics Serial/Store Image/Text "Failure" string Serial Console
Audit Activity Log Resource Event log Critical failures Log Analytics
graph TD
    VM[Virtual Machine] -->|Host Metrics| Monitor[Azure Monitor]
    VM -->|Guest Logs/Metrics| Agent[AMA Agent]
    Agent -->|Store| LogA[Log Analytics]
    Monitor -->|Analyze| Dashboard[Azure Dashboards]
    LogA -->|Analyze| Insights[VM Insights]
    Monitor -->|Threshold| Alerts[Azure Alerts]
    LogA -->|Threshold| Alerts

Note

The Azure Monitor Agent (AMA) replaces the legacy Log Analytics agent and provides enhanced security and flexibility for log collection.

See Also

Sources