Notes from MS Learn AZ-700 Module 8: Design and Implement Network Monitoring – Unit 2: Monitor Your Networks Using Azure Monitor
- What is Azure Monitor
- Helps to maximize availability/performance of apps and services
- Delivers comprehensive solution for collecting/analyzing/acting on telemetry from cloud and on-prem
- Info helps understand how apps are performing and identify issues affecting them and depended upon resources
- Sample list of features
- Use App Insights to detect/diagnose issues across apps
- Correlate infra issues with VM and Container insights
- Use log analytics for troubleshooting/diagnostics of monitoring data
- Support ops at scale with smart alerts/automated actions
- Create visual dashboards and workbooks in Azure
- Collect data from monitored resources using. Azure Monitor Metrics
- Diagram below offers high-level overview of Azure Monitor
- Two fundamental types of data
- Metrics
- Logs
- Left Panel is monitoring source to populate data stores
- Right Panel is actions that can be performed on said data
- Analysis
- Alerting
- Stream to ext sources
- (Image taken from MS Learn)
- Monitor data types in Azure Monitor
- Two types of data collected
- Metrics
- Azure Monitor Metrics is feature collecting numeric data from monitored resources into time series DB
- Metrics are numerical and collected at regular intervals
- Describe aspects of system at a particular time
- Lightweight/Capable of supporting near real-time – useful for alerting
- Logs
- Contain diff kinds of data organized into records with diff properties for each type
- Telemetry(events and traces) stored as logs in addition for perf data for analysis
- Metrics
- Two types of data collected
- Azure Monitor metric sources
- 4 key sources of collected
- Azure resources
- Create platform metrics giving visibility into resource health/perf
- Each resource type creates distinct metrics
- Platform metrics collected from Azure resource at 1min unless defined differently
- Applications
- App insights show metrics for monitored apps to aid in perf issue and trend detection
- Includes server response time and browser exceptions
- VM Agents
- Collected from guest OS
- Enable guest OS for Windows VM with Windows Diagnostic Extension (WAD)
- Linux VMs with InfluxData Telegraf Agent
- Custom
- Defined metrics in app monitored with App Insights
- Also custom metrics for Azure service using custom API
- Azure resources
- 4 key sources of collected
- Azure Monitor Metric tasks
- Table summarized tasks possible for using metrics
- (Table below from MS Learn)
- Metrics Explorer
- Data collected is shown in the Monitoring tab of a resource Overview Page
- Example for a VM
- CPU
- Network
- Disk
- Monitor network resources with Azure Monitor Network Insights
- Use Insights > Networks section of Azure Monitor for high-level network resource health and metrics
- Provides access to network monitoring features
- Connection Monitor
- Flow Logging for NSG
- Traffic Analytics
- Key components
- Network Health/Metrics
- Offers simple method of visualizing inventory of net resources along with resource health and alerts
- Health divided into 4 areas
- Search & filtering
- Resource health & metrics
- Alerts
- Dependency view
- Connectivity
- Providing visualization of Connection Monitor tests
- Test grouped by SRC/DST
- Traffic
- Providing access to NSG flow logs & Traffic analytics for selected subscriptions
- Grouped by location
- Searchable by any IP addr in environment
- Diagnostic Toolkit
- Providing access to all diag feature available for shoot of network/components
- Most user Network Watcher
- Capture packets for anomaly diagnoses
- Troubleshoot VPN
- Diagnose Virtual Network Gateway/Connection health
- Troubleshoot connectivity
- Check TCP connections
- Includes FQDN, URI, or IPv4 Addr
- Identify Next Hops
- Obtain next hop type/IP of packet from VM and NIC
- Knowing can help establish traffic direction to expected DST or drops
- Diagnose traffic filtering issues
- Verify packet is allowed/denied to/from VM
- Security group decision and name of rule denying traffic returned
- Network Health/Metrics