Skip to content

monitoring-quick-reference.md

Latest commit

 

History

History
149 lines (112 loc) · 4.7 KB

monitoring-quick-reference.md

File metadata and controls

149 lines (112 loc) · 4.7 KB

Monitoring Quick Reference

Azlin provides monitoring through dedicated subcommands — there is no azlin monitor namespace.

Commands

# Health Dashboard (Four Golden Signals)
azlin health                                       # All VMs in default RG
azlin health --vm dev-vm-01                        # Single VM
azlin health --tui                                 # Interactive TUI dashboard
azlin health --tui --interval 5                    # TUI with 5s refresh
azlin health --resource-group my-rg                # Filter by RG

# Distributed Real-Time Monitoring
azlin top                                          # Live monitoring dashboard
azlin top --interval 5                             # 5s refresh
azlin top --vm dev-vm-01                           # Single VM
azlin top --ip 10.0.1.4                            # Direct IP (skip Azure lookup)

# Cost Intelligence
azlin costs dashboard                              # Current spending dashboard
azlin costs history                                # Historical cost trends
azlin costs budget                                 # Budget management
azlin costs recommend                              # Optimization recommendations
azlin costs actions                                # Execute cost-saving actions

# VM Status & Processes
azlin status VM_NAME                               # Detailed VM status
azlin w                                            # Who's logged in (all VMs)
azlin ps                                           # Process list (all VMs)

# Logs
azlin logs VM_NAME                                 # View syslog (default, 100 lines)
azlin logs VM_NAME -t auth                         # Auth logs
azlin logs VM_NAME -t cloud-init                   # Cloud-init logs
azlin logs VM_NAME -t azlin                        # Azlin application logs
azlin logs VM_NAME -t all                          # All log sources
azlin logs VM_NAME -f                              # Follow logs (tail -f)
azlin logs VM_NAME -n 500                          # Last 500 lines

Health Dashboard Flags

Flag Description Default
--vm Check a single VM by name all VMs
--tui Launch interactive TUI dashboard off
--interval Refresh interval in seconds (TUI only) 5
--resource-group Filter to resource group default RG

Top Dashboard Flags

Flag Description Default
-i, --interval Refresh interval in seconds 10
-t, --timeout SSH timeout per VM 5
--vm Target a single VM all VMs
--ip Direct IP (skip Azure lookup) —

Color Coding

  • 🟢 Green (<70%): Normal
  • 🟡 Yellow (70-85%): Elevated
  • 🔴 Red (>85%): High

Default Alert Rules

Rule Metric Threshold Severity
high_cpu cpu_percent >80% warning
critical_cpu cpu_percent >95% critical
high_memory memory_percent >85% warning
critical_memory memory_percent >95% critical
disk_space disk_percent >90% warning

Metrics

Metric Description Unit
cpu_percent CPU utilization 0-100%
memory_percent Memory utilization 0-100%
disk_read_bytes Disk read throughput Bytes/sec
disk_write_bytes Disk write throughput Bytes/sec
network_in_bytes Network ingress Bytes/sec
network_out_bytes Network egress Bytes/sec

Configuration Files

  • ~/.azlin/metrics.db - Historical metrics database
  • ~/.azlin/alert_rules.yaml - Alert rules configuration

Troubleshooting

# Check Azure authentication
az account show

# Verify VM access
az vm list --output table

# View VM status
azlin status VM_NAME

# Check health for a single VM
azlin health --vm VM_NAME --verbose

Common Tasks

Monitor a Specific VM

# Quick health check
azlin health --vm dev-vm-01

# Live resource monitoring
azlin top --vm dev-vm-01 --interval 5

# View recent logs
azlin logs dev-vm-01 -n 200

Fleet-Wide Overview

# Health dashboard for all VMs
azlin health --tui

# Cost overview
azlin costs dashboard

# Who's logged in
azlin w

Performance Tips

  1. Increase refresh interval for many VMs: azlin top --interval 30
  2. Filter by resource group to reduce API calls: --resource-group my-rg
  3. Run in screen/tmux for long-running dashboards
  4. Use --vm flag to focus on a single VM when debugging

Security Best Practices

  • Never commit alert_rules.yaml with secrets to version control
  • Use environment variables for webhook tokens
  • Rotate SMTP passwords regularly if using email-based alerts
  • Restrict database file permissions: chmod 600 ~/.azlin/metrics.db