Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Diagnostics

The diagnostics commands provide tools for monitoring and troubleshooting Redis Enterprise cluster health, running diagnostic checks, and generating diagnostic reports.

Overview

Redis Enterprise includes a built-in diagnostics system that performs various health checks on the cluster, nodes, and databases. These checks help identify potential issues before they become critical problems.

Available Commands

Get Diagnostics Configuration

Retrieve the current diagnostics configuration:

# Get full diagnostics config
redisctl enterprise diagnostics get

# Get specific configuration fields
redisctl enterprise diagnostics get -q "enabled_checks"

Update Diagnostics Configuration

Modify diagnostics settings:

# Update from JSON file
redisctl enterprise diagnostics update --data @diagnostics-config.json

# Update from stdin
echo '{"check_interval": 300}' | redisctl enterprise diagnostics update --data -

# Disable specific checks
redisctl enterprise diagnostics update --data '{"disabled_checks": ["memory_check", "disk_check"]}'

Run Diagnostics Checks

Trigger diagnostic checks manually:

# Run all diagnostics
redisctl enterprise diagnostics run

# Run with specific parameters
redisctl enterprise diagnostics run --data '{"checks": ["connectivity", "resources"]}'

List Available Checks

View all available diagnostic checks:

# List all checks
redisctl enterprise diagnostics list-checks

# Output as table
redisctl enterprise diagnostics list-checks -o table

Get Latest Report

Retrieve the most recent diagnostics report:

# Get latest report
redisctl enterprise diagnostics last-report

# Get specific sections
redisctl enterprise diagnostics last-report -q "cluster_health"

Get Specific Report

Retrieve a diagnostics report by ID:

# Get report by ID
redisctl enterprise diagnostics get-report <report_id>

# Get report summary only
redisctl enterprise diagnostics get-report <report_id> -q "summary"

List All Reports

View all available diagnostics reports:

# List all reports
redisctl enterprise diagnostics list-reports

# List recent reports only
redisctl enterprise diagnostics list-reports --data '{"limit": 10}'

# Filter by date range
redisctl enterprise diagnostics list-reports --data '{"start_date": "2024-01-01", "end_date": "2024-01-31"}'

Diagnostic Check Types

Common diagnostic checks include:

  • Resource Checks

    • Memory utilization
    • CPU usage
    • Disk space
    • Network bandwidth

  • Cluster Health

    • Node connectivity
    • Replication status
    • Shard distribution
    • Quorum status

  • Database Health

    • Endpoint availability
    • Persistence status
    • Backup status
    • Module functionality

  • Security Checks

    • Certificate expiration
    • Authentication status
    • Encryption settings
    • ACL configuration

Configuration Examples

Enable Automatic Diagnostics

{
  "enabled": true,
  "auto_run": true,
  "check_interval": 3600,
  "retention_days": 30,
  "email_alerts": true,
  "alert_recipients": ["ops@example.com"]
}

Configure Check Thresholds

{
  "thresholds": {
    "memory_usage_percent": 80,
    "disk_usage_percent": 85,
    "cpu_usage_percent": 75,
    "certificate_expiry_days": 30
  }
}

Disable Specific Checks

{
  "disabled_checks": [
    "backup_validation",
    "module_check"
  ],
  "check_timeout": 30
}

Practical Examples

Daily Health Check Script

#!/bin/bash
# Run daily diagnostics and email report

# Run diagnostics
redisctl enterprise diagnostics run

# Get latest report
REPORT=$(redisctl enterprise diagnostics last-report)

# Check for critical issues
CRITICAL=$(echo "$REPORT" | jq '.issues | map(select(.severity == "critical")) | length')

if [ "$CRITICAL" -gt 0 ]; then
  # Send alert for critical issues
  echo "$REPORT" | mail -s "Redis Enterprise: Critical Issues Found" ops@example.com
fi

Monitor Cluster Health

# Continuous health monitoring
watch -n 60 'redisctl enterprise diagnostics last-report -q "summary" -o table'

Generate Monthly Report

# Get all reports for the month
redisctl enterprise diagnostics list-reports \
  --data '{"start_date": "2024-01-01", "end_date": "2024-01-31"}' \
  -o json > monthly-diagnostics.json

# Extract key metrics
jq '.[] | {date: .timestamp, health_score: .summary.health_score}' monthly-diagnostics.json

Pre-Maintenance Check

# Run comprehensive diagnostics before maintenance
redisctl enterprise diagnostics run --data '{
  "comprehensive": true,
  "include_logs": true,
  "validate_backups": true
}'

# Wait for completion and check results
sleep 30
redisctl enterprise diagnostics last-report -q "ready_for_maintenance"

Report Structure

Diagnostics reports typically include:

{
  "report_id": "diag-12345",
  "timestamp": "2024-01-15T10:30:00Z",
  "cluster_id": "cluster-1",
  "summary": {
    "health_score": 95,
    "total_checks": 50,
    "passed": 48,
    "warnings": 1,
    "failures": 1
  },
  "cluster_health": {
    "nodes": [...],
    "databases": [...],
    "replication": {...}
  },
  "resource_usage": {
    "memory": {...},
    "cpu": {...},
    "disk": {...}
  },
  "issues": [
    {
      "severity": "warning",
      "component": "node-2",
      "message": "Disk usage at 82%",
      "recommendation": "Consider adding storage"
    }
  ],
  "recommendations": [...]
}

Best Practices

  1. Schedule Regular Checks - Run diagnostics daily or weekly
  2. Monitor Trends - Track health scores over time
  3. Set Up Alerts - Configure email alerts for critical issues
  4. Archive Reports - Keep historical reports for trend analysis
  5. Pre-Maintenance Checks - Always run diagnostics before maintenance
  6. Custom Thresholds - Adjust thresholds based on your environment

Integration with Monitoring

The diagnostics system can be integrated with external monitoring tools:

# Export to Prometheus format
redisctl enterprise diagnostics last-report -q "metrics" | \
  prometheus-push-gateway

# Send to logging system
redisctl enterprise diagnostics last-report | \
  logger -t redis-diagnostics

# Create JIRA ticket for issues
ISSUES=$(redisctl enterprise diagnostics last-report -q "issues")
if [ -n "$ISSUES" ]; then
  create-jira-ticket --project OPS --summary "Redis Diagnostics Issues" --description "$ISSUES"
fi

Troubleshooting

Diagnostics Not Running

# Check if diagnostics are enabled
redisctl enterprise diagnostics get -q "enabled"

# Enable diagnostics
redisctl enterprise diagnostics update --data '{"enabled": true}'

Reports Not Generated

# Check last run time
redisctl enterprise diagnostics get -q "last_run"

# Trigger manual run
redisctl enterprise diagnostics run

Missing Checks

# List disabled checks
redisctl enterprise diagnostics get -q "disabled_checks"

# Re-enable all checks
redisctl enterprise diagnostics update --data '{"disabled_checks": []}'