Best Practices
Get the most out of Uptime Monitor with optimal configuration settings and monitoring strategies.
Configure Fail Thresholds
Uptime Monitor uses a fail threshold algorithm to prevent alerts from temporary network glitches. Set the number of consecutive failures required before alerting.
Recommended Settings:
- Threshold 1: Not recommended - even AWS network has occasional blips
- Threshold 2-3: Good balance for most services (recommended)
- Threshold 4+: For services with known intermittent issues
Example: With threshold 2 and 10-second checks, you'll detect downtime in 10-20 seconds.
Optimize Check Intervals
Balance between quick detection and resource usage by setting appropriate check intervals.
Interval Guidelines:
- 10-30 seconds: Critical services requiring fastest detection
- 1 minute: Important production services
- 2-5 minutes: Standard services
- 5-15 minutes: Low priority (internal tools, docs)
Check Frequency Impact:
- 10-second checks: 8,640 checks per day (fastest detection)
- 30-second checks: 2,880 checks per day
- 1-minute checks: 1,440 checks per day
- 5-minute checks: 288 checks per day
Configure Notification Channels
Set up different notification methods based on severity and avoid channel limitations:
- •Critical alerts:PagerDuty, incident.io
- •Important alerts: Slack, Discord, webhooks
- •Informational: Email notifications
Tip: Use unlimited alert delivery methods (webhooks, Slack, Discord) instead of limited ones (SMS, email) to ensure you never miss critical alerts.
Test Your Contacts: Use the Test button next to each contact in Settings → Contacts to verify notification delivery is working before you need it.
Dashboard Best Practices
- Delete or pause monitors that are no longer needed
- Use consistent naming conventions for monitors
- Use descriptive names for monitors
- Set appropriate timeout values (3-10 seconds)
- Use status pages to group related services
- Pause monitors during maintenance windows
Quick Optimization Checklist
- Set fail threshold to 2-3 for most monitors
- Use 10-second to 1-minute checks for critical services for fastest downtime detection
- Configure multiple notification channels for redundancy
- Use unlimited alert delivery methods (webhooks, Slack, Discord) instead of limited ones (SMS, email)
- Pause or delete monitors for deprecated services
Ready to optimize your monitoring setup?