- 2 minutes to read

How do I configure alerts for different severity levels (Warning vs. Error)?

Eliminate alert fatigue with Nodinite graduated severity levels. Configure intelligent dual-threshold monitoring that provides early warnings without crying wolf, and critical alerts that demand immediate action.

✅ Set Warning threshold - Early signal (75% of normal) for proactive investigation
✅ Set Error threshold - Critical level (50% of normal) for immediate escalation
✅ Graduate response actions - Different recipients, frequencies, and escalation paths
✅ Prevent false escalations - Gap between thresholds allows appropriate responses

Set dual thresholds per Non-Events resource. E-commerce order monitoring example:

Non-Events resource: "Order Confirmations - Business Hours"

  • Warning threshold: Min 600 orders/hour (75% of normal 800, early heads-up)

    • Alert recipient: Operations team email + Slack #operations-alerts channel
    • Alert frequency: Every 30 minutes while condition persists
    • Escalation: None (operational awareness, investigate when convenient)
    • Action: Check order processing service health, review error logs, monitor trend (is volume dropping further?)

  • Error threshold: Min 400 orders/hour (50% of normal 800, critical data outage)

    • Alert recipient: Operations team pager + SMS to on-call engineer + Slack #critical-alerts
    • Alert frequency: Immediately, repeat every 5 minutes until resolved
    • Escalation: If not acknowledged within 15 minutes, escalate to engineering manager + CTO
    • Action: Immediate investigation, assume critical outage (payment gateway down, integration service crashed, API authentication failure), engage incident response procedure

Threshold tuning rationale:

  • Warning at 75%: Provides early signal (volume trending down but not catastrophic, proactive investigation prevents Error)
  • Error at 50%: Clear critical threshold (half normal volume = obvious data outage, not normal variance)
  • Gap between Warning/Error: Allows graduated response (Warning = watch, Error = act immediately), reduces false escalations

AckNak severity example:

  • Warning: Payment acknowledgment >25 seconds (trending toward 30-second SLA violation, proactive capacity planning)
  • Error: Payment acknowledgment >30 seconds (SLA violated, regulatory compliance issue, customer escalation risk)

Next Step

Need more help? Check the Troubleshooting Overview for all FAQs, or refer to the Configuration Guide to fine-tune your Non-Events monitoring setup.