What is an incident?

An incident is any issue that needs to be investigated and resolved, often one that affects your customers.

An incident is any issue that needs to be investigated and resolved. When a connected monitoring tool detects a problem, Spike creates an incident and alerts the right person based on the escalation policy attached to that integration.

Examples

Common examples of incidents include:

• CPU or memory utilization crossing a critical threshold

• Database monitoring alerts

• Database backup failures

• Billing process failures

• Notification delivery failures

• Dashboard loading errors

• Website downtime

You can also create incidents directly from your code using a webhook integration.

Incident statuses

Every incident has one of three statuses. See incident statuses for the full breakdown.

The three incident statuses: Triggered, Acknowledged, and Resolved.

Triggered

A triggered incident means Spike has detected a problem and is actively alerting. Spike works through the escalation policy, sending alerts until someone acknowledges or resolves the incident. Repeat incidents are automatically suppressed to reduce alert fatigue.

Acknowledged

An acknowledged incident means someone is working on a fix. Spike stops alerting and pauses the escalation policy. You can optionally set an acknowledge timeout: a time limit after which Spike moves the incident back to triggered and resumes alerting if it hasn't been resolved.

Resolved

A resolved incident means the issue is fixed. Spike stops all alerts and resets the escalation policy for any future incidents on that integration.