Core Concepts
Incidents
Incidents track service disruptions and communicate status updates to your users. They follow a structured lifecycle from detection to resolution.
Incident Lifecycle
Every incident progresses through four stages. At each stage, you can post an update message that subscribers receive via email.
Investigating
The issue has been detected and the team is looking into it. This is the default status when an incident is created.
Identified
The root cause has been identified and the team is working on a fix.
Monitoring
A fix has been applied and the team is monitoring to ensure stability.
Resolved
The incident has been fully resolved. A resolution timestamp is recorded.
Impact Levels
Each incident has an impact level that determines how it's displayed on the status page:
| Impact | Description | Status Page Banner |
|---|---|---|
| None | Informational, no user impact | Grey |
| Minor | Small subset of users affected | Yellow |
| Major | Significant impact on service | Orange |
| Critical | Complete service disruption | Red |
Creating an Incident
Navigate to Dashboard → Incidents and click Create Incident. You'll need to provide:
- Title — A clear, concise summary (e.g., "API Latency Increase")
- Impact — None, Minor, Major, or Critical
- Message — Initial update describing the issue
- Affected Components — Select which components are impacted
Incident Updates
As you work through an incident, post updates to keep your users informed. Each update includes a new status and a message. Updates are displayed chronologically on the status page and sent to subscribers.
Best practice: Post updates every 30 minutes during active incidents, even if there's no new information. Users appreciate knowing you're still working on it.
Auto-Created Incidents
When a monitor detects failures exceeding the configured threshold, PingBase automatically creates an incident with the "Investigating" status and "Minor" impact. The affected component's status is updated to reflect the outage. When the monitor recovers, the incident is automatically resolved.