API

IMPulse provides simple API and WebSocket endpoints for incident management and system interaction.

Info

All endpoints use the HTTP_PREFIX prefix if configured.

General endpoints

HTTP / [GET]

Main page of the IMPulse web interface.

HTTP / [POST]

Send a new alert for processing.

Requirements:

• Server must be in primary mode
• Request body must contain valid JSON with alert data

Requirements:

• UI must be enabled in configuration ([ui section])

HTTP /app [POST]

HTTP /app [PUT]

Handle button interactions in messengers (Slack, Mattermost, Telegram).

HTTP /incidents [GET]

Get list of all incidents.

HTTP /metrics [GET]

Prometheus metrics endpoint. Returns metrics in Prometheus format for monitoring and observability.

Responses:

• 200 OK - Returns metrics in Prometheus format

HTTP /queue [GET]

Get current processing queue state.

WebSocket /ws

WebSocket connection for receiving real-time incident updates.

Requirements:

• Server must be in primary mode (see High Availability)
• Connection will be closed with code 1008 if server is in standby mode

Service endpoints

HTTP /livez [GET]

Server liveness check. Used for Kubernetes liveness probes to determine if the container is alive.

Responses:

• 200 OK - Container is alive (returns 200 in both primary and standby modes)

HTTP /readyz [GET]

Server readiness check. Used for health checks and determining server state (see High Availability).

Responses:

• 200 OK - Server is ready and running in primary mode
• 503 Service Unavailable - Server is in standby mode or initializing

HTTP /-/reload [POST]

Reload server configuration without restart.

Requirements:

• Server must be in primary mode (returns 503 in standby mode)

Responses:

• 200 OK - Configuration reloaded successfully
• 400 Bad Request - Configuration reload failed
• 500 Internal Server Error - Unexpected reload error