Scheduled Maintenance
Database patching starts at 22:00 UTC. Brief read-only window expected.
Self-hosted uptime & availability monitoring — HTTP checks, Docker image probing, live status dashboards and Prometheus metrics out of the box.
A live view of your monitored resources. Each row shows current status, resource type and a 90-check timeline bar. Green = up, red = down, yellow = degraded.
Kairos now captures response time metrics per check, helping you detect performance regression before it becomes customer-facing downtime.
Keep users informed during maintenance windows and incidents with visible, severity-tagged announcements right on your status page.
Database patching starts at 22:00 UTC. Brief read-only window expected.
Some checks show slower response times in eu-central. Team is actively monitoring.
API Gateway outage mitigated. Full post-incident report will follow.
Track incidents with clear start/end windows, impacted resources and status progression over time. The timeline view gives operators and stakeholders instant context during active events.
API Gateway failed checks in eu-central cluster
Failover enabled, error rate dropping
Service recovered, all checks operational
Query all monitored resources and their current status
Trigger an immediate health check for any resource
Ad-hoc check any URL or Docker image — no resource needed
Retrieve active or historical outages across all resources
Kairos ships with a native Model Context Protocol (MCP) server, letting AI assistants like Claude, GitHub Copilot and Cursor query and control your monitoring instance directly — no manual API calls required.
Full REST API at /api with interactive Swagger UI auto-generated from the OpenAPI spec.
| Method | Path | Auth | Description |
|---|---|---|---|
| GET | /api/resources | Public | List all active resources |
| GET | /api/resources/{id} | Public | Resource details + latest health status |
| POST | /api/resources | Admin | Add a new resource |
| DELETE | /api/resources/{id} | Admin | Delete a resource |
| GET | /api/resources/{id}/history | Auth | Full check history for a resource |
| GET | /api/announcements | Public | List all announcements |
| POST | /api/announcements | Admin | Create an announcement |
| PUT | /api/announcements/{id} | Admin | Update an announcement |
| DELETE | /api/announcements/{id} | Admin | Delete an announcement |
Interactive Swagger UI available at
/api
on your running Kairos instance.
Kairos is batteries-included — no cloud dependencies, no per-check fees, fully self-hosted.
Periodic GET checks with configurable intervals and parallelism limits. Instant checks trigger on startup.
Validate image pullability via the OCI/Docker Registry HTTP API — manifest & blob probe, no Docker socket needed.
Provide a registry prefix and Kairos auto-creates resources for all discovered images with optional recursive traversal.
24-hour timelines, uptime percentages (24h / 7d / 30d), per-resource check history with filterable tables.
Publish rich-text announcements with INFORMATION, WARNING, or PROBLEM severity — with optional auto-expiry.
Native kairos_resource_status gauge per resource at /actuator/prometheus — plug straight into Grafana.
Organise services into named groups with drag-and-drop reordering within and across groups.
Machine-to-machine access via named API keys. OIDC/OAuth2 login with any OpenID Connect provider.
Start instantly with the embedded H2 database or switch to PostgreSQL with a single property. Flyway handles migrations.
Dedicated outage display with timeline view to follow incidents from start to resolution and understand impact quickly.
Response time measurements for every check execution, including percentile-friendly visibility for spotting regressions early.
Native Model Context Protocol server — let AI assistants like Claude and GitHub Copilot query resources, trigger checks and manage outages directly.
All your services at a glance — uptime percentages, 24-hour timelines, and instant drill-down to full check history.
Choose your preferred deployment method — from source, Docker, or Helm on Kubernetes.
git clone https://github.com/wenisch-tech/kairos.git cd kairos maven spring-boot:run
docker run -d \
--name kairos \
-p 8080:8080 \
-v kairos-data:/app/data \
ghcr.io/wenisch-tech/kairos:latest
helm repo add kairos https://charts.wenisch.tech
helm repo update
helm install kairos kairos/kairos --version 1.0.4 \
-n kairos --create-namespace \
--set persistence.enabled=true
Run with Docker, from source with Maven, or deploy to Kubernetes with the bundled Helm chart. Kairos is ready on port 8080.
Default credentials: admin@kairos.local / admin. ⚠ Change the password immediately after first login via Admin → Users.
Add HTTP services or Docker images via the Admin panel or the REST API. Kairos starts checking immediately — no wait for the first interval tick.
The public dashboard shows 24-hour timelines, current status and uptime percentages for all your resources.
Explore the full documentation with setup walkthroughs, configuration guides, and operational best practices.
Kairos is production-ready — deploy it, plug it into Prometheus, and forget about it.
# HELP kairos_resource_status 1=up 0=down -1=unknown kairos_resource_status{resource_name="Website",resource_type="HTTP"} 1.0 kairos_resource_status{resource_name="API Gateway",resource_type="HTTP"} 1.0 kairos_resource_status{resource_name="Container App",resource_type="DOCKER"} 1.0
Kairos is open-source under the GNU GPLv3 licence. Self-host it in minutes with Docker or Helm, no cloud account required.