Instance Management in Kestra Enterprise – Services and Maintenance

Available on:

The Instance menu gives you a centralized view of your Kestra deployment's health along with controls for upgrades, maintenance, and user notifications.

Overview

The Instance menu provides centralized control over your platform’s infrastructure so administrators can monitor service health, manage configurations, and communicate announcements (like planned maintenance downtime) to users without relying on additional observability tools. The System Overview tab gives a high-level snapshot of your instance’s operational status. Here, you can:

View License Details: validity, usage statistics, and installed secrets and storage plugins.

Instance System Overview

Services

Kestra tracks the health of critical components, including:

Workers: Execute tasks.
Schedulers: Trigger workflows.
Executors: Manage task execution.
Webservers: Host the UI and API.

Instance Overview Table

Each service displays:

State: Whether the service is active (RUNNING) or unresponsive.
Host Name: Identifier of the server/pod (e.g., kafka-ee-preview-79fb7755f8-zhlhq).
Server Type: For example, STANDALONE.
Version
Start Date
Health Check Date

Server information and liveness

Each service instance provides technical details for debugging when clicked on:

Hostname: Identifier of the server/pod (e.g., kafka-ee-preview-79fb7755f8-zhlhq).
Session Timeout: Time before an unresponsive service is marked offline (e.g., 60 seconds).
Heartbeat Interval: The expected time between heartbeats.
Last Heartbeat: Timestamp of the latest health check.
Termination Grace Period: The expected time for this service to complete all its tasks before initiating a graceful shutdown.

Services Overview

Additional tabs include Configuration to display port configuration, Metrics such as CPU Usage and Executor Thread Count:

Service Metrics

And an Events Timeline to give an overview of the service's lifecycle:

Service Events

Announcements

Notify users about planned maintenance or updates:

Create Announcements: Specify a title, message, and date range.
Choose Type: Define severity of the announcement (e.g., info, warning, error).

Announcements appear in the UI during the selected period, ensuring users stay informed.

Maintenance Mode

From the Instance - Services tab, enter Maintenance Mode to temporarily pause all workflows and services for upgrades:

Services enter a paused state, and new executions are blocked.
Combine it with the Announcements feature so users see a maintenance banner while running workflows gracefully terminate.

Worker Groups

Create Worker Groups to isolate workloads or delegate tasks to specific workers:

Add Worker Groups: Define groups with specific resource limits or labels.
Assign Tasks: Route workflows to designated groups via worker group key within a task or trigger.