Configuration
Learn how to configure and customize the Giant Swarm observability platform for your organization's needs.
The Giant Swarm observability platform offers extensive configuration options to tailor the system to your organization’s specific monitoring, alerting, and visualization needs. This overview covers the key areas you can configure to create a customized observability experience that fits your operational requirements.
Platform-level configuration
Configure foundational platform settings that affect the entire observability stack:
Initial platform setup
Work with Giant Swarm to configure core platform components:
- OIDC provider integration: Connect your identity provider for authentication
- Base domain configuration: Set up DNS and ingress for platform access
Multi-tenancy and access control
Multi-tenancy forms the foundation of platform configuration, enabling you to:
- Isolate data between teams, environments, or projects through tenant separation
- Control access by mapping Grafana organizations to specific tenants and user groups
- Manage permissions with role-based access controls (RBAC) from your identity provider
- Configure tenant lifecycle including creation, management, and removal processes
Multi-tenancy configuration affects all other platform components, making it essential to understand before configuring other areas.
Organization management
Manage your observability platform’s organizational structure through GrafanaOrganization resources. Organization management enables you to:
- Create isolated workspaces: Set up dedicated Grafana organizations for teams, environments, or projects
- Configure access controls: Map identity provider groups to appropriate roles (admin, editor, viewer)
- Define tenant access: Control which data tenants each organization can access
- Provision datasources: Automatically configure tenant-scoped access to metrics, logs, and alerting
Effective organization management provides teams with secure, isolated access to their observability data while maintaining proper governance and data separation. Contact your Giant Swarm account engineer for guidance on creating and configuring Grafana organizations with practical examples and step-by-step instructions.
Data ingestion configuration
Control what observability data flows into your platform and how it’s processed:
Metrics collection
Configure metrics ingestion to collect custom application and infrastructure metrics:
- ServiceMonitors: Define which services to scrape for metrics and how frequently
- PodMonitors: Configure direct pod-level metric collection for specialized use cases
- Tenant routing: Specify which tenant receives metrics through labeling
- Metric transformation: Apply relabeling rules to modify, filter, or enrich metrics before storage
Log collection
Configure log ingestion to capture application and system logs:
- PodLogs: Select which pods to collect logs from using label selectors
- Pod labels: Enable automatic log collection by labeling pods with tenant information
- Log parsing: Configure structured parsing and label extraction from log content
- Namespace filtering: Control log collection scope by namespace
External data sources
Integrate data from sources outside Giant Swarm managed clusters through the Observability Platform API:
- OIDC authentication: Configure your identity provider for secure API access
- Data ingestion endpoints: Send metrics, logs, and traces from external systems
- OpenTelemetry Protocol (OTLP): Configure standardized telemetry data ingestion
Visualization and dashboard configuration
Once you’ve created Grafana organizations as part of your multi-tenancy setup, you can leverage them for powerful visualization capabilities:
Custom dashboards
Create custom dashboards tailored to your specific monitoring needs:
- GitOps approach: Version-controlled dashboard definitions using ConfigMaps
- Dashboard provisioning: Automatically deploy dashboards to specific organizations
- UI creation: Build dashboards directly in Grafana with persistent storage
- Cross-tenant dashboards: Combine data from multiple tenants when authorized
Alerting and notification configuration
Set up comprehensive alerting to provide timely notification of issues:
Alert rules
Create alerting and recording rules for proactive monitoring:
- Metric-based alerts: PromQL expressions for infrastructure and application monitoring
- Log-based alerts: LogQL expressions for application error detection and log pattern analysis
- Recording rules: Pre-computed metrics for improved dashboard performance
- Tenant federation: Create rules that query data across multiple tenants
Alert management
Configure Alertmanager for comprehensive alert handling and operational workflows:
- Notification channels: Integration with Slack, email, PagerDuty, and other services
- Contact points: Configure notification destinations and templates
- Routing rules: Define which alerts go to which teams based on labels and severity
- Notification policies: Define routing logic based on alert characteristics
- Alert grouping: Combine related alerts to reduce notification fatigue
- Silences: Manage alert suppression for planned maintenance or known issues
- Alert templates: Customize notification formatting and content
Configuration best practices
Start simple and iterate
Begin with basic configurations and gradually add complexity:
- Start with environment-based tenants (
production,staging,development) - Implement essential alerting rules before expanding to comprehensive monitoring
- Create basic dashboards before building complex visualization workflows
Plan for scale
Design configurations that grow with your organization:
- Choose tenant strategies that align with your organizational structure
- Design alert routing that scales with team growth
- Plan dashboard organization that remains manageable as teams expand
Maintain consistency
Establish patterns and standards across your configuration:
- Use consistent naming conventions for tenants, alerts, and dashboards
- Standardize alert severity levels and routing patterns
- Create reusable templates for common monitoring patterns
Getting started
To begin configuring your observability platform:
- Understand multi-tenancy: Work with your Giant Swarm account engineer to plan your data organization strategy
- Set up Grafana organizations: Create organizations for your teams with assistance from Giant Swarm
- Configure data ingestion: Set up metrics and log collection for your applications
- Create essential alerts: Implement basic alerting rules for critical system health
- Build dashboards: Create custom dashboards for your specific monitoring needs
For platform-level configuration assistance, contact your Giant Swarm account engineer who can help with OIDC provider setup, resource allocation, and other foundational configurations.