Skip to main content

Service Catalog

The RunOS marketplace provides 20+ pre-configured open-source services ready to deploy with a single click. Each service includes intelligent defaults, automatic configuration, and integrated monitoring.

Databases​

MySQL​

Description: Popular open-source relational database management system known for reliability and ease of use.

Good for:

  • Web applications (WordPress, Drupal, etc.)
  • Content management systems
  • E-commerce platforms
  • Structured data with ACID compliance

Commonly used for:

  • Traditional web application backends
  • Application state storage
  • Reporting and analytics workloads
  • Multi-tenant applications

High Availability: Deploy with 2+ replicas on multiple worker nodes for primary-replica replication with automatic failover. Read replicas distribute read traffic across multiple instances.

Backups: Automated backups via MySQL operator with S3 storage. Daily backups with configurable retention. Point-in-time recovery available through binary log archiving.

PostgreSQL​

Description: Advanced open-source relational database with extensive features including JSONB support, full-text search, and robust data integrity.

Good for:

  • Complex queries and analytics
  • JSON document storage with relational features
  • Geospatial data (PostGIS extension)
  • Financial and accounting systems requiring strict ACID compliance

Commonly used for:

  • Modern web application backends
  • Data warehousing and analytics
  • Time-series data
  • Multi-tenant SaaS applications

High Availability: Deploy with 2+ replicas on multiple worker nodes. CloudNativePG operator provides automatic failover, synchronous replication, and read replicas. Self-healing capabilities with automatic pod recovery.

Backups: Industry-leading backup solution via CloudNativePG. Automated daily backups with WAL archiving for point-in-time recovery. Supports local storage and S3-compatible destinations. Near-zero data loss with continuous archiving.

FerretDB​

Description: MongoDB-compatible database built on PostgreSQL, providing MongoDB API while storing data in PostgreSQL.

Good for:

  • MongoDB applications requiring PostgreSQL reliability
  • Teams familiar with MongoDB wanting PostgreSQL benefits
  • Applications needing document database flexibility with SQL backend
  • Migration from MongoDB to open-source alternative

Commonly used for:

  • Document-oriented applications
  • Rapid prototyping and development
  • Applications with flexible schema requirements
  • MongoDB to PostgreSQL migrations

High Availability: Leverages underlying PostgreSQL HA configuration. Benefits from PostgreSQL operator's automatic failover and replication.

Backups: Uses PostgreSQL backup infrastructure. All PostgreSQL backup features available including point-in-time recovery.

Valkey​

Description: High-performance in-memory data structure store, fully compatible with Redis. Open-source alternative following Redis license change.

Good for:

  • Caching frequently accessed data
  • Session storage for web applications
  • Real-time analytics and leaderboards
  • Message queuing and pub/sub

Commonly used for:

  • Application-level caching
  • Session management
  • Rate limiting
  • Real-time data processing

High Availability: Deploy with 2+ replicas on multiple worker nodes for Redis Sentinel with automatic failover. Multiple replicas ensure data availability during failures.

Backups: RDB snapshots and AOF (Append-Only File) persistence. Automatic snapshots based on configurable save policies. Manual backups via BGSAVE command.

ClickHouse​

Description: Column-oriented database optimized for analytics workloads, providing extremely fast query performance on large datasets.

Good for:

  • Real-time analytics and reporting
  • Log aggregation and analysis
  • Time-series data at scale
  • Business intelligence and data warehousing

Commonly used for:

  • Web analytics platforms
  • Application performance monitoring
  • IoT data analysis
  • Financial market data analysis

High Availability: Supports replication across multiple nodes. Distributed tables provide automatic data distribution and parallel query execution.

Backups: Manual backups using clickhouse-backup tool. Can export to S3-compatible storage. Consider replication for high-availability scenarios.

Data & Streaming​

Kafka​

Description: Distributed event streaming platform for building real-time data pipelines and streaming applications.

Good for:

  • Event-driven architectures
  • Log aggregation at scale
  • Real-time data pipelines
  • Microservice communication

Commonly used for:

  • Stream processing applications
  • Activity tracking and monitoring
  • Messaging between microservices
  • Data integration and ETL pipelines

High Availability: Distributed architecture with topic replication across multiple brokers. Automatic leader election ensures availability during failures.

Backups: MirrorMaker for topic replication to secondary cluster. Configuration backups for topics and schemas. Consider replication for disaster recovery.

Zookeeper​

Description: Centralized service for maintaining configuration, naming, distributed synchronization, and group services.

Good for:

  • Configuration management for distributed systems
  • Service discovery and coordination
  • Distributed locking and leader election
  • Cluster coordination (required for Kafka)

Commonly used for:

  • Kafka cluster coordination
  • Distributed application configuration
  • Service registry
  • Distributed consensus

High Availability: Ensemble of multiple nodes (typically 3 or 5) provides fault tolerance. Quorum-based architecture ensures consistency.

Backups: Export configuration snapshots. Primarily used for coordination, not data storage, so backup needs are minimal.

RabbitMQ​

Description: Message broker implementing Advanced Message Queuing Protocol (AMQP) with support for multiple messaging protocols.

Good for:

  • Reliable message delivery between applications
  • Task queues and work distribution
  • RPC (Remote Procedure Call) patterns
  • Complex routing scenarios

Commonly used for:

  • Background job processing
  • Email sending queues
  • Notification systems
  • Microservice communication

High Availability: Mirrored queues across multiple nodes. Automatic queue replication ensures messages survive node failures.

Backups: Export queue and exchange definitions. Message persistence to disk available. Consider mirroring for critical queues.

Vector Pipelines​

Description: High-performance observability data pipeline for collecting, transforming, and routing logs, metrics, and traces.

Good for:

  • Centralized log aggregation
  • Log parsing and enrichment
  • Routing logs to multiple destinations
  • Log-based metrics generation

Commonly used for:

  • Centralized logging infrastructure
  • Log transformation and filtering
  • Multi-destination log routing
  • Observability data processing

High Availability: Deploy multiple instances with load balancing. Stateless architecture allows horizontal scaling.

Backups: Configuration backups. Logs are transient data flowing through the pipeline, stored at destinations.

Storage & Registry​

MinIO​

Description: High-performance, S3-compatible object storage built for cloud-native applications.

Good for:

  • Object storage for applications
  • Backup and archive storage
  • Media and asset storage
  • Machine learning dataset storage

Commonly used for:

  • Application file storage
  • Backup destinations (PostgreSQL, MySQL, etc.)
  • Static asset hosting
  • Data lake storage

High Availability: Distributed mode with erasure coding provides data redundancy across multiple nodes. Supports multi-site replication for disaster recovery.

Backups: Object versioning maintains multiple versions of each object. Replication to secondary MinIO instance for geographic redundancy. Lifecycle policies for automated version management.

Longhorn​

Description: Cloud-native distributed block storage for Kubernetes with built-in backup and disaster recovery.

Good for:

  • Persistent storage for stateful applications
  • Storage that needs to move between nodes
  • Distributed storage with replication
  • Snapshot and backup capabilities

Commonly used for:

  • Distributed application storage
  • Storage requiring high availability
  • Multi-node deployments
  • Applications needing storage snapshots

High Availability: Replicates data across multiple nodes (typically 3 replicas). Storage survives node failures with automatic failover.

Backups: Built-in snapshot capabilities. Backup to S3-compatible storage. Incremental backups reduce storage requirements.

Harbor​

Description: Open-source container registry with security scanning, image signing, and replication.

Good for:

  • Private container image storage
  • CI/CD pipelines requiring image registry
  • Image vulnerability scanning
  • Multi-registry replication

Commonly used for:

  • Storing application container images
  • Build artifact storage
  • Helm chart repository
  • Internal image distribution

High Availability: Supports database and Redis backend HA configurations. Multi-node deployment with load balancing.

Backups: Database backups (PostgreSQL) for registry metadata. Storage backend (local or S3) for image layers. Replication to secondary Harbor instance for DR.

Networking & Security​

Traefik​

Description: Modern HTTP reverse proxy and load balancer with automatic HTTPS and dynamic configuration.

Good for:

  • Ingress controller for Kubernetes
  • Automatic SSL/TLS certificate management
  • Load balancing HTTP/HTTPS traffic
  • Service routing and discovery

Commonly used for:

  • Exposing services externally (port 8890 for dashboards)
  • Automatic Let's Encrypt certificates
  • Traffic routing to services
  • API gateway functionality

High Availability: Multiple replicas distribute traffic. Session affinity and connection pooling maintain reliability.

Backups: Configuration stored in Kubernetes resources. Dynamic configuration from service discovery.

Cert Manager​

Description: Kubernetes certificate management controller for automated SSL/TLS certificate issuance and renewal.

Good for:

  • Automatic Let's Encrypt certificates
  • Certificate lifecycle management
  • Internal CA certificate issuance
  • Wildcard certificate management

Commonly used for:

  • Cluster domain wildcard certificates
  • Service HTTPS endpoints
  • Internal service mTLS
  • Certificate rotation automation

High Availability: Single controller instance sufficient. Certificate storage in Kubernetes secrets provides redundancy.

Backups: Certificates stored in Kubernetes secrets. Automatic renewal prevents expiration. Backup secrets for disaster recovery.

WireGuard​

Description: Fast, modern, and secure VPN tunnel with minimal configuration and high performance.

Good for:

  • Secure remote cluster access (User VPN on port 32768)
  • Inter-node cluster networking (wg0 on port 51820)
  • Encrypted communication channels
  • Site-to-site VPN connections

Commonly used for:

  • Remote access to cluster services
  • Kubernetes node mesh networking
  • Secure administrator access
  • Encrypted service communication

High Availability: Mesh topology connects all nodes. Multiple endpoints provide redundancy.

Backups: Configuration stored in node configuration. VPN keys in Kubernetes secrets and node filesystem.

Observability​

Grafana​

Description: Multi-platform analytics and interactive visualization platform with support for multiple data sources.

Good for:

  • Metrics visualization and dashboards
  • Alerting and notification
  • Multi-source data correlation
  • Team collaboration on monitoring

Commonly used for:

  • Infrastructure monitoring dashboards
  • Application metrics visualization
  • Service-level indicator (SLI) tracking
  • Custom business metrics

High Availability: Multiple instances with shared database backend. Session storage in database or Redis.

Backups: Dashboard and configuration stored in database (PostgreSQL or MySQL). Export dashboards as JSON. Database backups preserve all configuration.

Prometheus​

Description: Open-source monitoring and alerting toolkit with powerful query language and time-series database.

Good for:

  • Metrics collection and storage
  • Service monitoring and alerting
  • Infrastructure monitoring
  • Custom application metrics

Commonly used for:

  • Kubernetes cluster monitoring
  • Service health checks
  • Resource utilization tracking
  • Performance monitoring

High Availability: Federation and remote write for multi-instance setup. Alertmanager clustering for alert distribution.

Backups: Time-series data typically ephemeral (15-30 day retention). Export critical metrics to long-term storage. Configuration in Kubernetes resources.

LangFuse​

Description: Open-source observability platform for Large Language Model (LLM) applications with tracing and analytics.

Good for:

  • LLM application monitoring
  • Prompt and completion tracking
  • Cost analysis for LLM APIs
  • Performance optimization

Commonly used for:

  • AI application debugging
  • LLM usage analytics
  • Prompt engineering optimization
  • Production LLM monitoring

High Availability: Scalable backend with database storage. Multiple instances handle increased load.

Backups: PostgreSQL backend stores all traces and analytics. Standard database backups preserve all observability data.

AI Models & Serving​

Ollama​

Description: Platform for running Large Language Models locally with simple API for model management and inference.

Good for:

  • Local LLM deployment
  • Privacy-focused AI applications
  • Custom model fine-tuning
  • Development and testing AI features

Commonly used for:

  • Self-hosted ChatGPT alternatives
  • AI-powered applications
  • Model experimentation
  • On-premise AI requirements

High Availability: Multiple instances behind load balancer. Model storage on shared storage for consistency.

Backups: Model files stored in persistent volumes. Backup model directory and configuration. Models can be re-downloaded if needed.

LiteLLM​

Description: Unified API proxy for multiple LLM providers (OpenAI, Anthropic, Cohere, etc.) with load balancing and fallback.

Good for:

  • Multi-provider LLM applications
  • Load balancing across LLM endpoints
  • Fallback to alternative providers
  • Cost optimization across providers

Commonly used for:

  • Production LLM applications
  • Provider abstraction layer
  • LLM request routing
  • Multi-model applications

High Availability: Stateless proxy allows multiple instances. Built-in failover to alternative providers.

Backups: Configuration-focused service. Backup configuration and API keys. No persistent data storage.

AI Applications​

Open WebUI​

Description: Feature-rich web interface for local LLMs with chat interface, model management, and document integration.

Good for:

  • ChatGPT-like interface for local LLMs
  • Team access to self-hosted AI
  • Document chat and RAG applications
  • Model comparison and testing

Commonly used for:

  • Internal AI assistants
  • Document question answering
  • Team collaboration with AI
  • AI application prototyping

High Availability: Multiple instances with shared backend. Session storage in database or Redis.

Backups: PostgreSQL backend stores chat history and user data. Backup database and uploaded documents. Configuration in environment variables.

LibreChat​

Description: Open-source ChatGPT alternative supporting multiple AI providers with conversation management.

Good for:

  • Self-hosted chat applications
  • Multi-provider AI chat interface
  • Privacy-focused AI interactions
  • Custom AI chat deployments

Commonly used for:

  • Internal ChatGPT replacement
  • Multi-model chat interface
  • AI conversation management
  • Team AI access

High Availability: Multiple instances with shared MongoDB/PostgreSQL backend. Load balancing distributes traffic.

Backups: Database stores conversations and user data. Backup database regularly. Configuration and API keys in environment variables.

Search & Discovery​

Meilisearch​

Description: Fast, typo-tolerant search engine with instant results and simple API, built in Rust.

Good for:

  • Application search functionality
  • E-commerce product search
  • Document search and discovery
  • Auto-complete and suggestions

Commonly used for:

  • Website search bars
  • Product catalogs
  • Documentation search
  • Content discovery

High Availability: Experimental multi-node support. Typically deployed with load balancing and data replication.

Backups: Export search indices via API. Snapshot and dump functionality. Re-indexing from source data possible.

Deployment Recommendations​

For Production:

  • Use High Availability or Enterprise tiers for critical services
  • Configure automated backups with off-site storage
  • Enable monitoring and alerting
  • Use node tags for dedicated hardware (databases on SSD nodes)
  • Plan for disaster recovery with regular backup testing

For Development:

  • Lightweight tier sufficient for most services
  • Local storage acceptable
  • Fewer backups needed
  • Can run multiple services on same nodes

Storage Selection:

  • Databases: OpenEBS for best performance
  • Distributed applications: Longhorn for multi-node availability
  • Backup targets: MinIO or external S3-compatible storage

High Availability Essentials:

  • Run on at least 3 nodes for true HA
  • Use distributed storage or replication
  • Configure automated backups
  • Test failover scenarios
  • Monitor health and resource usage

Getting Started​

Ready to deploy? Navigate to Services → Add Service in the RunOS console and start deploying production-ready open-source services in minutes.