Satellite Communication - DeployStack - Enterprise Control Plane for MCP

The DeployStack backend implements satellite management APIs that handle registration, command orchestration, and configuration distribution. The system supports both global satellites (serving all teams) and team satellites (serving specific teams) through a polling-based communication architecture.

Implementation Status

Current Status: Fully implemented and operational The satellite communication system includes:

Satellite Registration: Working registration endpoint with API key generation
Command Orchestration: Complete command polling and result reporting endpoints
Configuration Management: Team-aware MCP server configuration distribution
Status Monitoring: Heartbeat collection with automatic satellite activation
Authentication: Argon2-based API key validation middleware

MCP Server Distribution Architecture

Global Satellite Model: Currently implemented approach where global satellites serve all teams with process isolation. Team-Aware Configuration Distribution:

Global satellites receive ALL team MCP server installations
Each team installation becomes separate per-user processes with unique identifiers
Process ID format: {server_slug}-{team_slug}-{user_slug}-{installation_id}
User-specific configurations (args, environment, headers) merged per user instance

Configuration Merging Process:

Template-level configuration (from MCP server definition)
Team-level configuration (from team installation)
User-level configuration (from user preferences)
Final merged configuration sent to satellite

Multi-Transport Support:

stdio transport: Command and arguments for subprocess execution
http transport: URL and headers for HTTP proxy
sse transport: URL and headers for Server-Sent Events

Satellite Lifecycle Management

Registration Process:

Satellites register with backend and receive API keys
Initial status set to ‘inactive’ for security
API keys stored as Argon2 hashes in database
Satellite URL sent during registration (optional - auto-detected if not provided)

Activation Process:

Satellites send heartbeat after registration
Backend automatically sets status to ‘active’ on first heartbeat
Active satellites begin receiving actual commands

Command Processing:

Inactive satellites receive empty command arrays (no 403 errors)
Active satellites receive pending commands based on priority
Command results reported back to backend for status tracking

Architecture Pattern

Polling-Based Communication

Satellites use outbound-only HTTPS polling to communicate with the backend, making them compatible with restrictive corporate firewalls:

┌─────────────────┐    Outbound HTTPS     ┌─────────────────┐
│   Satellite     │ ───────────────────► │  DeployStack    │
│   (Edge)        │                      │  Backend        │
│                 │ ◄─────────────────── │  (Cloud)        │
└─────────────────┘    Command Response   └─────────────────┘

Communication Channels

The system uses three distinct communication patterns: Command Polling (Backend → Satellite):

Backend creates commands, satellites poll and execute
Adaptive intervals: 2-60 seconds based on priority
Used for: MCP server configuration, process management

Heartbeat (Satellite → Backend, Periodic):

Satellites report status every 30 seconds
Contains: System metrics, process counts, resource usage, satellite URL (first heartbeat only)
Used for: Health monitoring, capacity planning, satellite URL updates

Events (Satellite → Backend, Immediate):

Satellites emit events when actions occur, batched every 3 seconds
Contains: Point-in-time occurrences with precise timestamps
Used for: Real-time UI updates, audit trails, user notifications
See Satellite Events for detailed implementation

Dual Deployment Models

Global Satellites: Cloud-hosted by DeployStack team

Serve all teams with resource isolation
Managed through global satellite management endpoints

Team Satellites: Customer-deployed within corporate networks

Serve specific teams exclusively
Managed through team-scoped satellite management endpoints

Satellite Pairing Process

Security Architecture

The satellite pairing process implements a secure two-step JWT-based authentication system that prevents unauthorized satellite connections. For complete implementation details, see API Security - Registration Token Authentication. Step 1: Token Generation

Administrators generate temporary registration tokens through admin APIs
Scope-specific tokens (global vs team) with cryptographic signatures
Token management endpoints for generation, listing, and revocation

Step 2: Satellite Registration

Satellites authenticate using Authorization: Bearer deploystack_satellite_* headers
Backend validates JWT tokens with single-use consumption
Permanent API keys issued after successful token validation
Token consumed to prevent replay attacks

Note: All new satellite registrations require valid registration tokens. The open registration system has been secured.

Registration Middleware

The secure registration process is implemented through specialized middleware. For technical implementation details, see services/backend/src/middleware/registrationTokenMiddleware.ts. Key Security Features:

JWT signature verification with HMAC-SHA256
Scope validation (global vs team tokens)
Security event logging for failed attempts
Structured error responses with actionable instructions

Command Orchestration

Command Queue Architecture

The backend maintains a priority-based command queue system: Command Types:

spawn: Start new MCP server process
kill: Terminate MCP server process
restart: Restart existing MCP server
configure: Update MCP server configuration
health_check: Request process health status

Priority Levels:

immediate: High-priority commands requiring instant execution
high: Important commands processed within minutes
normal: Standard commands processed during regular polling
low: Background maintenance commands

Command Event Types

Configure commands include an event field in the payload for tracking purposes:

mcp_installation_created - New installation
mcp_installation_updated - Configuration change
mcp_installation_deleted - Installation removed
mcp_recovery - Server recovery

See Satellite Commands for complete event type documentation and usage patterns.

Adaptive Polling Strategy

Satellites adjust polling behavior based on backend signals: Polling Modes:

Immediate Mode: 2-second intervals for urgent commands
Normal Mode: 30-second intervals for standard operations
Backoff Mode: Exponential backoff during errors or low activity

Optimization Features:

Conditional polling based on last poll timestamp
Command batching to reduce API calls
Cache headers for efficient bandwidth usage
Circuit breaker patterns for error recovery

Command Lifecycle

Command Flow:

User action triggers command creation in backend
Command added to priority queue with team context
Satellite polls and retrieves pending commands
Satellite executes command with team isolation
Satellite reports execution results back to backend
Backend updates command status and notifies user interface

Team Context Integration:

All commands include team scope information
Team satellites only receive commands for their team
Global satellites process commands with team isolation
Audit trail with team attribution

Status Monitoring

Heartbeat System

Satellites report health and performance metrics: System Metrics:

CPU usage percentage and memory consumption
Disk usage and network connectivity status
Process count and resource utilization
Uptime and stability indicators
Satellite URL (first heartbeat after startup only) - Updates public URL for satellite discovery

Process Metrics:

Individual MCP server process status
Health indicators (healthy/unhealthy/unknown)
Performance metrics (request count, response times)
Resource consumption per process

Real-Time Status Tracking

The backend provides real-time satellite status information: Satellite Health Monitoring:

Connection status and last heartbeat timestamps
System resource usage trends
Process health aggregation
Alert generation for issues

Performance Analytics:

Historical performance data collection
Usage pattern analysis for capacity planning
Team-specific metrics and reporting
Audit trail generation

Configuration Management

Dynamic Configuration Updates

Satellites retrieve configuration updates without requiring restarts: Configuration Categories:

Polling Settings: Interval configuration and optimization parameters
Resource Limits: CPU, memory, and process count restrictions
Team Settings: Team-specific policies and allowed MCP servers
Security Policies: Access control and compliance requirements

Configuration Distribution:

Push-based updates through command queue
Pull-based configuration refresh during polling
Version-controlled configuration management
Rollback capabilities for configuration errors

Team-Aware Configuration

Configuration respects team boundaries and isolation: Global Satellite Configuration:

Platform-wide settings and resource allocation
Multi-tenant isolation policies
Global resource limits and quotas
Cross-team security boundaries

Team Satellite Configuration:

Team-specific MCP server configurations
Custom resource limits per team
Team-defined security policies
Internal resource access settings

Frontend API Endpoints

The backend provides REST and SSE endpoints for frontend access to installation status, logs, and requests.

Status & Monitoring Endpoints

GET /api/teams/{teamId}/mcp/installations/{installationId}/status

Returns authenticated user’s instance status only (not installation-level)
Each user sees ONLY their own instance status, not other team members’ statuses
Response: { status, status_message, status_updated_at, last_health_check_at }
Used by frontend for real-time status badges and progress indicators

GET /api/teams/{teamId}/mcp/installations/{installationId}/logs

Returns paginated server logs (stderr output, connection errors)
Query params: limit, offset for pagination
Limited to 100 lines per installation (enforced by cleanup cron job)

GET /api/teams/{teamId}/mcp/installations/{installationId}/requests

Returns paginated request logs (tool execution history)
Includes request params, duration, success status
Response data included if request_logging_enabled=true

GET /api/teams/{teamId}/mcp/installations/{installationId}/requests/{requestId}

Returns detailed request log for specific execution
Includes full request/response payloads when available

Settings Management

PATCH /api/teams/{teamId}/mcp/installations/{installationId}/settings

Updates installation settings (stored in mcpServerInstallations.settings jsonb column)
Settings distributed to satellites via config endpoint
Current settings:
- request_logging_enabled (boolean) - Controls capture of tool responses

Real-Time Streaming (SSE)

GET /api/teams/{teamId}/mcp/installations/{installationId}/logs/stream

Server-Sent Events endpoint for real-time log streaming
Frontend subscribes for live stderr output
Auto-reconnects on connection loss

GET /api/teams/{teamId}/mcp/installations/{installationId}/requests/stream

Server-Sent Events endpoint for real-time request log streaming
Frontend subscribes for live tool execution updates
Includes duration, status, and optionally response data

SSE vs REST Comparison:

Feature	REST Endpoints	SSE Endpoints
Use Case	Historical data, pagination	Real-time updates
Connection	Request/response	Persistent connection
Data Flow	Pull (client requests)	Push (server sends)
Frontend Usage	Initial load, manual refresh	Live monitoring

SSE Controller Implementation: services/backend/src/controllers/mcp/sse.controller.ts Routes Implementation: services/backend/src/routes/api/teams/mcp/installations.routes.ts

Health Check & Recovery Systems

Cumulative Health Check System

Purpose: Template-level health aggregation across all installations of an MCP server. McpHealthCheckService (services/backend/src/services/mcp-health-check.service.ts):

Aggregates health status from all installations of each MCP server template
Updates mcpServers.health_status based on installation health
Provides template-level health visibility in admin dashboard

Cron Job: mcp-health-check runs every 3 minutes

Implementation: services/backend/src/jobs/mcp-health-check.job.ts
Checks all MCP server templates
Updates template health status for admin visibility

Credential Validation System

Purpose: Per-installation OAuth token validation to detect expired/revoked credentials. McpCredentialValidationWorker (services/backend/src/workers/mcp-credential-validation.worker.ts):

Validates OAuth tokens for each installation
Sends health_check command to satellite with check_type: 'credential_validation'
Satellite performs OAuth validation and reports status

Cron Job: mcp-credential-validation runs every 1 minute

Implementation: services/backend/src/jobs/mcp-credential-validation.job.ts
Validates installations on 15-minute rotation
Triggers requires_reauth status on validation failure

Health Check Command Payload:

{
  "commandType": "health_check",
  "priority": "immediate",
  "payload": {
    "check_type": "credential_validation",
    "installation_id": "inst_123",
    "team_id": "team_xyz"
  }
}

Satellite validates credentials and emits mcp.server.status_changed with status:

online - Credentials valid
requires_reauth - OAuth token expired/revoked
error - Validation failed with error

Auto-Recovery System

Recovery Trigger:

Health check system detects offline installations
Backend calls notifyMcpRecovery(installation_id, team_id)
Sends command to satellite: Set status=connecting, rediscover tools
Status progression: offline → connecting → discovering_tools → online

Tool Execution Recovery:

Satellite detects recovery during tool execution (offline server responds)
Emits immediate status change event (doesn’t wait for health check)
Triggers asynchronous re-discovery

For satellite-side recovery implementation, see Satellite Recovery System.

Background Cron Jobs

The backend runs three MCP-related cron jobs for maintenance and monitoring: cleanup-mcp-server-logs:

Schedule: Every 10 minutes
Purpose: Enforce 100-line limit per installation in mcpServerLogs table
Action: Deletes oldest logs beyond 100-line limit
Implementation: services/backend/src/jobs/cleanup-mcp-server-logs.job.ts

mcp-health-check:

Schedule: Every 3 minutes
Purpose: Template-level health aggregation
Action: Updates mcpServers.health_status column
Implementation: services/backend/src/jobs/mcp-health-check.job.ts

mcp-credential-validation:

Schedule: Every 1 minute
Purpose: Detect expired/revoked OAuth tokens
Action: Sends health_check commands to satellites
Implementation: services/backend/src/jobs/mcp-credential-validation.job.ts

Database Schema Integration

Core Table Structure

The satellite system integrates with existing DeployStack schema through 5 specialized tables. For detailed schema definitions, see services/backend/src/db/schema.ts. Satellite Registry (satellites):

Central registration of all satellites
Type classification (global/team) and ownership
Capability tracking and status monitoring
API key management and authentication
Satellite URL tracking - Publicly accessible URL updated during registration and first heartbeat

Command Queue (satelliteCommands):

Priority-based command orchestration
Team context and correlation tracking
Expiration and retry management
Command lifecycle tracking

Process Tracking (satelliteProcesses):

Real-time MCP server process monitoring
Health status and performance metrics
Team isolation and resource usage
Integration with existing MCP configuration system

Usage Analytics (satelliteUsageLogs):

Audit trail for compliance
User attribution and team tracking
Performance analytics and billing data
Device tracking for enterprise security

Health Monitoring (satelliteHeartbeats):

System metrics and resource monitoring
Process health aggregation
Alert generation and notification triggers
Historical health trend analysis

Database Integration

Per-User Instance Model: Each team member gets their own instance with independent status tracking. Status exists ONLY at the instance level in mcpServerInstances table. Key Tables:

mcpServerInstances - Per-user instance status tracking
satellites - Satellite registry
satelliteCommands - Command queue
satelliteProcesses - Process tracking
satelliteUsageLogs - Usage analytics
satelliteHeartbeats - Health monitoring

For complete database schema details, see Database Schema. For complete instance lifecycle documentation, see Instance Lifecycle.

Team Isolation in Data Model

All satellite data respects team boundaries: Team-Scoped Data:

Team satellites linked to specific teams
Process isolation per team context
Usage logs with team attribution
Configuration scoped to team access

Global Data with Team Context:

Global satellites serve all teams with isolation
Cross-team usage tracking and analytics
Team-aware resource allocation
Compliance reporting per team

Authentication & Security

Multi-Layer Security Model

Registration Security:

Temporary JWT tokens for initial pairing
Scope validation preventing privilege escalation
Single-use tokens with automatic expiration
Audit trail for security compliance

Operational Security:

Permanent API keys for ongoing communication
Request authentication and authorization
Rate limiting and abuse prevention
IP whitelisting support for team satellites

Team Isolation Security:

Team boundary enforcement
Resource isolation and access control
Cross-team data leakage prevention
Compliance with enterprise security policies

Role-Based Access Control Integration

The satellite system integrates with DeployStack’s existing role framework: global_admin:

Satellite system oversight
Global satellite registration and management
Cross-team analytics and monitoring
System-wide configuration control

team_admin:

Team satellite registration and management
Team-scoped MCP server installation
Team resource monitoring and configuration
Team member access control

team_user:

Satellite-hosted MCP server usage
Team satellite status visibility
Personal usage analytics access

global_user:

Team satellite registration within memberships
Cross-team satellite usage through teams
Limited administrative capabilities

Integration Points

Existing DeployStack Systems

User Management Integration:

Leverages existing authentication and session management
Integrates with current permission and role systems
Uses established user and team membership APIs
Maintains consistency with platform security model

MCP Configuration Integration:

Builds on existing MCP server installation system
Extends current team-based configuration management
Integrates with established credential management
Maintains compatibility with existing MCP workflows

Monitoring Integration:

Uses existing structured logging infrastructure
Integrates with current metrics collection system
Leverages established alerting and notification systems
Maintains consistency with platform observability

Development Implementation

Route Structure

Satellite communication endpoints are organized in services/backend/src/routes/satellites/:

satellites/
├── index.ts              # Route registration
├── register.ts           # Satellite registration endpoint
├── commands.ts           # Command polling and result reporting
├── config.ts             # Configuration distribution
├── heartbeat.ts          # Health monitoring and status updates
└── manage/               # Management endpoints for frontend
    ├── list.ts          # Satellite listing
    └── status.ts        # Satellite status queries

Authentication Middleware

Satellite authentication uses dedicated middleware in services/backend/src/middleware/satelliteAuthMiddleware.ts: Key Features:

Argon2 hash verification for API key validation
Satellite context injection for route handlers
Dual authentication support (user cookies + satellite API keys)
Comprehensive error handling and logging

Usage Pattern:

import { requireSatelliteAuth } from '../../middleware/satelliteAuthMiddleware';

server.get('/satellites/:satelliteId/commands', {
  preValidation: [requireSatelliteAuth()],
  // Route implementation
});

Database Integration

The satellite system uses specialized tables for registry, commands, processes, usage logs, and heartbeats. Per-User Instance Model: mcpServerInstances table tracks status for each user’s instance independently, with CASCADE delete when installation is removed. For complete database schema and relationships, see Database Schema.

Configuration Query Implementation

The configuration endpoint implements complex queries to merge team-specific MCP server configurations: Query Strategy:

Join mcpServerInstallations, mcpServers, and teams tables
Global satellites: Query ALL team installations
Team satellites: Query only specific team installations
JSON field parsing with comprehensive error handling

Configuration Merging Logic:

// Parse template and team configurations
const templateArgs = JSON.parse(installation.template_args || '[]');
const teamArgs = JSON.parse(installation.team_args || '[]');
const templateEnv = JSON.parse(installation.template_env || '{}');
const teamEnv = JSON.parse(installation.team_env || '{}');

// Merge configurations with team overrides
const finalArgs = [...templateArgs, ...teamArgs];
const finalEnv = { ...templateEnv, ...teamEnv };

Error Handling Patterns

Graceful Degradation:

Inactive satellites receive empty command arrays instead of 403 errors
Invalid JSON configurations are skipped with warning logs
Failed satellite authentication returns 401 with structured error messages

Comprehensive Logging:

Structured logging with operation identifiers
Error context preservation for debugging
Performance metrics collection (response times, success rates)

Development Workflow

Local Development Setup:

# Backend setup
cd services/backend
npm install
npm run dev  # Starts on http://localhost:3000

# Satellite setup (separate terminal)
cd services/satellite  
npm install
npm run dev  # Starts on http://localhost:3001

Testing Satellite Communication:

Start backend server
Start satellite (automatically registers)
Monitor logs for successful polling and configuration retrieval
Use database tools to inspect satellite tables and command queue

Database Inspection:

# View registered satellites
psql deploystack
> SELECT id, name, satellite_type, status FROM satellites;

# View MCP server installations
> SELECT installation_name, team_id FROM "mcpServerInstallations";

API Documentation

For detailed API endpoints, request/response formats, and authentication patterns, see the API Specification generated from the backend OpenAPI schema. For detailed satellite architecture and implementation:

Satellite Events - Real-time event processing system
API Security - Security patterns and authorization
Database Management - Schema and data management
OAuth2 Server - OAuth2 implementation details

Basics

Advanced

API

Database

Auth

Satellite

​Implementation Status

​MCP Server Distribution Architecture

​Satellite Lifecycle Management

​Architecture Pattern

​Polling-Based Communication

​Communication Channels

​Dual Deployment Models

​Satellite Pairing Process

​Security Architecture

​Registration Middleware

​Command Orchestration

​Command Queue Architecture

​Command Event Types

​Adaptive Polling Strategy

​Command Lifecycle

​Status Monitoring

​Heartbeat System

​Real-Time Status Tracking

​Configuration Management

​Dynamic Configuration Updates

​Team-Aware Configuration

​Frontend API Endpoints

​Status & Monitoring Endpoints

​Settings Management

​Real-Time Streaming (SSE)

​Health Check & Recovery Systems

​Cumulative Health Check System

​Credential Validation System

​Auto-Recovery System

​Background Cron Jobs

​Database Schema Integration

​Core Table Structure

​Database Integration

​Team Isolation in Data Model

​Authentication & Security

​Multi-Layer Security Model

​Role-Based Access Control Integration

​Integration Points

​Existing DeployStack Systems

​Development Implementation

​Route Structure

​Authentication Middleware

​Database Integration

​Configuration Query Implementation

​Error Handling Patterns

​Development Workflow

​API Documentation

​Related Documentation

Implementation Status

MCP Server Distribution Architecture

Satellite Lifecycle Management

Architecture Pattern

Polling-Based Communication

Communication Channels

Dual Deployment Models

Satellite Pairing Process

Security Architecture

Registration Middleware

Command Orchestration

Command Queue Architecture

Command Event Types

Adaptive Polling Strategy

Command Lifecycle

Status Monitoring

Heartbeat System

Real-Time Status Tracking

Configuration Management

Dynamic Configuration Updates

Team-Aware Configuration

Frontend API Endpoints

Status & Monitoring Endpoints

Settings Management

Real-Time Streaming (SSE)

Health Check & Recovery Systems

Cumulative Health Check System

Credential Validation System

Auto-Recovery System

Background Cron Jobs

Database Schema Integration

Core Table Structure

Database Integration

Team Isolation in Data Model

Authentication & Security

Multi-Layer Security Model

Role-Based Access Control Integration

Integration Points

Existing DeployStack Systems

Development Implementation

Route Structure

Authentication Middleware

Database Integration

Configuration Query Implementation

Error Handling Patterns

Development Workflow

API Documentation

Related Documentation