Skip to main content
Amp provides comprehensive management and administration capabilities for operating blockchain data infrastructure. Administrators can manage datasets, control extraction jobs, monitor workers, and configure providers through both the ampctl CLI and the Admin API.

​
Admin Interfaces

Amp offers two complementary interfaces for administration:

ampctl CLI

Command-line tool for interactive administration and scripting

Admin API

RESTful HTTP API for programmatic access and automation

​
Admin API Endpoint

The Admin API is served by the ampd controller process on port 1610: 
# Default endpoint
http://localhost:1610

# Example API calls
curl http://localhost:1610/datasets
curl http://localhost:1610/jobs
curl http://localhost:1610/workers

​
CLI vs API Usage

Both interfaces provide access to the same functionality: 
# Using ampctl CLI
ampctl dataset list
ampctl job list
ampctl worker list

# Equivalent API calls
curl http://localhost:1610/datasets
curl http://localhost:1610/jobs
curl http://localhost:1610/workers
The CLI provides:
  • Human-readable output formatting
  • Interactive workflows
  • Built-in validation and error messages
  • Convenient aliases and shortcuts
The API provides:
  • Programmatic access for automation
  • Integration with monitoring systems
  • Custom tooling and dashboards
  • Direct access from applications

​
Core Administration Operations

Amp administration covers four main areas:

Dataset Management

Register, deploy, and manage blockchain dataset configurations

Job Control

Monitor, start, stop, and manage extraction jobs

Worker Management

Monitor worker health, versions, and availability

Monitoring

Track metrics, health, and performance

​
Dataset Management

Manage the lifecycle of blockchain datasets:
  • Register datasets - Store dataset manifests with version tags
  • Deploy datasets - Schedule extraction jobs to sync blockchain data
  • List datasets - View all registered datasets and versions
  • Inspect datasets - Get detailed configuration and metadata
  • Restore datasets - Recover metadata from object storage

​
Job Control

Control and monitor extraction jobs:
  • List jobs - View active and historical jobs with pagination
  • Inspect jobs - Get detailed job status and configuration
  • Monitor progress - Track sync state and block numbers
  • Stop jobs - Gracefully terminate running jobs
  • Remove jobs - Clean up completed or failed jobs

​
Worker Management

Monitor the distributed worker fleet:
  • List workers - View all registered workers
  • Inspect workers - Get detailed worker information and build version
  • Monitor health - Track heartbeats and worker availability
  • Troubleshoot - Debug worker issues and version mismatches

​
Monitoring & Observability

Track system health and performance:
  • Health checks - Verify system components are operational
  • Metrics - Collect and export operational metrics
  • OpenTelemetry - Integration with observability platforms
  • Progress tracking - Monitor data sync status

​
Quick Start Examples

​
Using ampctl CLI

# List all datasets
ampctl dataset list

# Register a new dataset
ampctl dataset register ethereum/mainnet ./manifest.json --tag 1.0.0

# Deploy a dataset for extraction
ampctl dataset deploy ethereum/mainnet@1.0.0

# Check job status
ampctl job list

# Monitor job progress
ampctl job progress 123

# Check worker health
ampctl worker list

​
Using Admin API

# List all datasets
curl http://localhost:1610/datasets

# Get dataset details
curl http://localhost:1610/datasets/ethereum/mainnet/versions/1.0.0

# Deploy a dataset
curl -X POST http://localhost:1610/datasets/ethereum/mainnet/versions/1.0.0/deploy \
  -H "Content-Type: application/json" \
  -d '{"end_block": null, "parallelism": 1}'

# List active jobs
curl http://localhost:1610/jobs?status=active

# Get job progress
curl http://localhost:1610/jobs/123/progress

# List workers
curl http://localhost:1610/workers

​
JSON Output for Automation

All ampctl commands support --json flag for machine-readable output: 
# Get JSON output
ampctl dataset list --json
ampctl job list --json
ampctl worker list --json

# Pipe to jq for processing
ampctl job list --json | jq '.jobs[] | select(.status == "RUNNING")'
ampctl worker list --json | jq -r '.workers[] | "\(.node_id): \(.heartbeat_at)"'

​
Architecture

Administration flows through the following components: 
Operator → ampctl CLI → Admin API (port 1610) → Controller → Metadata DB
                ↓
         Admin Client Library
Components:
  • ampctl - CLI tool that calls the Admin API
  • Admin API - RESTful HTTP service served by ampd controller
  • Admin Client - Rust library for programmatic API access
  • Controller - Orchestrates operations and manages metadata
  • Metadata DB - PostgreSQL database storing system state

​
Next Steps

Dataset Management

Learn how to register and deploy datasets

Job Management

Control and monitor extraction jobs

Worker Management

Monitor and troubleshoot workers

Monitoring

Set up metrics and observability

Build docs developers (and LLMs) love

Get started for freeTalk to us