Merge pull request 'afd' (#978) from afd into main

Reviewed-on: cerc-io/stack-orchestrator#978
This commit is contained in:
AFDudley 2026-01-17 23:21:24 +00:00
commit 55d6c5b495
4 changed files with 1242 additions and 0 deletions

151
AI-FRIENDLY-PLAN.md Normal file
View File

@ -0,0 +1,151 @@
# Plan: Make Stack-Orchestrator AI-Friendly
## Goal
Make the stack-orchestrator repository easier for AI tools (Claude Code, Cursor, Copilot) to understand and use for generating stacks, including adding a `create-stack` command.
---
## Part 1: Documentation & Context Files
### 1.1 Add CLAUDE.md
Create a root-level context file for AI assistants.
**File:** `CLAUDE.md`
Contents:
- Project overview (what stack-orchestrator does)
- Stack creation workflow (step-by-step)
- File naming conventions
- Required vs optional fields in stack.yml
- Common patterns and anti-patterns
- Links to example stacks (simple, medium, complex)
### 1.2 Add JSON Schema for stack.yml
Create formal validation schema.
**File:** `schemas/stack-schema.json`
Benefits:
- AI tools can validate generated stacks
- IDEs provide autocomplete
- CI can catch errors early
### 1.3 Add Template Stack with Comments
Create an annotated template for reference.
**File:** `stack_orchestrator/data/stacks/_template/stack.yml`
```yaml
# Stack definition template - copy this directory to create a new stack
version: "1.2" # Required: 1.0, 1.1, or 1.2
name: my-stack # Required: lowercase, hyphens only
description: "Human-readable description" # Optional
repos: # Git repositories to clone
- github.com/org/repo
containers: # Container images to build (must have matching container-build/)
- cerc/my-container
pods: # Deployment units (must have matching docker-compose-{pod}.yml)
- my-pod
```
### 1.4 Document Validation Rules
Create explicit documentation of constraints currently scattered in code.
**File:** `docs/stack-format.md`
Contents:
- Container names must start with `cerc/`
- Pod names must match compose file: `docker-compose-{pod}.yml`
- Repository format: `host/org/repo[@ref]`
- Stack directory name should match `name` field
- Version field options and differences
---
## Part 2: Add `create-stack` Command
### 2.1 Command Overview
```bash
laconic-so create-stack --repo github.com/org/my-app [--name my-app] [--type webapp]
```
**Behavior:**
1. Parse repo URL to extract app name (if --name not provided)
2. Create `stacks/{name}/stack.yml`
3. Create `container-build/cerc-{name}/Dockerfile` and `build.sh`
4. Create `compose/docker-compose-{name}.yml`
5. Update list files (repository-list.txt, container-image-list.txt, pod-list.txt)
### 2.2 Files to Create
| File | Purpose |
|------|---------|
| `stack_orchestrator/create/__init__.py` | Package init |
| `stack_orchestrator/create/create_stack.py` | Command implementation |
### 2.3 Files to Modify
| File | Change |
|------|--------|
| `stack_orchestrator/main.py` | Add import and `cli.add_command()` |
### 2.4 Command Options
| Option | Required | Description |
|--------|----------|-------------|
| `--repo` | Yes | Git repository URL (e.g., github.com/org/repo) |
| `--name` | No | Stack name (defaults to repo name) |
| `--type` | No | Template type: webapp, service, empty (default: webapp) |
| `--force` | No | Overwrite existing files |
### 2.5 Template Types
| Type | Base Image | Port | Use Case |
|------|------------|------|----------|
| webapp | node:20-bullseye-slim | 3000 | React/Vue/Next.js apps |
| service | python:3.11-slim | 8080 | Python backend services |
| empty | none | none | Custom from scratch |
---
## Part 3: Implementation Summary
### New Files (6)
1. `CLAUDE.md` - AI assistant context
2. `schemas/stack-schema.json` - Validation schema
3. `stack_orchestrator/data/stacks/_template/stack.yml` - Annotated template
4. `docs/stack-format.md` - Stack format documentation
5. `stack_orchestrator/create/__init__.py` - Package init
6. `stack_orchestrator/create/create_stack.py` - Command implementation
### Modified Files (1)
1. `stack_orchestrator/main.py` - Register create-stack command
---
## Verification
```bash
# 1. Command appears in help
laconic-so --help | grep create-stack
# 2. Dry run works
laconic-so --dry-run create-stack --repo github.com/org/test-app
# 3. Creates all expected files
laconic-so create-stack --repo github.com/org/test-app
ls stack_orchestrator/data/stacks/test-app/
ls stack_orchestrator/data/container-build/cerc-test-app/
ls stack_orchestrator/data/compose/docker-compose-test-app.yml
# 4. Build works with generated stack
laconic-so --stack test-app build-containers
```

413
STACK-CREATION-GUIDE.md Normal file
View File

@ -0,0 +1,413 @@
# Implementing `laconic-so create-stack` Command
A plan for adding a new CLI command to scaffold stack files automatically.
---
## Overview
Add a `create-stack` command that generates all required files for a new stack:
```bash
laconic-so create-stack --name my-stack --type webapp
```
**Output:**
```
stack_orchestrator/data/
├── stacks/my-stack/stack.yml
├── container-build/cerc-my-stack/
│ ├── Dockerfile
│ └── build.sh
└── compose/docker-compose-my-stack.yml
Updated: repository-list.txt, container-image-list.txt, pod-list.txt
```
---
## CLI Architecture Summary
### Command Registration Pattern
Commands are Click functions registered in `main.py`:
```python
# main.py (line ~70)
from stack_orchestrator.create import create_stack
cli.add_command(create_stack.command, "create-stack")
```
### Global Options Access
```python
from stack_orchestrator.opts import opts
if not opts.o.quiet:
print("message")
if opts.o.dry_run:
print("(would create files)")
```
### Key Utilities
| Function | Location | Purpose |
|----------|----------|---------|
| `get_yaml()` | `util.py` | YAML parser (ruamel.yaml) |
| `get_stack_path(stack)` | `util.py` | Resolve stack directory path |
| `error_exit(msg)` | `util.py` | Print error and exit(1) |
---
## Files to Create
### 1. Command Module
**`stack_orchestrator/create/__init__.py`**
```python
# Empty file to make this a package
```
**`stack_orchestrator/create/create_stack.py`**
```python
import click
import os
from pathlib import Path
from shutil import copy
from stack_orchestrator.opts import opts
from stack_orchestrator.util import error_exit, get_yaml
# Template types
STACK_TEMPLATES = {
"webapp": {
"description": "Web application with Node.js",
"base_image": "node:20-bullseye-slim",
"port": 3000,
},
"service": {
"description": "Backend service",
"base_image": "python:3.11-slim",
"port": 8080,
},
"empty": {
"description": "Minimal stack with no defaults",
"base_image": None,
"port": None,
},
}
def get_data_dir() -> Path:
"""Get path to stack_orchestrator/data directory"""
return Path(__file__).absolute().parent.parent.joinpath("data")
def validate_stack_name(name: str) -> None:
"""Validate stack name follows conventions"""
import re
if not re.match(r'^[a-z0-9][a-z0-9-]*[a-z0-9]$', name) and len(name) > 2:
error_exit(f"Invalid stack name '{name}'. Use lowercase alphanumeric with hyphens.")
if name.startswith("cerc-"):
error_exit("Stack name should not start with 'cerc-' (container names will add this prefix)")
def create_stack_yml(stack_dir: Path, name: str, template: dict, repo_url: str) -> None:
"""Create stack.yml file"""
config = {
"version": "1.2",
"name": name,
"description": template.get("description", f"Stack: {name}"),
"repos": [repo_url] if repo_url else [],
"containers": [f"cerc/{name}"],
"pods": [name],
}
stack_dir.mkdir(parents=True, exist_ok=True)
with open(stack_dir / "stack.yml", "w") as f:
get_yaml().dump(config, f)
def create_dockerfile(container_dir: Path, name: str, template: dict) -> None:
"""Create Dockerfile"""
base_image = template.get("base_image", "node:20-bullseye-slim")
port = template.get("port", 3000)
dockerfile_content = f'''# Build stage
FROM {base_image} AS builder
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build
# Production stage
FROM {base_image}
WORKDIR /app
COPY package*.json ./
RUN npm ci --only=production
COPY --from=builder /app/dist ./dist
EXPOSE {port}
CMD ["npm", "run", "start"]
'''
container_dir.mkdir(parents=True, exist_ok=True)
with open(container_dir / "Dockerfile", "w") as f:
f.write(dockerfile_content)
def create_build_script(container_dir: Path, name: str) -> None:
"""Create build.sh script"""
build_script = f'''#!/usr/bin/env bash
# Build cerc/{name}
source ${{CERC_CONTAINER_BASE_DIR}}/build-base.sh
SCRIPT_DIR=$( cd -- "$( dirname -- "${{BASH_SOURCE[0]}}" )" &> /dev/null && pwd )
docker build -t cerc/{name}:local \\
-f ${{SCRIPT_DIR}}/Dockerfile \\
${{build_command_args}} \\
${{CERC_REPO_BASE_DIR}}/{name}
'''
build_path = container_dir / "build.sh"
with open(build_path, "w") as f:
f.write(build_script)
# Make executable
os.chmod(build_path, 0o755)
def create_compose_file(compose_dir: Path, name: str, template: dict) -> None:
"""Create docker-compose file"""
port = template.get("port", 3000)
compose_content = {
"version": "3.8",
"services": {
name: {
"image": f"cerc/{name}:local",
"restart": "unless-stopped",
"ports": [f"${{HOST_PORT:-{port}}}:{port}"],
"environment": {
"NODE_ENV": "${NODE_ENV:-production}",
},
}
}
}
with open(compose_dir / f"docker-compose-{name}.yml", "w") as f:
get_yaml().dump(compose_content, f)
def update_list_file(data_dir: Path, filename: str, entry: str) -> None:
"""Add entry to a list file if not already present"""
list_path = data_dir / filename
# Read existing entries
existing = set()
if list_path.exists():
with open(list_path, "r") as f:
existing = set(line.strip() for line in f if line.strip())
# Add new entry
if entry not in existing:
with open(list_path, "a") as f:
f.write(f"{entry}\n")
@click.command()
@click.option("--name", required=True, help="Name of the new stack (lowercase, hyphens)")
@click.option("--type", "stack_type", default="webapp",
type=click.Choice(list(STACK_TEMPLATES.keys())),
help="Stack template type")
@click.option("--repo", help="Git repository URL (e.g., github.com/org/repo)")
@click.option("--force", is_flag=True, help="Overwrite existing files")
@click.pass_context
def command(ctx, name: str, stack_type: str, repo: str, force: bool):
"""Create a new stack with all required files.
Examples:
laconic-so create-stack --name my-app --type webapp
laconic-so create-stack --name my-service --type service --repo github.com/org/repo
"""
# Validate
validate_stack_name(name)
template = STACK_TEMPLATES[stack_type]
data_dir = get_data_dir()
# Define paths
stack_dir = data_dir / "stacks" / name
container_dir = data_dir / "container-build" / f"cerc-{name}"
compose_dir = data_dir / "compose"
# Check for existing files
if not force:
if stack_dir.exists():
error_exit(f"Stack already exists: {stack_dir}\nUse --force to overwrite")
if container_dir.exists():
error_exit(f"Container build dir exists: {container_dir}\nUse --force to overwrite")
# Dry run check
if opts.o.dry_run:
print(f"Would create stack '{name}' with template '{stack_type}':")
print(f" - {stack_dir}/stack.yml")
print(f" - {container_dir}/Dockerfile")
print(f" - {container_dir}/build.sh")
print(f" - {compose_dir}/docker-compose-{name}.yml")
print(f" - Update repository-list.txt")
print(f" - Update container-image-list.txt")
print(f" - Update pod-list.txt")
return
# Create files
if not opts.o.quiet:
print(f"Creating stack '{name}' with template '{stack_type}'...")
create_stack_yml(stack_dir, name, template, repo)
if opts.o.verbose:
print(f" Created {stack_dir}/stack.yml")
create_dockerfile(container_dir, name, template)
if opts.o.verbose:
print(f" Created {container_dir}/Dockerfile")
create_build_script(container_dir, name)
if opts.o.verbose:
print(f" Created {container_dir}/build.sh")
create_compose_file(compose_dir, name, template)
if opts.o.verbose:
print(f" Created {compose_dir}/docker-compose-{name}.yml")
# Update list files
if repo:
update_list_file(data_dir, "repository-list.txt", repo)
if opts.o.verbose:
print(f" Added {repo} to repository-list.txt")
update_list_file(data_dir, "container-image-list.txt", f"cerc/{name}")
if opts.o.verbose:
print(f" Added cerc/{name} to container-image-list.txt")
update_list_file(data_dir, "pod-list.txt", name)
if opts.o.verbose:
print(f" Added {name} to pod-list.txt")
# Summary
if not opts.o.quiet:
print(f"\nStack '{name}' created successfully!")
print(f"\nNext steps:")
print(f" 1. Edit {stack_dir}/stack.yml")
print(f" 2. Customize {container_dir}/Dockerfile")
print(f" 3. Run: laconic-so --stack {name} build-containers")
print(f" 4. Run: laconic-so --stack {name} deploy-system up")
```
### 2. Register Command in main.py
**Edit `stack_orchestrator/main.py`**
Add import:
```python
from stack_orchestrator.create import create_stack
```
Add command registration (after line ~78):
```python
cli.add_command(create_stack.command, "create-stack")
```
---
## Implementation Steps
### Step 1: Create module structure
```bash
mkdir -p stack_orchestrator/create
touch stack_orchestrator/create/__init__.py
```
### Step 2: Create the command file
Create `stack_orchestrator/create/create_stack.py` with the code above.
### Step 3: Register in main.py
Add the import and `cli.add_command()` line.
### Step 4: Test the command
```bash
# Show help
laconic-so create-stack --help
# Dry run
laconic-so --dry-run create-stack --name test-app --type webapp
# Create a stack
laconic-so create-stack --name test-app --type webapp --repo github.com/org/test-app
# Verify
ls -la stack_orchestrator/data/stacks/test-app/
cat stack_orchestrator/data/stacks/test-app/stack.yml
```
---
## Template Types
| Type | Base Image | Port | Use Case |
|------|------------|------|----------|
| `webapp` | node:20-bullseye-slim | 3000 | React/Vue/Next.js apps |
| `service` | python:3.11-slim | 8080 | Python backend services |
| `empty` | none | none | Custom from scratch |
---
## Future Enhancements
1. **Interactive mode** - Prompt for values if not provided
2. **More templates** - Go, Rust, database stacks
3. **Template from existing** - `--from-stack existing-stack`
4. **External stack support** - Create in custom directory
5. **Validation command** - `laconic-so validate-stack --name my-stack`
---
## Files Modified
| File | Change |
|------|--------|
| `stack_orchestrator/create/__init__.py` | New (empty) |
| `stack_orchestrator/create/create_stack.py` | New (command implementation) |
| `stack_orchestrator/main.py` | Add import and `cli.add_command()` |
---
## Verification
```bash
# 1. Command appears in help
laconic-so --help | grep create-stack
# 2. Dry run works
laconic-so --dry-run create-stack --name verify-test --type webapp
# 3. Full creation works
laconic-so create-stack --name verify-test --type webapp
ls stack_orchestrator/data/stacks/verify-test/
ls stack_orchestrator/data/container-build/cerc-verify-test/
ls stack_orchestrator/data/compose/docker-compose-verify-test.yml
# 4. Build works
laconic-so --stack verify-test build-containers
# 5. Cleanup
rm -rf stack_orchestrator/data/stacks/verify-test
rm -rf stack_orchestrator/data/container-build/cerc-verify-test
rm stack_orchestrator/data/compose/docker-compose-verify-test.yml
```

View File

@ -0,0 +1,550 @@
# Docker Compose Deployment Guide
## Introduction
### What is a Deployer?
In stack-orchestrator, a **deployer** provides a uniform interface for orchestrating containerized applications. This guide focuses on Docker Compose deployments, which is the default and recommended deployment mode.
While stack-orchestrator also supports Kubernetes (`k8s`) and Kind (`k8s-kind`) deployments, those are out of scope for this guide. See the [Kubernetes Enhancements](./k8s-deployment-enhancements.md) documentation for advanced deployment options.
## Prerequisites
To deploy stacks using Docker Compose, you need:
- Docker Engine (20.10+)
- Docker Compose plugin (v2.0+)
- Python 3.8+
- stack-orchestrator installed (`laconic-so`)
**That's it!** No additional infrastructure is required. If you have Docker installed, you're ready to deploy.
## Deployment Workflow
The typical deployment workflow consists of four main steps:
1. **Setup repositories and build containers** (first time only)
2. **Initialize deployment specification**
3. **Create deployment directory**
4. **Start and manage services**
## Quick Start Example
Here's a complete example using the built-in `test` stack:
```bash
# Step 1: Setup (first time only)
laconic-so --stack test setup-repositories
laconic-so --stack test build-containers
# Step 2: Initialize deployment spec
laconic-so --stack test deploy init --output test-spec.yml
# Step 3: Create deployment directory
laconic-so --stack test deploy create \
--spec-file test-spec.yml \
--deployment-dir test-deployment
# Step 4: Start services
laconic-so deployment --dir test-deployment start
# View running services
laconic-so deployment --dir test-deployment ps
# View logs
laconic-so deployment --dir test-deployment logs
# Stop services (preserves data)
laconic-so deployment --dir test-deployment stop
```
## Deployment Workflows
Stack-orchestrator supports two deployment workflows:
### 1. Deployment Directory Workflow (Recommended)
This workflow creates a persistent deployment directory that contains all configuration and data.
**When to use:**
- Production deployments
- When you need to preserve configuration
- When you want to manage multiple deployments
- When you need persistent volume data
**Example:**
```bash
# Initialize deployment spec
laconic-so --stack fixturenet-eth deploy init --output eth-spec.yml
# Optionally edit eth-spec.yml to customize configuration
# Create deployment directory
laconic-so --stack fixturenet-eth deploy create \
--spec-file eth-spec.yml \
--deployment-dir my-eth-deployment
# Start the deployment
laconic-so deployment --dir my-eth-deployment start
# Manage the deployment
laconic-so deployment --dir my-eth-deployment ps
laconic-so deployment --dir my-eth-deployment logs
laconic-so deployment --dir my-eth-deployment stop
```
### 2. Quick Deploy Workflow
This workflow deploys directly without creating a persistent deployment directory.
**When to use:**
- Quick testing
- Temporary deployments
- Simple stacks that don't require customization
**Example:**
```bash
# Start the stack directly
laconic-so --stack test deploy up
# Check service status
laconic-so --stack test deploy port test 80
# View logs
laconic-so --stack test deploy logs
# Stop (preserves volumes)
laconic-so --stack test deploy down
# Stop and remove volumes
laconic-so --stack test deploy down --delete-volumes
```
## Real-World Example: Ethereum Fixturenet
Deploy a local Ethereum testnet with Geth and Lighthouse:
```bash
# Setup (first time only)
laconic-so --stack fixturenet-eth setup-repositories
laconic-so --stack fixturenet-eth build-containers
# Initialize with default configuration
laconic-so --stack fixturenet-eth deploy init --output eth-spec.yml
# Create deployment
laconic-so --stack fixturenet-eth deploy create \
--spec-file eth-spec.yml \
--deployment-dir fixturenet-eth-deployment
# Start the network
laconic-so deployment --dir fixturenet-eth-deployment start
# Check status
laconic-so deployment --dir fixturenet-eth-deployment ps
# Access logs from specific service
laconic-so deployment --dir fixturenet-eth-deployment logs fixturenet-eth-geth-1
# Stop the network (preserves blockchain data)
laconic-so deployment --dir fixturenet-eth-deployment stop
# Start again - blockchain data is preserved
laconic-so deployment --dir fixturenet-eth-deployment start
# Clean up everything including data
laconic-so deployment --dir fixturenet-eth-deployment stop --delete-volumes
```
## Configuration
### Passing Configuration Parameters
Configuration can be passed in three ways:
**1. At init time via `--config` flag:**
```bash
laconic-so --stack test deploy init --output spec.yml \
--config PARAM1=value1,PARAM2=value2
```
**2. Edit the spec file after init:**
```bash
# Initialize
laconic-so --stack test deploy init --output spec.yml
# Edit spec.yml
vim spec.yml
```
Example spec.yml:
```yaml
stack: test
config:
PARAM1: value1
PARAM2: value2
```
**3. Docker Compose defaults:**
Environment variables defined in the stack's `docker-compose-*.yml` files are used as defaults. Configuration from the spec file overrides these defaults.
### Port Mapping
By default, services are accessible on randomly assigned host ports. To find the mapped port:
```bash
# Find the host port for container port 80 on service 'webapp'
laconic-so deployment --dir my-deployment port webapp 80
# Output example: 0.0.0.0:32768
```
To configure fixed ports, edit the spec file before creating the deployment:
```yaml
network:
ports:
webapp:
- '8080:80' # Maps host port 8080 to container port 80
api:
- '3000:3000'
```
Then create the deployment:
```bash
laconic-so --stack my-stack deploy create \
--spec-file spec.yml \
--deployment-dir my-deployment
```
### Volume Persistence
Volumes are preserved between stop/start cycles by default:
```bash
# Stop but keep data
laconic-so deployment --dir my-deployment stop
# Start again - data is still there
laconic-so deployment --dir my-deployment start
```
To completely remove all data:
```bash
# Stop and delete all volumes
laconic-so deployment --dir my-deployment stop --delete-volumes
```
Volume data is stored in `<deployment-dir>/data/`.
## Common Operations
### Viewing Logs
```bash
# All services, continuous follow
laconic-so deployment --dir my-deployment logs --follow
# Last 100 lines from all services
laconic-so deployment --dir my-deployment logs --tail 100
# Specific service only
laconic-so deployment --dir my-deployment logs webapp
# Combine options
laconic-so deployment --dir my-deployment logs --tail 50 --follow webapp
```
### Executing Commands in Containers
```bash
# Execute a command in a running service
laconic-so deployment --dir my-deployment exec webapp ls -la
# Interactive shell
laconic-so deployment --dir my-deployment exec webapp /bin/bash
# Run command with specific environment variables
laconic-so deployment --dir my-deployment exec webapp env VAR=value command
```
### Checking Service Status
```bash
# List all running services
laconic-so deployment --dir my-deployment ps
# Check using Docker directly
docker ps
```
### Updating a Running Deployment
If you need to change configuration after deployment:
```bash
# 1. Edit the spec file
vim my-deployment/spec.yml
# 2. Regenerate configuration
laconic-so deployment --dir my-deployment update
# 3. Restart services to apply changes
laconic-so deployment --dir my-deployment stop
laconic-so deployment --dir my-deployment start
```
## Multi-Service Deployments
Many stacks deploy multiple services that work together:
```bash
# Deploy a stack with multiple services
laconic-so --stack laconicd-with-console deploy init --output spec.yml
laconic-so --stack laconicd-with-console deploy create \
--spec-file spec.yml \
--deployment-dir laconicd-deployment
laconic-so deployment --dir laconicd-deployment start
# View all services
laconic-so deployment --dir laconicd-deployment ps
# View logs from specific services
laconic-so deployment --dir laconicd-deployment logs laconicd
laconic-so deployment --dir laconicd-deployment logs console
```
## ConfigMaps
ConfigMaps allow you to mount configuration files into containers:
```bash
# 1. Create the config directory in your deployment
mkdir -p my-deployment/data/my-config
echo "database_url=postgres://localhost" > my-deployment/data/my-config/app.conf
# 2. Reference in spec file
vim my-deployment/spec.yml
```
Add to spec.yml:
```yaml
configmaps:
my-config: ./data/my-config
```
```bash
# 3. Restart to apply
laconic-so deployment --dir my-deployment stop
laconic-so deployment --dir my-deployment start
```
The files will be mounted in the container at `/config/` (or as specified by the stack).
## Deployment Directory Structure
A typical deployment directory contains:
```
my-deployment/
├── compose/
│ └── docker-compose-*.yml # Generated compose files
├── config.env # Environment variables
├── deployment.yml # Deployment metadata
├── spec.yml # Deployment specification
└── data/ # Volume mounts and configs
├── service-data/ # Persistent service data
└── config-maps/ # ConfigMap files
```
## Troubleshooting
### Common Issues
**Problem: "Cannot connect to Docker daemon"**
```bash
# Ensure Docker is running
docker ps
# Start Docker if needed (macOS)
open -a Docker
# Start Docker (Linux)
sudo systemctl start docker
```
**Problem: "Port already in use"**
```bash
# Either stop the conflicting service or use different ports
# Edit spec.yml before creating deployment:
network:
ports:
webapp:
- '8081:80' # Use 8081 instead of 8080
```
**Problem: "Image not found"**
```bash
# Build containers first
laconic-so --stack your-stack build-containers
```
**Problem: Volumes not persisting**
```bash
# Check if you used --delete-volumes when stopping
# Volume data is in: <deployment-dir>/data/
# Don't use --delete-volumes if you want to keep data:
laconic-so deployment --dir my-deployment stop
# Only use --delete-volumes when you want to reset completely:
laconic-so deployment --dir my-deployment stop --delete-volumes
```
**Problem: Services not starting**
```bash
# Check logs for errors
laconic-so deployment --dir my-deployment logs
# Check Docker container status
docker ps -a
# Try stopping and starting again
laconic-so deployment --dir my-deployment stop
laconic-so deployment --dir my-deployment start
```
### Inspecting Deployment State
```bash
# Check deployment directory structure
ls -la my-deployment/
# Check running containers
docker ps
# Check container details
docker inspect <container-name>
# Check networks
docker network ls
# Check volumes
docker volume ls
```
## CLI Commands Reference
### Stack Operations
```bash
# Clone required repositories
laconic-so --stack <name> setup-repositories
# Build container images
laconic-so --stack <name> build-containers
```
### Deployment Initialization
```bash
# Initialize deployment spec with defaults
laconic-so --stack <name> deploy init --output <spec-file>
# Initialize with configuration
laconic-so --stack <name> deploy init --output <spec-file> \
--config PARAM1=value1,PARAM2=value2
```
### Deployment Creation
```bash
# Create deployment directory from spec
laconic-so --stack <name> deploy create \
--spec-file <spec-file> \
--deployment-dir <dir>
```
### Deployment Management
```bash
# Start all services
laconic-so deployment --dir <dir> start
# Stop services (preserves volumes)
laconic-so deployment --dir <dir> stop
# Stop and remove volumes
laconic-so deployment --dir <dir> stop --delete-volumes
# List running services
laconic-so deployment --dir <dir> ps
# View logs
laconic-so deployment --dir <dir> logs [--tail N] [--follow] [service]
# Show mapped port
laconic-so deployment --dir <dir> port <service> <private-port>
# Execute command in service
laconic-so deployment --dir <dir> exec <service> <command>
# Update configuration
laconic-so deployment --dir <dir> update
```
### Quick Deploy Commands
```bash
# Start stack directly
laconic-so --stack <name> deploy up
# Stop stack
laconic-so --stack <name> deploy down [--delete-volumes]
# View logs
laconic-so --stack <name> deploy logs
# Show port mapping
laconic-so --stack <name> deploy port <service> <port>
```
## Related Documentation
- [CLI Reference](./cli.md) - Complete CLI command documentation
- [Adding a New Stack](./adding-a-new-stack.md) - Creating custom stacks
- [Specification](./spec.md) - Internal structure and design
- [Kubernetes Enhancements](./k8s-deployment-enhancements.md) - Advanced K8s deployment options
- [Web App Deployment](./webapp.md) - Deploying web applications
## Examples
For more examples, see the test scripts:
- `scripts/quick-deploy-test.sh` - Quick deployment example
- `tests/deploy/run-deploy-test.sh` - Comprehensive test showing all features
## Summary
- Docker Compose is the default and recommended deployment mode
- Two workflows: deployment directory (recommended) or quick deploy
- The standard workflow is: setup → build → init → create → start
- Configuration is flexible with multiple override layers
- Volume persistence is automatic unless explicitly deleted
- All deployment state is contained in the deployment directory
- For Kubernetes deployments, see separate K8s documentation
You're now ready to deploy stacks using stack-orchestrator with Docker Compose!

View File

@ -0,0 +1,128 @@
# Deploying to the Laconic Network
## Overview
The Laconic network uses a **registry-based deployment model** where everything is published as blockchain records.
## Key Documentation in stack-orchestrator
- `docs/laconicd-with-console.md` - Setting up a laconicd network
- `docs/webapp.md` - Webapp building/running
- `stack_orchestrator/deploy/webapp/` - Implementation (14 modules)
## Core Concepts
### LRN (Laconic Resource Name)
Format: `lrn://laconic/[namespace]/[name]`
Examples:
- `lrn://laconic/deployers/my-deployer-name`
- `lrn://laconic/dns/example.com`
- `lrn://laconic/deployments/example.com`
### Registry Record Types
| Record Type | Purpose |
|-------------|---------|
| `ApplicationRecord` | Published app metadata |
| `WebappDeployer` | Deployment service offering |
| `ApplicationDeploymentRequest` | User's request to deploy |
| `ApplicationDeploymentAuction` | Optional bidding for deployers |
| `ApplicationDeploymentRecord` | Completed deployment result |
## Deployment Workflows
### 1. Direct Deployment
```
User publishes ApplicationDeploymentRequest
→ targets specific WebappDeployer (by LRN)
→ includes payment TX hash
→ Deployer picks up request, builds, deploys, publishes result
```
### 2. Auction-Based Deployment
```
User publishes ApplicationDeploymentAuction
→ Deployers bid (commit/reveal phases)
→ Winner selected
→ User publishes request targeting winner
```
## Key CLI Commands
### Publish a Deployer Service
```bash
laconic-so publish-webapp-deployer --laconic-config config.yml \
--api-url https://deployer-api.example.com \
--name my-deployer \
--payment-address laconic1... \
--minimum-payment 1000alnt
```
### Request Deployment (User Side)
```bash
laconic-so request-webapp-deployment --laconic-config config.yml \
--app lrn://laconic/apps/my-app \
--deployer lrn://laconic/deployers/xyz \
--make-payment auto
```
### Run Deployer Service (Deployer Side)
```bash
laconic-so deploy-webapp-from-registry --laconic-config config.yml --discover
```
## Laconic Config File
All tools require a laconic config file (`laconic.toml`):
```toml
[cosmos]
address_prefix = "laconic"
chain_id = "laconic_9000-1"
endpoint = "http://localhost:26657"
key = "<account-name>"
password = "<account-password>"
```
## Setting Up a Local Laconicd Network
```bash
# Clone and build
laconic-so --stack fixturenet-laconic-loaded setup-repositories
laconic-so --stack fixturenet-laconic-loaded build-containers
laconic-so --stack fixturenet-laconic-loaded deploy create
laconic-so deployment --dir laconic-loaded-deployment start
# Check status
laconic-so deployment --dir laconic-loaded-deployment exec cli "laconic registry status"
```
## Key Implementation Files
| File | Purpose |
|------|---------|
| `publish_webapp_deployer.py` | Register deployment service on network |
| `publish_deployment_auction.py` | Create auction for deployers to bid on |
| `handle_deployment_auction.py` | Monitor and bid on auctions (deployer-side) |
| `request_webapp_deployment.py` | Create deployment request (user-side) |
| `deploy_webapp_from_registry.py` | Process requests and deploy (deployer-side) |
| `request_webapp_undeployment.py` | Request app removal |
| `undeploy_webapp_from_registry.py` | Process removal requests |
| `util.py` | LaconicRegistryClient - all registry interactions |
## Payment System
- **Token Denom**: `alnt` (Laconic network tokens)
- **Payment Options**:
- `--make-payment`: Create new payment with amount (or "auto" for deployer's minimum)
- `--use-payment`: Reference existing payment TX
## What's NOT Well-Documented
1. No end-to-end tutorial for full deployment workflow
2. Stack publishing (vs webapp) process unclear
3. LRN naming conventions not formally specified
4. Payment economics and token mechanics