Deprecated: nself destroy is a compatibility wrapper. Use nself infra destroy instead. This stub will be removed in v1.0.0.

Command: nself infra destroy (was: nself destroy)

Safe destruction of project infrastructure with configurable scope and comprehensive safety features.

The destroy command provides a controlled way to tear down nself infrastructure with multiple safety mechanisms, selective destruction options, and intelligent cleanup.

• Multi-Level Safety: Double confirmation for data destruction
• Selective Destruction: Target specific resources (containers, volumes, networks, files)
• Dry Run Mode: Preview destruction without executing
• Force Mode: Non-interactive destruction for automation
• Intelligent Cleanup: Removes Docker resources and generated files
• Data Preservation: Option to keep volumes while removing other resources
• Detailed Reporting: Shows exactly what will be destroyed

# Interactive destruction (safest)
nself destroy

# Preview what will be destroyed
nself destroy --dry-run

# Destroy but preserve data volumes
nself destroy --keep-volumes

# Only remove containers
nself destroy --containers-only

# Only remove generated files
nself destroy --generated-only

# Non-interactive (for scripts)
nself destroy --force

nself destroy [OPTIONS]

Docker Resources:

• ✗ All Docker containers (running and stopped)
• ✗ All Docker volumes (databases, files, cache) - DATA LOSS!
• ✗ All Docker networks

Generated Files:

• ✗ docker-compose.yml
• ✗ nginx/ directory
• ✗ services/ directory (custom services)
• ✗ monitoring/ directory (if exists)
• ✗ ssl/ directory (if exists)
• ✗ postgres/ directory (init scripts)
• ✗ .env.runtime

Configuration:

• ✓ .env files (all variants)
• ✓ .env.dev, .env.staging, .env.prod
• ✓ .env.secrets

Code:

• ✓ Source code and custom files
• ✓ Version control (.git/)
• ✓ Node modules and dependencies

nself destroy

Output:

╔════════════════════════════════════════════════════════════════╗
║  nself destroy                                                 ║
║  Safe destruction of project infrastructure                   ║
╚════════════════════════════════════════════════════════════════╝

Destruction Plan:

  ✗ 25 containers (24 running, 1 stopped)
  ✗ 8 volumes (ALL DATA WILL BE LOST)
  ✗ 2 networks
  ✗ Generated files:
      - docker-compose.yml
      - nginx/ directory
      - services/ directory (custom services)
      - monitoring/ directory (if exists)
      - ssl/ directory (if exists)

Preserved:
  ✓ .env files (configuration)
  ✓ Source code and custom files

WARNING: This will permanently delete all data!
This includes:
  - PostgreSQL databases
  - Redis cache data
  - MinIO stored files
  - All other persistent data

Are you sure you want to destroy project 'myapp'? (yes/no): yes

FINAL WARNING: Type the project name 'myapp' to confirm: myapp

→ Beginning destruction sequence...

✓ Removed 25 containers
✓ Removed 8 volumes
✓ Removed 2 networks
✓ Removed 7 generated files/directories

✓ Destruction complete!
✓ Project 'myapp' has been destroyed

Next steps:

  nself init   - Initialize a new project
  nself build  - Rebuild from existing .env

nself destroy --dry-run

Shows exactly what would be destroyed without making any changes. Perfect for:

• Understanding impact before execution
• Verifying selective destruction options
• Automation testing

nself destroy --keep-volumes

Removes containers, networks, and generated files but preserves all data volumes:

• PostgreSQL databases remain intact
• Redis data preserved
• MinIO files preserved
• Can rebuild and restart without data loss

Use case: Clean rebuild while preserving data

nself destroy --containers-only

Only stops and removes Docker containers. Leaves volumes, networks, and files intact.

Use case: Force restart all containers

nself destroy --volumes-only

Only removes Docker volumes. Requires confirmation due to data loss.

Use case: Clean data while keeping configuration

nself destroy --networks-only

Only removes Docker networks. Safe operation with no data loss.

Use case: Fix network conflicts

nself destroy --generated-only

Only removes generated files (docker-compose.yml, nginx/, services/, etc.). No Docker operations.

Use case: Force regeneration of configs via nself build

nself destroy --force

⚠️ DANGEROUS: Skips all confirmations. Use only in automation with proper safeguards.

Production Safety: Never use --force for production environments without manual review.

By default, requires explicit confirmation:

Are you sure you want to destroy project 'myapp'? (yes/no):

When destroying volumes, requires typing the project name:

FINAL WARNING: Type the project name 'myapp' to confirm:

Preview destruction without making changes:

nself destroy --dry-run

• Red (✗): Items to be destroyed
• Yellow (⚠): Data loss warnings
• Green (✓): Preserved items

Use --*-only flags to limit scope:

• Reduces blast radius
• Allows targeted cleanup
• Safer for production

# 1. Destroy infrastructure but keep data
nself destroy --keep-volumes

# 2. Rebuild configuration
nself build

# 3. Start services (data intact)
nself start

# 1. Only remove containers
nself destroy --containers-only

# 2. Restart services
nself start

# 1. Full destruction (interactive)
nself destroy

# 2. Re-initialize
nself init

# 3. Build and start
nself build && nself start

#!/bin/bash
# Automated environment teardown (with safety backup)

# 1. Backup first (safety)
nself backup create

# 2. Destroy (non-interactive)
nself destroy --force

# 3. Clean Docker system (optional)
docker system prune -af --volumes

Symptom: Some containers/volumes remain after destruction

Solutions:

# 1. Force removal manually
docker ps -a | grep myapp | awk '{print $1}' | xargs docker rm -f
docker volume ls | grep myapp | awk '{print $2}' | xargs docker volume rm

# 2. Use Docker system prune
docker system prune -af --volumes

# 3. Check with verbose mode
nself destroy --verbose

Symptom: Cannot remove certain files or volumes

Solutions:

# 1. Check file ownership
ls -la

# 2. Fix permissions
sudo chown -R $USER:$USER .

# 3. Use sudo for Docker operations (if needed)
sudo nself destroy

Symptom: "volume is in use" errors

Solutions:

# 1. Stop all containers first
nself stop

# 2. Remove containers before volumes
nself destroy --containers-only
nself destroy --volumes-only

# 3. Force Docker cleanup
docker container prune -f
docker volume prune -f

# Create backup before destruction
nself backup create
nself destroy

# Preview before executing
nself destroy --dry-run

# If satisfied, execute
nself destroy

# Keep databases during rebuilds
nself destroy --keep-volumes

# ❌ WRONG (production)
nself destroy --force

# ✓ RIGHT (production)
nself destroy  # Interactive with confirmations

# Step-by-step safe rebuild
nself backup create          # Safety backup
nself destroy --keep-volumes # Remove infra, keep data
nself build                  # Regenerate configs
nself start                  # Restart with data intact

# Backup before destruction
nself backup create
nself destroy

# If needed, rollback
nself backup restore <backup-id>

# Force config regeneration
nself destroy --generated-only
nself build

# Graceful vs destructive
nself stop               # Stops services (safe)
nself destroy            # Removes everything (destructive)
nself destroy --keep-volumes  # Middle ground

# Different reset strategies
nself backup reset       # Database only
nself destroy --volumes-only  # All volumes
nself destroy            # Complete teardown

• nself stop - Stop services without removal
• nself backup create - Create backup before destruction
• nself backup restore - Restore from backup after destruction
• nself build - Rebuild configuration
• nself init - Initialize new project

1. Always backup production data:

   nself backup create --env prod

2. Use --keep-volumes for development:

   nself destroy --keep-volumes

3. Never use --force without review:

   # Review first
   nself destroy --dry-run
   # Then execute
   nself destroy

Before running destroy in production:

• Backup created and verified
• Team notified of downtime
• Maintenance mode enabled
• Confirmed correct environment (not production by mistake)
• Reviewed dry-run output
• Rollback plan prepared

#!/bin/bash
set -e

# Backup with error handling
if ! nself backup create; then
  echo "Backup failed - aborting destruction"
  exit 1
fi

# Destroy with confirmation
if nself destroy --dry-run; then
  read -p "Proceed with destruction? (yes/no): " confirm
  if [[ "$confirm" == "yes" ]]; then
    nself destroy --force
  fi
fi

# .github/workflows/teardown.yml
name: Teardown Environment

on:
  workflow_dispatch:
    inputs:
      environment:
        description: 'Environment to destroy'
        required: true
        type: choice
        options:
          - staging
          - dev

jobs:
  teardown:
    runs-on: ubuntu-latest
    steps:
      - name: Backup
        run: nself backup create --env ${{ inputs.environment }}

      - name: Destroy
        run: nself destroy --force --keep-volumes

Q: Can I undo a destroy operation? A: Only if you created a backup first. Use nself backup restore <backup-id> to recover.

Q: What's the difference between stop and destroy? A: stop pauses services (reversible), destroy removes them (requires rebuild).

Q: Is it safe to use destroy in development? A: Yes, with --keep-volumes flag to preserve databases.

Q: Can I destroy only one service? A: No, destroy operates on the entire project. Use docker rm <container> for single services.

Q: What happens to .env files? A: They are always preserved (never deleted by destroy).

Q: Can I recover after accidental destruction? A: Only data in Docker volumes is lost. Rebuild with nself build && nself start. If you had backups, use nself backup restore.

Version: nself v1.0.0+ Command Type: Destructive Safety Level: Interactive (with --force bypass) Related Docs: Backup Guide, Stop Command