mirror of https://github.com/jeffcaldwellca/mkcertWeb.git synced 2025-12-29 00:39:35 -06:00

Files

Jeff Caldwell cc6483fcc9 2.0 initial commit

2025-08-08 02:31:25 -04:00

41 KiB

Raw Permalink Blame History

Testing mkcert Web UI v2.0 on Ubuntu

This document provides comprehensive testing procedures for the mkcert Web UI application version 2.0 on Ubuntu systems. This version includes significant security enhancements, modular architecture, and standardized API responses.

What's New in v2.0 Testing

Security Testing Requirements

Command injection protection validation
Path traversal prevention testing
Rate limiting verification across all endpoints
Standardized API response format validation

API Response Format Changes

All API endpoints now return standardized JSON format:

Success: {"success": true, "data": {...}, "message": "optional"}
Error: {"success": false, "error": "description"}

Prerequisites Verification

1. System Requirements Check

# Check if required tools are available
which node nodejs npm wget python3 openssl

# Verify versions
node --version    # Should be 16+
npm --version     # Should be 8+
wget --version    # Built into Ubuntu
openssl version   # Built into Ubuntu

2. Install Missing Prerequisites

# Update package list
sudo apt update

# Install Node.js 18 LTS (if not installed)
sudo apt install -y nodejs npm

# Install additional tools if needed
sudo apt install -y wget curl git libnss3-tools openssl

# Verify installations
node --version && npm --version

3. Install mkcert (Ubuntu-native approach)

# Install dependencies
sudo apt install -y libnss3-tools wget

# Download mkcert (avoiding curl)
wget -O mkcert https://github.com/FiloSottile/mkcert/releases/latest/download/mkcert-v1.4.4-linux-amd64
chmod +x mkcert
sudo mv mkcert /usr/local/bin/

# Verify installation
mkcert -version
which mkcert

Application Setup and Installation Testing

1. Project Setup

# Clone the repository
git clone https://github.com/jeffcaldwellca/mkcertWeb.git
cd mkcertWeb

# Verify project structure
find . -name "*.js" -o -name "*.json" -o -name "*.md" | head -10

# Install dependencies
npm install

# Verify dependencies installed correctly
ls node_modules/ | wc -l  # Should show many packages
npm list --depth=0        # Show direct dependencies

2. Environment Configuration

# Check if .env file exists
ls -la .env

# If .env doesn't exist, copy from example
cp .env.example .env

# View current configuration
cat .env

# Test with authentication enabled (default)
grep "ENABLE_AUTH=true" .env && echo "✓ Authentication enabled"

# Test with authentication disabled (optional)
# sed -i 's/ENABLE_AUTH=true/ENABLE_AUTH=false/' .env

3. Initialize mkcert CA

# Initialize the Certificate Authority
mkcert -install

# Verify CA was created
mkcert -CAROOT
ls -la $(mkcert -CAROOT)

# Should show rootCA.pem and rootCA-key.pem files

4. Start Application

# Start the server
npm start &
SERVER_PID=$!

# Wait for server to start
sleep 3

# Verify server is running
ps aux | grep node
netstat -tlnp | grep :3000

v2.0 API Response Format Testing

1. Health Check Endpoint (Standardized Response)

# Test health endpoint - should return standardized format
wget -qO- http://localhost:3000/api/health | python3 -m json.tool

# Expected v2.0 format:
# {
#   "success": true,
#   "status": "ok",
#   "timestamp": "2025-08-08T...",
#   "uptime": 30.054,
#   "version": "2.0.0"
# }

2. Commands Endpoint (New Standardized Format)

# Test commands endpoint - should return standardized format
wget -qO- http://localhost:3000/api/commands | python3 -m json.tool

# Expected v2.0 format:
# {
#   "success": true,
#   "commands": [
#     {
#       "name": "Install CA",
#       "key": "install-ca",
#       "description": "Install the local CA certificate",
#       "dangerous": false
#     },
#     ...
#   ]
# }

3. Error Response Format Testing

# Test invalid endpoint to verify error format
wget -qO- http://localhost:3000/api/nonexistent 2>/dev/null | python3 -m json.tool

# Expected v2.0 error format:
# {
#   "success": false,
#   "error": "API endpoint not found",
#   "path": "/api/nonexistent",
#   "method": "GET"
# }

4. Security Validation Testing

# Test command injection protection (should fail safely)
wget --post-data='{"command":"invalid; rm -rf /"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/execute \
     -O /tmp/security-test.json 2>/dev/null

cat /tmp/security-test.json | python3 -m json.tool

# Expected security response:
# {
#   "success": false,
#   "error": "Invalid command"
# }

Authentication Testing

1. Authentication Status Testing

# Test authentication status endpoint
wget -qO- http://localhost:3000/api/auth/status | python3 -m json.tool

# Expected output with authentication enabled:
# {
#   "authEnabled": true,
#   "authenticated": false,
#   "username": null
# }

# Test login with correct credentials
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/login-response.json

# Verify successful login
cat /tmp/login-response.json | python3 -m json.tool

# Expected output:
# {
#   "success": true,
#   "message": "Login successful",
#   "redirectTo": "/"
# }

# Test authentication status after login
wget --load-cookies=/tmp/cookies.txt \
     -qO- http://localhost:3000/api/auth/status | python3 -m json.tool

# Expected output after login:
# {
#   "authEnabled": true,
#   "authenticated": true,
#   "username": "admin"
# }

# Test login with incorrect credentials
wget --post-data='{"username":"admin","password":"wrong"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/auth/login \
     -O /tmp/login-fail.json 2>&1

# Verify login failure
cat /tmp/login-fail.json | python3 -m json.tool

# Should contain:
# {
#   "success": false,
#   "error": "Invalid username or password"
# }

# Test with missing credentials
wget --post-data='{"username":"admin"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/auth/login \
     -O /tmp/login-missing.json 2>&1

# Should return 400 error for missing password

4. Logout Testing

# Login first to get valid session
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json

# Test logout
wget --load-cookies=/tmp/cookies.txt \
     --method=POST \
     http://localhost:3000/api/auth/logout \
     -O /tmp/logout-response.json

# Verify successful logout
cat /tmp/logout-response.json | python3 -m json.tool

# Expected output:
# {
#   "success": true,
#   "message": "Logout successful",
#   "redirectTo": "/login"
# }

# Verify session is invalidated
wget --load-cookies=/tmp/cookies.txt \
     -qO- http://localhost:3000/api/auth/status | python3 -m json.tool

# Should show authenticated: false

5. Protected Routes Testing

# Test accessing protected API without authentication
wget http://localhost:3000/api/status -O /tmp/unauth-test.json 2>&1

# Should return 401 Unauthorized when auth is enabled

# Test with valid session
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json

# Now test protected route with authentication
wget --load-cookies=/tmp/cookies.txt \
     -qO- http://localhost:3000/api/status | python3 -m json.tool

# Should work and return system status

# Test login page loads
wget -qO- http://localhost:3000/login | grep -q "<title>" && echo "✓ Login page loads"

# Test redirect to login when not authenticated
wget -qO- http://localhost:3000/ 2>&1 | grep -q "302\|login" && echo "✓ Redirects to login"

# Test redirect to main page when already authenticated
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json

wget --load-cookies=/tmp/cookies.txt \
     http://localhost:3000/login 2>&1 | grep -q "302\|/" && echo "✓ Redirects authenticated users from login"

7. Session Persistence Testing

# Login and save session
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/session-cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json

# Test session persists across requests
for i in {1..3}; do
    echo "Request $i:"
    wget --load-cookies=/tmp/session-cookies.txt \
         -qO- http://localhost:3000/api/auth/status | python3 -c "
import json, sys
data = json.load(sys.stdin)
print('Authenticated:', data.get('authenticated', False))
"
    sleep 1
done

8. Authentication Disabled Testing

# Backup current .env
cp .env .env.backup

# Disable authentication
sed -i 's/ENABLE_AUTH=true/ENABLE_AUTH=false/' .env

# Restart server
kill $SERVER_PID 2>/dev/null
npm start &
SERVER_PID=$!
sleep 3

# Test that authentication is disabled
wget -qO- http://localhost:3000/api/auth/status | python3 -m json.tool

# Expected output:
# {
#   "authEnabled": false
# }

# Test direct access to main page (should work without login)
wget -qO- http://localhost:3000/ | grep -q "<title>" && echo "✓ Main page accessible without auth"

# Test API access without authentication (should work)
wget -qO- http://localhost:3000/api/status | python3 -m json.tool

# Restore authentication settings
cp .env.backup .env

# Restart server with auth enabled
kill $SERVER_PID 2>/dev/null
npm start &
SERVER_PID=$!
sleep 3

OpenID Connect (OIDC) SSO Testing

Note: OIDC testing requires a configured OIDC provider. For development/testing purposes, you can use a public OIDC test provider or set up a local identity server.

1. OIDC Configuration Testing

# Create OIDC test configuration
cat > .env.oidc << 'EOF'
# Copy existing configuration
ENABLE_AUTH=true
ENABLE_OIDC=true

# Test OIDC Configuration (example with a test provider)
OIDC_ISSUER=https://demo.identityserver.io
OIDC_CLIENT_ID=interactive.public
OIDC_CLIENT_SECRET=
OIDC_CALLBACK_URL=http://localhost:3000/auth/oidc/callback
OIDC_SCOPE=openid profile email
EOF

# Backup current configuration
cp .env .env.backup

# Apply OIDC configuration
cp .env.oidc .env

# Restart server with OIDC enabled
kill $SERVER_PID 2>/dev/null
npm start &
SERVER_PID=$!
sleep 5

2. OIDC Provider Discovery Testing

# Test OIDC configuration endpoint
wget -qO- http://localhost:3000/api/auth/status | python3 -m json.tool

# Expected output should include:
# {
#   "authEnabled": true,
#   "oidcEnabled": true,
#   "authenticated": false
# }

# Verify OIDC provider discovery is working
# (Check server logs for successful OIDC provider initialization)
echo "Check server logs for OIDC provider initialization..."

3. OIDC Authentication Flow Testing

# Test OIDC login initiation
# This should redirect to the OIDC provider
wget --max-redirect=0 http://localhost:3000/auth/oidc 2>&1 | grep -q "302\|Location" && echo "✓ OIDC login initiation redirects properly"

# Test OIDC callback endpoint exists
wget --spider http://localhost:3000/auth/oidc/callback 2>&1 | grep -q "200\|404" && echo "✓ OIDC callback endpoint accessible"

# Test that login page includes OIDC option when enabled
wget -qO- http://localhost:3000/login | grep -qi "sso\|oidc\|sign.*with" && echo "✓ Login page includes OIDC option"

# Test login page loads properly with OIDC enabled
wget -qO- http://localhost:3000/login | grep -q "<title>" && echo "✓ Login page loads with OIDC enabled"

5. OIDC Configuration Validation Testing

# Test with missing OIDC configuration
cat > .env.oidc-invalid << 'EOF'
ENABLE_AUTH=true
ENABLE_OIDC=true
# Missing required OIDC settings
OIDC_ISSUER=
OIDC_CLIENT_ID=
OIDC_CLIENT_SECRET=
EOF

cp .env.oidc-invalid .env

# Restart and check that server handles missing config gracefully
kill $SERVER_PID 2>/dev/null
npm start &
SERVER_PID=$!
sleep 3

# Check that app still works with invalid OIDC config
wget -qO- http://localhost:3000/api/auth/status | python3 -c "
import json, sys
try:
    data = json.load(sys.stdin)
    print('✓ Server handles invalid OIDC config gracefully')
    print('Auth enabled:', data.get('authEnabled', False))
    print('OIDC enabled:', data.get('oidcEnabled', False))
except:
    print('✗ Server error with invalid OIDC config')
"

6. Dual Authentication Testing

# Test both basic auth and OIDC enabled simultaneously
cat > .env.dual-auth << 'EOF'
ENABLE_AUTH=true
USERNAME=admin
PASSWORD=admin123
ENABLE_OIDC=true
OIDC_ISSUER=https://demo.identityserver.io
OIDC_CLIENT_ID=interactive.public
OIDC_CLIENT_SECRET=
OIDC_CALLBACK_URL=http://localhost:3000/auth/oidc/callback
OIDC_SCOPE=openid profile email
EOF

cp .env.dual-auth .env

# Restart server
kill $SERVER_PID 2>/dev/null
npm start &
SERVER_PID=$!
sleep 5

# Test that basic auth still works
wget --post-data='{"username":"admin","password":"admin123"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/basic-auth-cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/basic-auth-response.json

cat /tmp/basic-auth-response.json | python3 -m json.tool

# Test that login page shows both options
wget -qO- http://localhost:3000/login | grep -qi "username\|password" && echo "✓ Basic auth form present"
wget -qO- http://localhost:3000/login | grep -qi "sso\|oidc\|sign.*with" && echo "✓ OIDC option present"

7. OIDC Cleanup Testing

# Restore original configuration
cp .env.backup .env

# Restart server with original settings
kill $SERVER_PID 2>/dev/null
npm start &
SERVER_PID=$!
sleep 3

# Verify basic auth still works
wget -qO- http://localhost:3000/api/auth/status | python3 -m json.tool

# Clean up test files
rm -f .env.oidc .env.oidc-invalid .env.dual-auth .env.backup
rm -f /tmp/basic-auth-*.json /tmp/basic-auth-cookies.txt

echo "✓ OIDC testing completed and cleaned up"

Manual OIDC Testing Notes:

Complete OIDC authentication flow requires manual browser testing with a real OIDC provider
Test with Azure AD, Google, or other OIDC providers in your organization
Verify that user profile information is correctly extracted from OIDC tokens
Test OIDC logout and session management
Verify OIDC token refresh if implemented

Functional Testing Using Built-in Tools

1. System Status API Testing

# Login first if authentication is enabled
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/auth-cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json 2>/dev/null

# Test system status endpoint
wget --load-cookies=/tmp/auth-cookies.txt \
     -qO- http://localhost:3000/api/status | python3 -m json.tool

# Expected output should include:
# - "success": true
# - "mkcertInstalled": true  
# - "caExists": true
# - "caRoot": "/home/username/.local/share/mkcert"

2. Root CA Information Testing

# Test CA info endpoint (using session from previous test)
wget --load-cookies=/tmp/auth-cookies.txt \
     -qO- http://localhost:3000/api/rootca/info | python3 -m json.tool

# Expected output should include:
# - Certificate subject, issuer, serial
# - Valid from/to dates
# - SHA256 fingerprint
# - Days until expiry

3. Certificate Generation Testing

PEM Format Certificate

# Generate PEM format certificate using wget (with authentication)
wget --load-cookies=/tmp/auth-cookies.txt \
     --post-data='{"domains":["localhost","127.0.0.1","test.local"],"format":"pem"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/generate \
     -O /tmp/pem-response.json

# Verify response
cat /tmp/pem-response.json | python3 -m json.tool

# Expected output:
# - "success": true
# - "certFile": filename ending in .pem
# - "keyFile": filename ending in -key.pem
# - "folder": date-based folder path

CRT Format Certificate

# Generate CRT format certificate (with authentication)
wget --load-cookies=/tmp/auth-cookies.txt \
     --post-data='{"domains":["example.local","api.example.local"],"format":"crt"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/generate \
     -O /tmp/crt-response.json

# Verify response
cat /tmp/crt-response.json | python3 -m json.tool

# Expected output:
# - "success": true
# - "certFile": filename ending in .crt
# - "keyFile": filename ending in .key

4. Certificate Listing Testing

# List all certificates (with authentication)
wget --load-cookies=/tmp/auth-cookies.txt \
     -qO- http://localhost:3000/api/certificates | python3 -m json.tool > /tmp/certificates.json

# Verify certificate list
cat /tmp/certificates.json

# Check that both generated certificates appear
grep -c "localhost" /tmp/certificates.json     # Should be > 0
grep -c "example.local" /tmp/certificates.json  # Should be > 0

5. File Download Testing

# Create temporary directory for downloads
mkdir -p /tmp/mkcert-downloads
cd /tmp/mkcert-downloads

# Download root CA certificate (with authentication)
wget --load-cookies=/tmp/auth-cookies.txt \
     http://localhost:3000/api/download/rootca -O mkcert-rootCA.pem

# Verify it's a valid certificate
openssl x509 -in mkcert-rootCA.pem -text -noout | head -20

# Download certificate bundle (extract folder/filename from previous listing)
# Example folder format: 2025-07-25_2025-07-25T10-30-45_localhost_127-0-0-1_test-local
FOLDER=$(cat /tmp/certificates.json | python3 -c "
import json, sys
data = json.load(sys.stdin)
if data['certificates']:
    folder = data['certificates'][0]['folder'].replace('/', '_')
    name = data['certificates'][0]['name']
    print(f'{folder}')
")

CERT_NAME=$(cat /tmp/certificates.json | python3 -c "
import json, sys
data = json.load(sys.stdin)
if data['certificates']:
    print(data['certificates'][0]['name'])
")

if [ ! -z "$FOLDER" ] && [ ! -z "$CERT_NAME" ]; then
    # Download certificate bundle (with authentication)
    wget --load-cookies=/tmp/auth-cookies.txt \
         "http://localhost:3000/api/download/bundle/${FOLDER}/${CERT_NAME}" -O certificate-bundle.zip
    
    # Verify ZIP file
    unzip -l certificate-bundle.zip
    unzip certificate-bundle.zip
    
    # Verify certificate files
    ls -la *.pem *.crt *.key 2>/dev/null || echo "Certificate files extracted"
fi

File System Testing

1. Certificate Storage Structure

# Navigate back to project directory
cd - 

# Verify certificate organization
find certificates/ -type f -name "*.pem" -o -name "*.crt" -o -name "*.key" | head -10

# Check folder structure
find certificates/ -type d | sort

# Expected structure:
# certificates/
# certificates/YYYY-MM-DD/
# certificates/YYYY-MM-DD/YYYY-MM-DDTHH-MM-SS_domainnames/

2. File Permissions Testing

# Check file permissions
ls -la certificates/
find certificates/ -name "*.key" -exec ls -la {} \;

# Private keys should be readable by owner only (600 or similar)
# Certificates can be more permissive (644)

3. Root Certificate Protection Testing

# Try to delete a certificate from root (should fail)
if [ -d "certificates/root" ]; then
    # This should return 403 Forbidden
    wget --load-cookies=/tmp/auth-cookies.txt \
         --method=DELETE http://localhost:3000/api/certificates/root/test-cert \
         -O /tmp/delete-response.txt 2>&1
    cat /tmp/delete-response.txt
    grep -q "403" /tmp/delete-response.txt && echo "✓ Root protection working"
fi

# Try to delete a subfolder certificate (should succeed if any exist)
if [ ! -z "$FOLDER" ] && [ ! -z "$CERT_NAME" ]; then
    wget --load-cookies=/tmp/auth-cookies.txt \
         --method=DELETE "http://localhost:3000/api/certificates/${FOLDER}/${CERT_NAME}" \
         -O /tmp/delete-response2.txt 2>&1
    cat /tmp/delete-response2.txt
fi

Security Testing

1. Authentication Security Testing

# Test session timeout and security
# Login and get session
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/security-cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json

# Test session hijacking protection (cookies should be httpOnly)
grep -q "HttpOnly" /tmp/security-cookies.txt && echo "✓ HttpOnly cookies enabled"

# Test HTTPS session security (cookies should be secure in production)
# This would need HTTPS testing environment

# Test CSRF protection by attempting requests without proper session
wget --post-data='{"domains":["malicious.test"],"format":"pem"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/generate \
     -O /tmp/csrf-test.json 2>&1

# Should fail without proper authentication
grep -q "401\|403" /tmp/csrf-test.json && echo "✓ CSRF protection working"

2. Input Validation Testing

# Test invalid domain names (with authentication)
wget --load-cookies=/tmp/auth-cookies.txt \
     --post-data='{"domains":[],"format":"pem"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/generate \
     -O /tmp/invalid-empty.json 2>&1

grep -q "400" /tmp/invalid-empty.json && echo "✓ Empty domains rejected"

# Test invalid format (with authentication)
wget --load-cookies=/tmp/auth-cookies.txt \
     --post-data='{"domains":["test.local"],"format":"invalid"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/generate \
     -O /tmp/invalid-format.json 2>&1

grep -q "400" /tmp/invalid-format.json && echo "✓ Invalid format rejected"

# Test SQL injection attempts in authentication
wget --post-data='{"username":"admin'\'' OR 1=1--","password":"any"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/auth/login \
     -O /tmp/sql-injection.json 2>&1

grep -q "401\|400" /tmp/sql-injection.json && echo "✓ SQL injection protection working"

# Test XSS attempts in domain names
wget --load-cookies=/tmp/auth-cookies.txt \
     --post-data='{"domains":["<script>alert(\"xss\")</script>.test"],"format":"pem"}' \
     --header='Content-Type: application/json' \
     http://localhost:3000/api/generate \
     -O /tmp/xss-test.json 2>&1

# Should either reject or sanitize the input

3. Rate Limiting Security Testing

# Test authentication rate limiting (brute force protection)
echo "Testing authentication rate limiting..."

# Attempt rapid login attempts (should trigger rate limit)
echo "Testing auth rate limiting with invalid credentials..."
for i in {1..7}; do
    echo "Auth attempt $i:"
    wget --post-data='{"username":"admin","password":"wrong"}' \
         --header='Content-Type: application/json' \
         http://localhost:3000/api/auth/login \
         -O /tmp/auth-rate-test-$i.json 2>&1
    
    if grep -q "429\|Too many.*authentication" /tmp/auth-rate-test-$i.json; then
        echo "✓ Auth rate limit triggered at attempt $i"
        break
    elif [ $i -gt 5 ]; then
        echo "⚠ Auth rate limit may not be working (completed $i attempts)"
    fi
    sleep 1
done

# Test CLI rate limiting for certificate generation
echo "Testing CLI rate limiting..."

# Login with correct credentials for CLI testing
wget --post-data='{"username":"admin","password":"admin"}' \
     --header='Content-Type: application/json' \
     --save-cookies=/tmp/rate-limit-cookies.txt \
     http://localhost:3000/api/auth/login \
     -O /tmp/temp.json

# Test rapid certificate generation (should hit rate limit)
echo "Attempting rapid certificate generation to test CLI rate limiting..."
for i in {1..12}; do
    echo "Request $i:"
    wget --load-cookies=/tmp/rate-limit-cookies.txt \
         --post-data="{\"domains\":[\"test$i.local\"],\"format\":\"pem\"}" \
         --header='Content-Type: application/json' \
         http://localhost:3000/api/generate \
         -O /tmp/rate-limit-test-$i.json 2>&1
    
    if grep -q "429\|Too many.*CLI" /tmp/rate-limit-test-$i.json; then
        echo "✓ CLI rate limit triggered at request $i"
        break
    elif [ $i -gt 10 ]; then
        echo "⚠ CLI rate limit may not be working (completed $i requests)"
    fi
    sleep 1
done

# Test API rate limiting for general endpoints
echo "Testing general API rate limiting..."
for i in {1..105}; do
    wget --load-cookies=/tmp/rate-limit-cookies.txt \
         -qO- http://localhost:3000/api/certificates > /tmp/api-rate-$i.json 2>&1
    
    if grep -q "429\|Too many.*API" /tmp/api-rate-$i.json; then
        echo "✓ API rate limit working (triggered at request $i)"
        break
    fi
    
    if [ $((i % 25)) -eq 0 ]; then
        echo "Completed $i API requests..."
    fi
done

# Test rate limit configuration via environment variables
echo "Testing rate limit environment variable configuration..."
cat > .env.rate-test << 'EOF'
ENABLE_AUTH=true
AUTH_USERNAME=admin
AUTH_PASSWORD=admin
CLI_RATE_LIMIT_WINDOW=900000
CLI_RATE_LIMIT_MAX=5
API_RATE_LIMIT_WINDOW=900000
API_RATE_LIMIT_MAX=50
AUTH_RATE_LIMIT_WINDOW=900000
AUTH_RATE_LIMIT_MAX=3
EOF

echo "✓ Rate limiting configuration test completed"

# Test rate limit headers
echo "Testing rate limit headers..."
wget --load-cookies=/tmp/rate-limit-cookies.txt \
     --server-response \
     http://localhost:3000/api/status \
     -O /tmp/rate-headers.txt 2>&1

grep -E "X-RateLimit|RateLimit" /tmp/rate-headers.txt && echo "✓ Rate limit headers present"

echo "✓ Rate limiting tests completed"

# Clean up rate limiting test files
rm -f /tmp/rate-limit-*.json /tmp/api-rate-*.json /tmp/auth-rate-test-*.json /tmp/rate-headers.txt


### 4. File Access Security Testing
```bash
# Try to access files outside certificate directory (should fail)
wget --load-cookies=/tmp/auth-cookies.txt \
     http://localhost:3000/api/download/cert/root/../../../etc/passwd \
     -O /tmp/security-test.txt 2>&1

grep -q "404\|400" /tmp/security-test.txt && echo "✓ Path traversal protection working"

# Test unauthorized file access without authentication
wget http://localhost:3000/api/download/rootca \
     -O /tmp/unauth-download.txt 2>&1

grep -q "401" /tmp/unauth-download.txt && echo "✓ File download requires authentication"

5. Session Management Security Testing

# Test concurrent sessions
# Login from multiple "clients"
for i in {1..3}; do
    wget --post-data='{"username":"admin","password":"admin"}' \
         --header='Content-Type: application/json' \
         --save-cookies="/tmp/session-${i}.txt" \
         http://localhost:3000/api/auth/login \
         -O "/tmp/login-${i}.json" 2>/dev/null
done

# Test that all sessions work independently
for i in {1..3}; do
    STATUS=$(wget --load-cookies="/tmp/session-${i}.txt" \
                  -qO- http://localhost:3000/api/auth/status | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(data.get('authenticated', False))
")
    echo "Session $i authenticated: $STATUS"
done

# Test session invalidation after logout
wget --load-cookies="/tmp/session-1.txt" \
     --method=POST \
     http://localhost:3000/api/auth/logout \
     -O /tmp/logout-test.json

# Verify session is invalidated
STATUS=$(wget --load-cookies="/tmp/session-1.txt" \
              -qO- http://localhost:3000/api/auth/status | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(data.get('authenticated', False))
")
echo "Session after logout: $STATUS"
[ "$STATUS" = "False" ] && echo "✓ Session properly invalidated"

Performance Testing

1. Concurrent Certificate Generation

# Generate multiple certificates simultaneously (with authentication)
for i in {1..5}; do
    wget --load-cookies=/tmp/auth-cookies.txt \
         --post-data="{\"domains\":[\"test${i}.local\",\"api${i}.local\"],\"format\":\"pem\"}" \
         --header='Content-Type: application/json' \
         http://localhost:3000/api/generate \
         -O "/tmp/concurrent-${i}.json" &
done

# Wait for all to complete
wait

# Check all succeeded
for i in {1..5}; do
    if grep -q "success.*true" "/tmp/concurrent-${i}.json"; then
        echo "✓ Concurrent test $i passed"
    else
        echo "✗ Concurrent test $i failed"
    fi
done

2. Large Certificate List Testing

# Get certificate count (with authentication)
CERT_COUNT=$(wget --load-cookies=/tmp/auth-cookies.txt \
                  -qO- http://localhost:3000/api/certificates | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(len(data.get('certificates', [])))
")

echo "Certificate count: $CERT_COUNT"

# Measure response time for certificate listing
time wget --load-cookies=/tmp/auth-cookies.txt \
          -qO- http://localhost:3000/api/certificates > /dev/null

Browser Integration Testing

1. Web Interface Testing

# Test main page loads (should redirect to login when auth is enabled)
wget -qO- http://localhost:3000/ 2>&1 | grep -q "302\|login" && echo "✓ Redirects to login"

# Test login page loads
wget -qO- http://localhost:3000/login | grep -q "<title>" && echo "✓ Login page loads"

# Test authenticated access to main page
wget --load-cookies=/tmp/auth-cookies.txt \
     -qO- http://localhost:3000/ | grep -q "<title>" && echo "✓ Main page loads when authenticated"

# Test static assets
wget -qO- http://localhost:3000/styles.css | grep -q "body" && echo "✓ CSS loads"
wget -qO- http://localhost:3000/script.js | grep -q "function" && echo "✓ JavaScript loads"

2. Authentication Flow Testing (Manual Browser Testing)

echo "=== Manual Browser Testing Instructions ==="
echo "1. Open browser to http://localhost:3000"
echo "2. Should redirect to login page"
echo "3. Try invalid credentials - should show error"
echo "4. Login with admin/admin - should redirect to main page"
echo "5. Check that logout button appears"
echo "6. Test logout - should redirect back to login"
echo "7. Try accessing http://localhost:3000 again - should redirect to login"
echo "8. Test that browser back button doesn't bypass authentication"

3. Certificate Trust Testing (if desktop environment available)

# If running on desktop Ubuntu, test certificate trust
if command -v firefox &> /dev/null; then
    echo "Firefox detected - manual testing recommended:"
    echo "1. Generate a certificate for localhost"
    echo "2. Set up a test HTTPS server with the certificate"
    echo "3. Navigate to https://localhost"
    echo "4. Verify no security warnings appear"
    echo "5. Check certificate details show mkcert as issuer"
fi

Error Handling Testing

1. Service Unavailable Testing

# Stop the server
kill $SERVER_PID 2>/dev/null || echo "Server already stopped"

# Test connection to stopped server
wget http://localhost:3000/api/status -O /tmp/stopped-test.txt 2>&1 || echo "✓ Connection refused when server stopped"

# Restart server for cleanup
npm start &
SERVER_PID=$!
sleep 3

2. Missing mkcert Testing

# Temporarily rename mkcert to simulate missing installation
sudo mv /usr/local/bin/mkcert /usr/local/bin/mkcert.backup 2>/dev/null || echo "mkcert not in /usr/local/bin"

# Test API response
wget -qO- http://localhost:3000/api/status | python3 -c "
import json, sys
data = json.load(sys.stdin)
if not data.get('mkcertInstalled', True):
    print('✓ Missing mkcert detected correctly')
else:
    print('✗ Missing mkcert not detected')
"

# Restore mkcert
sudo mv /usr/local/bin/mkcert.backup /usr/local/bin/mkcert 2>/dev/null || echo "mkcert restored"

Cleanup and Validation

1. Test Cleanup

# Stop the server
kill $SERVER_PID 2>/dev/null

# Clean up test files
rm -rf /tmp/mkcert-downloads /tmp/*.json /tmp/*.txt

# Optionally clean up generated certificates
# rm -rf certificates/$(date +%Y-%m-%d)

echo "✓ Test cleanup completed"

2. Final Validation Report

echo "=== mkcert Web UI Test Summary ==="
echo "Date: $(date)"
echo "Ubuntu Version: $(lsb_release -d | cut -f2)"
echo "Node.js Version: $(node --version)"
echo "mkcert Version: $(mkcert -version 2>/dev/null || echo 'Not found')"
echo "Certificate Count: $(find certificates/ -name "*.pem" -o -name "*.crt" | wc -l)"
echo "Server Status: $(wget -q --spider http://localhost:3000 && echo 'Running' || echo 'Stopped')"
echo "=================================="

Automated Test Script

Create an automated test runner:

#!/bin/bash
# save as test-runner.sh
set -e

echo "Starting automated mkcert Web UI tests..."

# Source all the test commands above
# This script can be expanded to include all tests in sequence

echo "✓ All tests completed successfully!"

Troubleshooting Common Issues

Authentication Issues

# Check authentication configuration
grep "ENABLE_AUTH" .env
grep "AUTH_USERNAME" .env
grep "AUTH_PASSWORD" .env

# Reset authentication settings
cp .env.example .env
# Edit .env with your preferred settings

# Clear browser cookies if having session issues
# Or delete cookie files: rm /tmp/*cookies*.txt

Permission Issues

# Fix certificate directory permissions
chmod 755 certificates/
find certificates/ -type f -name "*.key" -exec chmod 600 {} \;
find certificates/ -type f -name "*.pem" -o -name "*.crt" -exec chmod 644 {} \;

Port Conflicts

# Check what's using port 3000
sudo netstat -tlnp | grep :3000

# Use alternative port
PORT=3001 npm start

Session Issues

# Check session configuration in development
# Session cookies should work over HTTP in development
# For production, ensure HTTPS is enabled for secure cookies

# Clear existing sessions
rm /tmp/*cookies*.txt

# Restart server to reset all sessions
kill $SERVER_PID
npm start &
SERVER_PID=$!

mkcert Issues

# Reinstall CA if needed
mkcert -uninstall
mkcert -install

# Check CA location and permissions
ls -la $(mkcert -CAROOT)

This comprehensive testing suite ensures the mkcert Web UI works correctly on Ubuntu systems using only built-in tools and package manager installations. mkcert -install

Verify CA installation

mkcert -CAROOT ls -la $(mkcert -CAROOT)


### 5. Start the Web UI
```bash
# Start the server
npm start

# Or for development with auto-restart
npm run dev

6. Access the Web Interface

Open your browser and go to: http://localhost:3000

7. Test Certificate Generation

Via Web Interface:

Open http://localhost:3000 in your browser
Check the system status (should show mkcert installed and CA installed)

Enter domains in the text area (one per line):

localhost
127.0.0.1
*.local.dev
test.example.com

Choose certificate format:
- PEM: Standard format (.pem, -key.pem) - good for most applications
- CRT/KEY: Web server format (.crt, .key) - common for Apache, Nginx, etc.
Click "Generate Certificate"
Download the generated certificates

Via API (using curl):

# Generate a PEM certificate via API
curl -X POST http://localhost:3000/api/generate \
  -H "Content-Type: application/json" \
  -d '{"domains": ["localhost", "127.0.0.1", "*.local.dev"], "format": "pem"}'

# Generate a CRT/KEY certificate via API  
curl -X POST http://localhost:3000/api/generate \
  -H "Content-Type: application/json" \
  -d '{"domains": ["web.local"], "format": "crt"}'

# List all certificates
curl http://localhost:3000/api/certificates

# Download a certificate (PEM format example)
curl -O http://localhost:3000/api/download/cert/localhost_127-0-0-1_local-dev.pem

# Download a certificate (CRT format example)  
curl -O http://localhost:3000/api/download/cert/web_local.crt

# Download a key file (PEM format example)
curl -O http://localhost:3000/api/download/key/localhost_127-0-0-1_local-dev-key.pem

# Download a key file (CRT format example)
curl -O http://localhost:3000/api/download/key/web_local.key

# Download bundle as ZIP
curl -O http://localhost:3000/api/download/bundle/localhost_127-0-0-1_local-dev

8. Verify Generated Certificates

# Check certificates directory
ls -la certificates/

# Verify certificate details and expiry (works for both .pem and .crt)
openssl x509 -in certificates/your-cert-file.pem -text -noout
openssl x509 -in certificates/your-cert-file.crt -text -noout

# Check certificate expiry specifically
openssl x509 -in certificates/your-cert-file.pem -noout -enddate
openssl x509 -in certificates/your-cert-file.crt -noout -enddate

# Check certificate domains
openssl x509 -in certificates/your-cert-file.pem -noout -text | grep -A1 "Subject Alternative Name"

# Test the certificate with a simple HTTPS server
# (Optional - requires additional setup)

9. Test Certificate Management and Expiry Display

View certificates: Check the web interface for the list
Verify format badges: Each certificate should show PEM or CRT/KEY format badge
Verify expiry information: Each certificate should show:
- Days until expiry with color coding:
  - Green: More than 30 days
  - Yellow/Orange: 7-30 days
  - Red: Less than 7 days or expired
- Exact expiry date
- Domain names covered by the certificate
Test both formats: Generate certificates in both PEM and CRT/KEY formats
Download individual files: Click download buttons for both formats
Download bundles: Test ZIP download functionality
Delete certificates: Use the delete button and verify files are removed

Troubleshooting on Ubuntu

Common Issues and Solutions:

mkcert command not found

# Add to PATH if needed
echo 'export PATH=$PATH:/usr/local/bin' >> ~/.bashrc
source ~/.bashrc

Permission denied errors

# Make sure user has write permissions
chmod 755 /path/to/mkcertWeb
chmod 755 /path/to/mkcertWeb/certificates

Port 3000 already in use

# Use a different port
PORT=3001 npm start

CA installation fails

# Install NSS tools for browser support
sudo apt install libnss3-tools
mkcert -install

Node.js version too old

# Update Node.js
sudo npm install -g n
sudo n stable

Testing Checklist

mkcert is installed and accessible
Node.js and npm are installed
Project dependencies are installed
Environment configuration (.env) is properly set
mkcert CA is installed (mkcert -install)
Web server starts without errors
Authentication Testing:
- Login page loads correctly
- Can login with correct credentials (admin/admin)
- Login fails with incorrect credentials
- Session persists across requests
- Logout invalidates session
- Protected routes require authentication
- Unauthenticated requests are redirected to login
- Session cookies are secure (HttpOnly)
Web interface loads at http://localhost:3000 (when authenticated)
System status shows green checkmarks
Can generate certificates for multiple domains
Can generate certificates in PEM format (.pem, -key.pem)
Can generate certificates in CRT/KEY format (.crt, .key)
Certificate format badges display correctly
Certificate expiry dates are displayed correctly
Expiry status shows proper color coding (green/yellow/red)
Domain names are extracted and displayed
Can download individual certificate files (both formats)
Can download certificate bundles (ZIP)
Can delete certificates (both formats)
API endpoints respond correctly (with authentication)
Expired certificates are clearly marked
Security Testing:
- Path traversal protection works
- Input validation prevents malicious input
- File downloads require authentication
- Sessions are properly managed
- Authentication can be disabled in development

Expected File Structure After Testing

mkcertWeb/
├── certificates/           # Generated certificates will appear here
│   ├── localhost.pem       # PEM format certificate
│   ├── localhost-key.pem   # PEM format key
│   ├── web_local.crt       # CRT format certificate  
│   ├── web_local.key       # CRT format key
│   ├── test_example_com.pem
│   └── test_example_com-key.pem
├── node_modules/          # Dependencies
├── public/                # Web interface files
├── server.js              # Main server
└── package.json           # Project configuration

Performance Testing

# Test with multiple concurrent requests
for i in {1..5}; do
  curl -X POST http://localhost:3000/api/generate \
    -H "Content-Type: application/json" \
    -d "{\"domains\": [\"test$i.local\"]}" &
done
wait

This should give you a comprehensive way to test the mkcert Web UI on your Ubuntu system!

41 KiB Raw Permalink Blame History

Testing mkcert Web UI v2.0 on Ubuntu

What's New in v2.0 Testing

Security Testing Requirements

API Response Format Changes

Prerequisites Verification

1. System Requirements Check

2. Install Missing Prerequisites

3. Install mkcert (Ubuntu-native approach)

Application Setup and Installation Testing

1. Project Setup

2. Environment Configuration

3. Initialize mkcert CA

4. Start Application

v2.0 API Response Format Testing

1. Health Check Endpoint (Standardized Response)

2. Commands Endpoint (New Standardized Format)

3. Error Response Format Testing

4. Security Validation Testing

Authentication Testing

1. Authentication Status Testing

2. Login Testing

3. Login Failure Testing

4. Logout Testing

5. Protected Routes Testing

6. Login Page Testing

7. Session Persistence Testing

8. Authentication Disabled Testing

OpenID Connect (OIDC) SSO Testing

1. OIDC Configuration Testing

2. OIDC Provider Discovery Testing

3. OIDC Authentication Flow Testing

4. OIDC Login Page Integration Testing

5. OIDC Configuration Validation Testing

6. Dual Authentication Testing

7. OIDC Cleanup Testing

Functional Testing Using Built-in Tools

1. System Status API Testing

2. Root CA Information Testing

3. Certificate Generation Testing

PEM Format Certificate

CRT Format Certificate

4. Certificate Listing Testing

5. File Download Testing

File System Testing

1. Certificate Storage Structure

2. File Permissions Testing

3. Root Certificate Protection Testing

Security Testing

1. Authentication Security Testing

2. Input Validation Testing

3. Rate Limiting Security Testing

5. Session Management Security Testing

Performance Testing

1. Concurrent Certificate Generation

2. Large Certificate List Testing

Browser Integration Testing

1. Web Interface Testing

2. Authentication Flow Testing (Manual Browser Testing)

3. Certificate Trust Testing (if desktop environment available)

Error Handling Testing

1. Service Unavailable Testing

2. Missing mkcert Testing

Cleanup and Validation

1. Test Cleanup

2. Final Validation Report

Automated Test Script

Troubleshooting Common Issues

Authentication Issues

Permission Issues

Port Conflicts

Session Issues

mkcert Issues

Verify CA installation

6. Access the Web Interface

7. Test Certificate Generation

Via Web Interface:

Via API (using curl):

8. Verify Generated Certificates

9. Test Certificate Management and Expiry Display

41 KiB

Raw Permalink Blame History