examples

Conductor Configuration Examples

This directory contains example configurations for common testing scenarios.

Available Examples

1. Web Load Test (web_load_test.cfg)

A comprehensive web application load test that coordinates:

• Web server monitoring
• Multiple load generators
• Performance monitoring

Components:

• web_server.cfg - The server under test
• load_generator1.cfg - First load generator
• monitor.cfg - Monitoring node

Usage:

# Start players on each machine
player web_server.cfg      # On web server
player load_generator1.cfg # On load gen 1
player monitor.cfg         # On monitor

# Run test from conductor
conduct web_load_test.cfg

2. Database Performance Test (database_test.cfg)

Tests database performance with multiple concurrent clients.

Components:

• database_server.cfg - Database server configuration
• database_client1.cfg - First database client
• database_client2.cfg - Second database client

Features:

• Connection pool testing
• Query performance monitoring
• Transaction throughput measurement

3. Network Latency Test (network_test.cfg)

Measures network latency and throughput between nodes.

Use Cases:

• Datacenter connectivity testing
• WAN link performance
• Network troubleshooting

4. Distributed System Test (distributed_test.cfg)

Tests a complete distributed system with multiple components.

Includes:

• Load balancer configuration
• Multiple application servers
• Database cluster
• Cache servers

Configuration Tips

Command Types

1. Normal Commands - Wait for completion

   step1 = echo "This runs and waits"

2. Spawn Commands - Run in background

   spawn1 = tail -f /var/log/app.log

3. Timeout Commands - Run with time limit

   timeout60 = stress --cpu 4 --timeout 60

Best Practices

1. Always include cleanup in the Reset phase
2. Create result directories in Startup phase
3. Use spawn for monitoring commands
4. Collect logs and metrics in Collect phase
5. Test locally first with localhost configurations

Common Patterns

Monitoring Pattern

[Run]
spawn1 = vmstat 1 > /tmp/vmstat.log
spawn2 = iostat -x 1 > /tmp/iostat.log
spawn3 = top -b -d 1 > /tmp/top.log

Health Check Pattern

[Run]
timeout5 = curl -f http://localhost/health || echo "Failed"

Result Collection Pattern

[Collect]
step1 = mkdir -p /tmp/results
step2 = cp /var/log/app.log /tmp/results/
step3 = tar -czf /tmp/results.tgz /tmp/results/

Customizing Examples

To customize these examples for your environment:

1. Update IP addresses in the [Coordinator] section
2. Modify commands for your specific applications
3. Adjust timeouts based on your test duration
4. Add more steps as needed

Creating Your Own

To create a new test configuration:

1. Copy an example that's similar to your use case
2. Update the [Coordinator] section with your network details
3. Modify the phases (Startup, Run, Collect, Reset)
4. Test with a single client first
5. Scale up to multiple clients

Troubleshooting

If examples don't work:

1. Check network connectivity between conductor and players
2. Verify ports 6970/6971 are open
3. Ensure commands exist on player machines
4. Check file permissions for output directories
5. Review player output for error messages

For more help, see the Installation Guide.