How to Adjust Kernel Tuning Parameters on a VPS? Print

Kernel tuning parameters in Linux control how the operating system manages resources like memory, network, and CPU. Adjusting these parameters on a Virtual Private Server (VPS) can optimize performance for specific workloads, such as web servers, databases, or high-traffic applications. This guide provides practical steps to tune kernel parameters using sysctl on a Linux VPS (e.g., Ubuntu or CentOS), focusing on common use cases like improving network performance and memory management.

Why Adjust Kernel Tuning Parameters?

• Optimize Performance: Enhance throughput for network-heavy applications.
• Improve Stability: Prevent resource exhaustion under high load.
• Tailor Resources: Match system behavior to specific workloads (e.g., web hosting, databases).
• Increase Security: Mitigate risks like SYN flood attacks.

Prerequisites

• Root Access: Full administrative access via SSH or console.
• Linux OS: Ubuntu, CentOS, or another distribution with sysctl support.
• Monitoring Tools: Install htop, iotop, or vmstat to monitor performance.
• Backup: Save current kernel configurations before changes.

Steps to Adjust Kernel Tuning Parameters on a Linux VPS

This guide uses Ubuntu 20.04, with notes for CentOS where applicable, and assumes a VPS hosting a web server with moderate traffic.

Step 1: Access the VPS

• Connect via SSH:

  ssh user@your-vps-ip
  

• If SSH is unavailable, use the provider’s console. For example, VPS.DO’s SolusVM control panel offers VNC console access for direct management.

Step 2: Check Current Kernel Parameters

• View current settings:

  sudo sysctl -a
  

• Filter for specific parameters (e.g., network-related):

  sudo sysctl -a | grep net.ipv4
  

Step 3: Adjust Network Parameters

Optimize network performance for web servers or high-traffic applications.

• Increase TCP buffer sizes for better throughput:

  sudo sysctl -w net.core.rmem_max=16777216
  sudo sysctl -w net.core.wmem_max=16777216
  sudo sysctl -w net.ipv4.tcp_rmem="4096 87380 16777216"
  sudo sysctl -w net.ipv4.tcp_wmem="4096 65536 16777216"
  

• Enable TCP window scaling:

  sudo sysctl -w net.ipv4.tcp_window_scaling=1
  

• Increase backlog for incoming connections:

  sudo sysctl -w net.core.somaxconn=65535
  

• Protect against SYN flood attacks:

  sudo sysctl -w net.ipv4.tcp_syncookies=1
  

Step 4: Optimize Memory Management

Adjust memory parameters to improve caching and swapping behavior.

• Increase file descriptor limit:

  sudo sysctl -w fs.file-max=2097152
  

• Reduce swappiness to prioritize RAM over swap (e.g., for database servers):

  sudo sysctl -w vm.swappiness=10
  

• Adjust dirty data writeback for better I/O performance:

  sudo sysctl -w vm.dirty_ratio=15
  sudo sysctl -w vm.dirty_background_ratio=5
  

Step 5: Make Changes Persistent

• Edit /etc/sysctl.conf or a file in /etc/sysctl.d/:

  sudo nano /etc/sysctl.d/99-custom.conf
  

• Add tuned parameters:

  net.core.rmem_max=16777216
  net.core.wmem_max=16777216
  net.ipv4.tcp_rmem=4096 87380 16777216
  net.ipv4.tcp_wmem=4096 65536 16777216
  net.ipv4.tcp_window_scaling=1
  net.core.somaxconn=65535
  net.ipv4.tcp_syncookies=1
  fs.file-max=2097152
  vm.swappiness=10
  vm.dirty_ratio=15
  vm.dirty_background_ratio=5
  

• Apply changes:

  sudo sysctl -p /etc/sysctl.d/99-custom.conf
  

Step 6: Monitor and Test Performance

• Monitor resource usage:

  htop  # CPU and memory
  iotop -o  # Disk I/O
  vmstat 1  # System performance
  

• Test network performance:

  iperf3 -c <remote-server-ip>  # Requires iperf3 installed
  

• Use VPS.DO’s SolusVM dashboard to track real-time CPU, RAM, and bandwidth usage, ensuring tuning aligns with available resources (e.g., 1–14 cores, 2–40 GB RAM).

Step 7: Verify and Fine-Tune

• Check applied parameters:

  sudo sysctl -a | grep -E "net.core.rmem_max|vm.swappiness"
  

• Stress test the system:

  stress --cpu 2 --io 2 --vm 1 --vm-bytes 512M
  

• Monitor logs for errors:

  sudo tail -f /var/log/syslog  # Ubuntu
  sudo tail -f /var/log/messages  # CentOS
  

• Adjust values incrementally if performance issues arise (e.g., reduce buffer sizes for low-memory VPS plans).

Troubleshooting

• Performance Degradation: Revert changes and test one parameter at a time.
• Syntax Errors: Verify /etc/sysctl.conf syntax and file permissions (chmod 644).
• Out of Memory: Reduce buffer sizes or increase swap if memory is limited.
• Provider Limits: Check for provider-imposed restrictions (e.g., CPU or network throttling) via logs or support.

Best Practices

• Tune Conservatively: Start with small changes to avoid instability.
• Match Workload: Adjust parameters based on specific needs (e.g., low swappiness for databases, high buffers for streaming).
• Backup Configurations: Save /etc/sysctl.conf before changes:

  sudo cp /etc/sysctl.conf /etc/sysctl.conf.bak
  

• Monitor Regularly: Use tools like htop or cron jobs to track performance.
• Test Incrementally: Apply changes to a test VPS before production.

When to Seek Help

If tuning causes instability or errors, review logs (/var/log/syslog or /var/log/messages) and revert to backups. Contact your VPS provider with details. Providers like VPS.DO offer 24/7 ticket-based support for troubleshooting performance issues.

Adjusting kernel parameters optimizes your VPS for specific workloads, improving performance and stability while ensuring efficient resource use.