DEV Community

Cover image for How to Fix High RAM Usage on a Linux Dedicated Server
Ethan Vance
Ethan Vance

Posted on Originally published at migservers.com AI-assisted

How to Fix High RAM Usage on a Linux Dedicated Server

#linux #devops #sysadmin #tutorial

Experiencing sluggish performance, unresponsive terminals, or unexpected application crashes on your MIG servers infrastructure usually points to memory exhaustion.

This guide provides the exact diagnostic commands and mitigation strategies to stabilize your Linux environment and permanently resolve memory leaks, using production-safe best practices.

1. Evaluate Current Memory Availability

Before taking action, you must understand how your server distributes its resources. Run the following command to get a snapshot of your system's memory:

free -h
Enter fullscreen mode Exit fullscreen mode

Total
Description: Physical RAM installed on your server.
Actionable Insight: Baseline reference for your hardware capacity.

Available
Description: Memory currently free and ready for new processes.
Actionable Insight: If consistently near zero (and swap is growing), your server is struggling.

Buff/Cache
Description: RAM used by the Linux kernel to cache files.
Actionable Insight: High numbers are healthy; Linux frees this automatically when apps need RAM.

Swap
Description: Disk space used as emergency overflow RAM.
Actionable Insight: Sustained, active swap usage severely degrades performance.

2. Identify Resource Hogs

Locate the specific applications draining your RAM using built-in Linux utilities.

  • Interactive Monitoring: Run top and press Shift + M to sort active tasks by memory usage. For a more readable interface, install and run htop (press F6 to sort by memory).

  • Targeted Process Snapshot: To immediately print the top 10 memory-consuming processes directly to your terminal:

ps aux --sort=-%mem | head -10
Enter fullscreen mode Exit fullscreen mode
  • Track Memory Leaks: If you suspect an application's Resident Set Size (RSS) is growing endlessly without releasing memory, monitor its Process ID (PID) over a 10-minute window: 
for i in $(seq 1 10); do ps -o pid,rss,vsz -p <PID>; sleep 60; done
Enter fullscreen mode Exit fullscreen mode

3. Immediate Remediation Actions

If your server is actively freezing, apply these commands to safely recover system resources.

  • Gracefully Terminate Rogue Processes: Send a standard termination signal to allow the application to clean up and shut down properly. 
kill <PID>
Enter fullscreen mode Exit fullscreen mode

Crucial Warning: Only use kill -9 as an absolute last resort if the process is completely frozen. It forces an immediate shutdown without cleanup, which can cause data corruption in databases or file writes.>

  • Restart Application Services: Restarting daemons (like Nginx, MySQL, or Node.js) flushes their allocated memory pools. 
sudo systemctl restart <service_name>
Enter fullscreen mode Exit fullscreen mode
  • Diagnostic Cache Clearing (Use Sparingly): Linux intentionally uses free RAM to cache files for better performance. Dropping the cache can prove whether RAM is truly locked up by applications, but doing this routinely will actually hurt server performance as the kernel is forced to re-read files from the disk. 
sync && echo 1 | sudo tee /proc/sys/vm/drop_caches
Enter fullscreen mode Exit fullscreen mode

4. Long-Term Server Optimization

Prevent future bottlenecks by tuning kernel parameters and configuring workload-specific limits.

  • Lower Swappiness: Depending on your workload, you can encourage the kernel to favor physical RAM before relying on slow disk swap. Open /etc/sysctl.conf, append vm.swappiness=10, and apply with sudo sysctl -p.

  • Deploy Emergency Swap Space: Swap prevents fatal Out-Of-Memory (OOM) crashes by providing overflow space, though it is not a replacement for sufficient physical RAM. Size this appropriately for your server (e.g., 2GB to 4GB):

sudo fallocate -l 2G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile
Enter fullscreen mode Exit fullscreen mode
  • Enforce Workload-Based Memory Limits: Prevent a single memory leak from taking down the entire server by setting Systemd limits. Run sudo systemctl edit <service_name> and configure limits based on your app's actual needs: 
[Service]
MemoryMax=4G
MemorySwapMax=0
Enter fullscreen mode Exit fullscreen mode

(Note: Set MemoryMax based on actual profiling; if set too low, Systemd will routinely kill your service. Setting MemorySwapMax=0 strictly prevents the service from using swap, which guarantees an OOM kill if physical RAM limits are exceeded).

  • Audit and Disable Unused Services: Background services consume baseline RAM. Identify unnecessary software and disable it from booting: 
sudo systemctl disable <service_name>
Enter fullscreen mode Exit fullscreen mode

Discussion:

What is your go-to terminal command when a server starts lagging? Drop your tips in the comments below! 👇

Top comments (0)