Skip to content
DevProc

Resource Monitoring

DevProc provides real-time resource monitoring for all running services, helping you identify performance issues and resource-hungry processes.

Overview

Resource monitoring shows:

  • CPU usage - Percentage of CPU time used
  • Memory usage - RAM consumed by the process

Data is collected every second using the system’s ps command.

Service List View

Running services display resource metrics inline:

● api         12.5%   45.2MB  :8080  4m
● worker       2.1%   23.1MB         3m
● web          8.3%   67.8MB  :5173  3m
  • CPU % - Color-coded: green (<50%), yellow (50-80%), red (>80%)
  • Memory - Displayed in appropriate units (KB, MB, GB)

Resources only show for running/healthy services.

Resource Graph View

Press m to toggle the resource graph for the selected service.

CPU Usage
▁▂▃▄▅▆▇█▇▆▅▄▃▂▁▂▃▄▅▆▇█▇▆▅▄▃▂▁▂▃▄▅▆▇█▇▆▅▄▃▂▁▂▃  12.5%


Memory Usage
▅▅▅▅▅▅▅▅▅▅▅▅▅▆▆▆▆▆▆▆▆▆▆▆▆▆▆▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇  45.2MB


History: 60 samples (60s)
Memory %: 0.8% of system

Sparkline Graph

The sparkline shows historical data:

  • Each character represents one sample (1 second)
  • Height indicates relative usage
  • Up to 60 seconds of history

Graph Information

  • Current value - Live reading next to the graph
  • Sample count - How much history is shown
  • Memory % - Percentage of total system memory

Press m again to return to log view.

Color Coding

CPU usage is color-coded for quick identification:

ColorCPU RangeIndication
Green/Cyan0-50%Normal
Yellow50-80%Elevated
Red80%+High

How It Works

DevProc uses the ps command to collect metrics:

Terminal window
ps -p <pid> -o %cpu=,rss=,%mem=
  • %cpu - CPU percentage
  • rss - Resident Set Size (physical memory)
  • %mem - Memory percentage of total system RAM

Metrics are polled every second for running processes.

Monitoring Multiple Services

While the graph shows one service at a time, the service list always shows current metrics for all running services.

Quick comparison:

  1. Look at the service list
  2. Identify services with high CPU (red) or high memory
  3. Select a service and press m for detailed history

Process Trees

For services that spawn child processes, only the main process is monitored. Child processes aren’t included in the metrics.

If your service uses workers or child processes, consider:

  • Monitoring at the OS level (htop, top)
  • Using application-level metrics

Docker Compose Services

For Docker Compose services, DevProc monitors the docker compose process, not the container itself.

To monitor actual container resources:

  • Use docker stats
  • Use Docker Desktop dashboard
  • Configure container resource limits

Limitations

Platform Support

Resource monitoring works on:

  • macOS - Full support
  • Linux - Full support
  • Windows/WSL - Limited support (may need additional tools)

Refresh Rate

Metrics update every second. Sub-second spikes may not be captured.

CPU Accuracy

CPU percentage is calculated over the sampling interval. Brief spikes may be averaged out.

Best Practices

Monitor During Development

Keep an eye on resource usage while developing:

  • Unexpected CPU spikes may indicate infinite loops
  • Memory growth may indicate leaks
  • Compare before/after code changes

Identify Resource Hogs

Use the service list to quickly identify which services are consuming the most resources.

Check After Deployment Changes

When you change configuration or code:

  1. Restart the service
  2. Monitor resource usage
  3. Compare to previous baseline

Set Resource Expectations

Know what’s normal for your services:

  • Database connections typically use more memory
  • Build/watch processes have CPU spikes
  • Idle services should use minimal resources