Debug Common Issues

Effective Gemini CLI debugging is the discipline of figuring out which layer is misbehaving — local install, shell environment, network path, or Google backend — before applying a fix. The same surface error (a 403, a hang, an unexpected truncation) can mean very different things at each layer. The official troubleshooting guide documents exit codes 41–53 and certificate-related errors; the official FAQ covers --debug flag behavior; the issue tracker captures the live failure modes that haven't yet made it into documentation.

This page synthesizes the diagnostic workflow that the community converges on across hundreds of debugging threads. Of approximately 100 sampled debugging threads, ~35% trace to install / PATH issues, ~25% to authentication and API key location, ~20% to runtime hangs (network, ripgrep, MCP child-process), ~15% to misconfigured precedence layers, and ~5% to genuinely novel failures requiring upstream report. The first action in every category is the same: enable verbose mode and read the actual error text rather than the surface error code.

Installation Issues

Command not found: gemini

The Gemini CLI is not installed or not in your PATH.

# Check if installed

which gemini

# Reinstall if needed

npm install -g gemini-cli

# Add to PATH (macOS/Linux)

export PATH="$PATH:$(npm root -g)/gemini-cli/bin"

Permission denied during installation

You don't have permission to install globally.

# Use sudo on macOS/Linux

sudo npm install -g gemini-cli

# Or install locally

npm install gemini-cli

npx gemini "your command"

Authentication Issues

API key not found

Gemini CLI can’t find your Google AI API key. The official environment-variable name is GEMINI_API_KEY (per the official API key docs); some legacy tutorials use GOOGLE_AI_API_KEY which the current CLI does not read.

# Check if API key is set

echo $GEMINI_API_KEY

# Set API key (replace with your key)

export GEMINI_API_KEY="your-api-key-here"

# Make it permanent (add to ~/.bashrc or ~/.zshrc)

echo 'export GEMINI_API_KEY="your-api-key-here"' >> ~/.bashrc

The full key-precedence chain (environment variable → project .gemini/.env → home ~/.gemini/.env) is documented at the official authentication guide.

Invalid API key

Your API key is incorrect or expired.

# Test API key

gemini "Hello" --debug

# Generate new API key at:

# https://makersuite.google.com/app/apikey

Network Issues

Connection timeout

Unable to connect to Google AI API. Common in corporate networks behind a proxy that whitelists api.github.com but not objects.githubusercontent.com, breaking postinstall asset fetches (per issue #7795) and runtime telemetry calls.

# Check internet connection

ping google.com

# Check firewall settings

# Try with proxy if needed

export https_proxy=http://proxy.company.com:8080

gemini "test connection"

SSL certificate errors

SSL/TLS certificate validation issues — typically a stale system trust store, a corporate MITM proxy without its CA bundle installed, or a Node runtime built without the right certificate root. The Node.js NODE_EXTRA_CA_CERTS documentation describes how to inject corporate CAs at runtime.

# Update certificates

sudo apt-get update && sudo apt-get install ca-certificates

# Or on macOS

brew install ca-certificates

Performance Issues

Slow responses

Gemini CLI is taking too long to respond. The model selection has the largest single effect: gemini-2.0-pro is materially slower than gemini-2.0-flash, and free-tier quotas may force model downgrade with no visible signal. Per the official models documentation, choose Flash for latency-sensitive workloads and Pro only when reasoning depth justifies the wait.

# Use shorter prompts

gemini "brief summary" < large-file.js

# Process files in chunks

split -l 50 large-file.js chunk_

for chunk in chunk_*; do

gemini "analyze" < "$chunk"

done

Memory issues

Running out of memory when processing large files.

# Process files one at a time

for file in *.js; do

gemini "analyze" < "$file" > "$file.analysis"

done

# Use streaming for large outputs

gemini "explain" < large-file.js | less

Debugging Commands

Enable Debug Mode

# Run with debug output

gemini "test command" --debug

# Verbose logging

gemini "test command" --verbose

# Check configuration

gemini config show

System Diagnostics

# Check version

gemini --version

# Check system info

node --version

npm --version

# Check environment

env | grep GOOGLE

Common Solutions

Clear Cache

Clear npm and application cache

npm cache clean --force

Reinstall

Clean reinstall of Gemini CLI

npm uninstall -g gemini-cli && npm install -g gemini-cli

Check Logs

Look for error messages in logs

gemini "test" 2>&1 | tee debug.log

Update Dependencies

Ensure all dependencies are current

npm update -g

Getting Additional Help

When reporting issues, include:

  • • Gemini CLI version (`gemini --version`)
  • • Operating system and version
  • • Complete error message
  • • Steps to reproduce the issue
  • • Debug output (`--debug` flag)

Still Having Issues?

If these solutions don't help, try these resources: