name: CI
on:
push:
branches: [main, master]
pull_request:
branches: [main, master]
env:
CARGO_TERM_COLOR: always
jobs:
build:
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, windows-latest, macos-latest]
rust: [stable]
steps:
- uses: actions/checkout@v4
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
components: clippy, rustfmt
- name: Cache cargo
uses: actions/cache@v4
with:
path: |
~/.cargo/bin/
~/.cargo/registry/index/
~/.cargo/registry/cache/
~/.cargo/git/db/
target/
key: ${{ runner.os }}-cargo-${{ hashFiles('**/Cargo.lock') }}
- name: Install Ollama (Linux)
if: runner.os == 'Linux'
run: |
# Install Ollama with retry logic for transient 503 errors
install_ollama() {
local max_retries=5
local retry=0
local delay=5
while [ $retry -lt $max_retries ]; do
echo "Attempting Ollama installation (attempt $((retry + 1))/$max_retries)..."
if curl -fsSL --retry 3 --retry-delay 5 https://ollama.com/install.sh | sh; then
echo "Ollama installation successful!"
return 0
fi
retry=$((retry + 1))
if [ $retry -lt $max_retries ]; then
echo "Installation failed, retrying in ${delay}s..."
sleep $delay
delay=$((delay * 2)) # Exponential backoff
fi
done
echo "ERROR: Ollama installation failed after $max_retries attempts"
return 1
}
install_ollama
OLLAMA_KEEP_ALIVE=60m ollama serve &
# Wait for server to be ready
for i in {1..30}; do
if curl -s http://localhost:11434/api/tags > /dev/null 2>&1; then
echo "Ollama server is ready"
break
fi
echo "Waiting for Ollama server... ($i/30)"
sleep 2
done
ollama pull llama3.2
# Warm up the model by loading it into memory - MUST complete before tests
echo "Warming up llama3.2 model (this loads it into memory)..."
curl -s --max-time 300 http://localhost:11434/api/generate -d '{"model": "llama3.2", "prompt": "Say hello", "stream": false}'
echo ""
echo "Model warm-up complete - model is now loaded"
- name: Install Ollama (macOS)
if: runner.os == 'macOS'
run: |
brew install ollama
OLLAMA_KEEP_ALIVE=60m ollama serve &
# Wait for server to be ready
for i in {1..30}; do
if curl -s http://localhost:11434/api/tags > /dev/null 2>&1; then
echo "Ollama server is ready"
break
fi
echo "Waiting for Ollama server... ($i/30)"
sleep 2
done
echo "Pulling llama3.2 model..."
ollama pull llama3.2
echo "Model pull complete. Checking available models:"
ollama list
# Warm up the model by loading it into memory - MUST complete before tests
echo "Warming up llama3.2 model (this loads it into memory)..."
echo "Start time: $(date)"
curl -v --max-time 600 http://localhost:11434/api/generate -d '{"model": "llama3.2", "prompt": "Say hello", "stream": false}' 2>&1 || {
echo "Warm-up curl failed with exit code $?"
echo "Checking if Ollama is still running..."
curl -s http://localhost:11434/api/tags || echo "Server not responding"
exit 1
}
echo ""
echo "End time: $(date)"
echo "Model warm-up complete - model is now loaded"
- name: Install Ollama (Windows)
if: runner.os == 'Windows'
run: |
# Download Ollama CLI directly (smaller than full zip)
$ollamaUrl = "https://github.com/ollama/ollama/releases/download/v0.13.2/ollama-windows-amd64.zip"
Write-Host "Downloading Ollama from $ollamaUrl"
Invoke-WebRequest -Uri $ollamaUrl -OutFile "ollama.zip" -TimeoutSec 600
# Extract
Expand-Archive -Path "ollama.zip" -DestinationPath "." -Force
# Verify ollama.exe exists
if (Test-Path ".\ollama.exe") {
Write-Host "ollama.exe found"
} else {
Write-Host "ERROR: ollama.exe not found after extraction"
Get-ChildItem -Recurse | Select-Object FullName
exit 1
}
# Start Ollama server in background with keep-alive to prevent model unloading
$env:OLLAMA_KEEP_ALIVE = "60m"
Start-Process -FilePath ".\ollama.exe" -ArgumentList "serve" -WindowStyle Hidden
# Wait for server to be ready
$maxRetries = 30
$retry = 0
while ($retry -lt $maxRetries) {
Start-Sleep -Seconds 2
try {
$response = Invoke-WebRequest -Uri "http://localhost:11434/api/tags" -TimeoutSec 5 -ErrorAction SilentlyContinue
if ($response.StatusCode -eq 200) {
Write-Host "Ollama server is ready"
break
}
} catch {
Write-Host "Waiting for Ollama server... ($retry/$maxRetries)"
}
$retry++
}
if ($retry -eq $maxRetries) {
Write-Host "ERROR: Ollama server failed to start"
exit 1
}
# Pull the model
.\ollama.exe pull llama3.2
# Warm up the model by loading it into memory - MUST complete before tests
Write-Host "Warming up llama3.2 model (this loads it into memory)..."
$body = '{"model": "llama3.2", "prompt": "Say hello", "stream": false}'
$response = Invoke-WebRequest -Uri "http://localhost:11434/api/generate" -Method POST -Body $body -ContentType "application/json" -TimeoutSec 300
Write-Host "Model warm-up complete - model is now loaded"
Write-Host "Response: $($response.Content)"
shell: pwsh
- name: Check formatting
run: cargo fmt -- --check
- name: Clippy
run: cargo clippy -- -D warnings
- name: Build
run: cargo build --verbose
- name: Build tests (pre-compile before Ollama warm-up)
run: cargo test --no-run --verbose
- name: Re-warm and run Ollama tests FIRST (Linux/macOS)
if: runner.os != 'Windows'
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
run: |
echo "Checking Ollama server health..."
if ! curl -s http://localhost:11434/api/tags > /dev/null 2>&1; then
echo "ERROR: Ollama server is not responding!"
echo "Attempting to restart Ollama..."
OLLAMA_KEEP_ALIVE=60m ollama serve &
sleep 5
fi
echo "Available models:"
curl -s http://localhost:11434/api/tags
echo ""
echo "Re-warming llama3.2 model..."
curl -s --max-time 300 http://localhost:11434/api/generate -d '{"model": "llama3.2", "prompt": "Ready", "stream": false}' || {
echo "Warm-up failed, checking server status..."
curl -s http://localhost:11434/api/tags || echo "Server not responding"
exit 1
}
echo ""
echo "Model loaded - running Ollama tests IMMEDIATELY while model is in memory..."
# Run both Ollama-specific tests that need the model loaded
cargo test --test ai_integration -- test_ollama_explain_finding --test-threads=1
cargo test --test ai_integration -- test_all_providers_explain_same_finding --test-threads=1
echo "Ollama tests passed!"
- name: Re-warm and run Ollama tests FIRST (Windows)
if: runner.os == 'Windows'
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
run: |
Write-Host "Checking Ollama server health..."
try {
$response = Invoke-WebRequest -Uri "http://localhost:11434/api/tags" -TimeoutSec 5 -ErrorAction Stop
Write-Host "Server is healthy"
} catch {
Write-Host "WARNING: Ollama server not responding, attempting restart..."
$env:OLLAMA_KEEP_ALIVE = "60m"
Start-Process -FilePath ".\ollama.exe" -ArgumentList "serve" -WindowStyle Hidden
Start-Sleep -Seconds 5
}
Write-Host "Re-warming llama3.2 model..."
$body = '{"model": "llama3.2", "prompt": "Ready", "stream": false}'
Invoke-WebRequest -Uri "http://localhost:11434/api/generate" -Method POST -Body $body -ContentType "application/json" -TimeoutSec 300
Write-Host "Model loaded - running Ollama tests IMMEDIATELY while model is in memory..."
# Run both Ollama-specific tests that need the model loaded
cargo test --test ai_integration -- test_ollama_explain_finding --test-threads=1
cargo test --test ai_integration -- test_all_providers_explain_same_finding --test-threads=1
Write-Host "Ollama tests passed!"
shell: pwsh
- name: Run remaining tests
run: cargo test --verbose -- --skip test_ollama_explain_finding --skip test_all_providers_explain_same_finding
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
release:
needs: build
runs-on: ubuntu-latest
if: startsWith(github.ref, 'refs/tags/')
steps:
- uses: actions/checkout@v4
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
- name: Build release
run: cargo build --release
- name: Create Release
uses: softprops/action-gh-release@v1
with:
files: target/release/mcplint
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}