feat: liveness probes, recovery strategies, and skip_if rename #492

Workflow file for this run

	name: CI

	on:
	pull_request:
	branches: [main]
	push:
	branches: [main]

	env:
	CARGO_TERM_COLOR: always
	RUSTFLAGS: "-D warnings"
	FORCE_JAVASCRIPT_ACTIONS_TO_NODE24: true

	jobs:
	commits:
	name: Conventional Commits
	runs-on: ubuntu-latest
	if: github.event_name == 'pull_request'
	steps:
	- uses: actions/checkout@v4
	with:
	fetch-depth: 0
	- name: Validate PR commits follow conventional format
	run: \|
	set -e
	BASE="${{ github.event.pull_request.base.sha }}"
	HEAD="${{ github.event.pull_request.head.sha }}"
	PATTERN='^(feat\|fix\|docs\|style\|refactor\|perf\|test\|build\|ci\|chore\|revert)($.+$)?!?: .+'
	FAILED=0
	while IFS= read -r sha; do
	MSG=$(git log --format='%s' -1 "$sha")
	if ! echo "$MSG" \| grep -qE "$PATTERN"; then
	echo "❌ $sha: $MSG"
	FAILED=1
	else
	echo "✅ $sha: $MSG"
	fi
	done < <(git rev-list "$BASE".."$HEAD")
	if [ "$FAILED" -eq 1 ]; then
	echo ""
	echo "Some commits do not follow conventional commit format."
	echo "Expected: type(scope)?: description"
	echo "Types: feat, fix, docs, style, refactor, perf, test, build, ci, chore, revert"
	exit 1
	fi

	website:
	name: Website Docker Build
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	- name: Build website container
	run: docker build -t veld-website website/

	schema:
	name: Validate JSON Schema
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	- uses: actions/setup-python@v5
	with:
	python-version: "3.12"
	- name: Validate schema and configs
	run: \|
	chmod +x tests/validate-schema.sh
	./tests/validate-schema.sh --install

	go-test:
	name: Go Module Tests
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	- uses: actions/setup-go@v5
	with:
	go-version: '1.22'
	- name: Test caddy/inject module
	working-directory: caddy/inject
	run: go test -v ./...

	check:
	name: Check & Lint
	runs-on: ubuntu-latest
	steps:
	- uses: actions/checkout@v4
	- uses: dtolnay/rust-toolchain@stable
	with:
	components: rustfmt, clippy
	- uses: Swatinem/rust-cache@v2
	- uses: actions/setup-node@v4
	with:
	node-version: '22'
	cache: 'npm'
	cache-dependency-path: crates/veld-daemon/frontend/package-lock.json
	- name: Install frontend dependencies
	run: cd crates/veld-daemon/frontend && npm ci
	- name: cargo fmt --check
	run: cargo fmt --all -- --check
	- name: cargo clippy
	run: cargo clippy --workspace -- -D warnings
	- name: cargo test
	run: cargo test --workspace

	frontend:
	name: Frontend (TypeScript)
	runs-on: ubuntu-latest
	defaults:
	run:
	working-directory: crates/veld-daemon/frontend
	steps:
	- uses: actions/checkout@v4
	- uses: actions/setup-node@v4
	with:
	node-version: '22'
	cache: 'npm'
	cache-dependency-path: crates/veld-daemon/frontend/package-lock.json
	- run: npm ci
	- name: Type-check
	run: npm run typecheck
	- name: Test
	run: npm test

	integration:
	name: Integration Tests (macOS)
	runs-on: macos-latest
	needs: check
	steps:
	- uses: actions/checkout@v4
	- uses: dtolnay/rust-toolchain@stable
	- uses: Swatinem/rust-cache@v2
	- uses: actions/setup-node@v4
	with:
	node-version: '22'
	cache: 'npm'
	cache-dependency-path: crates/veld-daemon/frontend/package-lock.json
	- name: Install frontend dependencies
	run: cd crates/veld-daemon/frontend && npm ci
	- uses: actions/setup-go@v5
	with:
	go-version: '1.22'

	- name: Build all binaries
	run: cargo build

	# Build Caddy with the veld_inject plugin from this branch.
	# install_caddy() just checks the binary exists — no markers needed.
	- name: Build Caddy with local veld_inject
	run: \|
	export PATH="$(go env GOPATH)/bin:$PATH"
	XCADDY_VERSION=$(cat .xcaddy-version)
	go install "github.com/caddyserver/xcaddy/cmd/xcaddy@v${XCADDY_VERSION}"
	mkdir -p "$HOME/.local/lib/veld"
	xcaddy build \
	--with github.com/prosperity-solutions/veld/caddy/inject=./caddy/inject \
	--output "$HOME/.local/lib/veld/caddy"
	chmod +x "$HOME/.local/lib/veld/caddy"
	# Stash a copy so we can restore after cleanup steps.
	cp "$HOME/.local/lib/veld/caddy" /tmp/caddy-preserve

	- name: Verify CLI basics
	run: \|
	./target/debug/veld --version
	./target/debug/veld --help

	- name: Test config parsing (testproject)
	working-directory: testproject
	run: \|
	../target/debug/veld nodes
	../target/debug/veld nodes --json
	../target/debug/veld graph frontend:local

	- name: Verify setup subcommands exist
	run: \|
	set -e
	./target/debug/veld setup --help \| grep -q "unprivileged"
	./target/debug/veld setup --help \| grep -q "privileged"
	./target/debug/veld setup --help \| grep -q "hammerspoon"
	./target/debug/veld doctor --help \| grep -q "json"
	echo "Setup subcommands and doctor verified."

	- name: Test Python servers directly
	run: \|
	set -e
	python3 testproject/backend/server.py 19555 &
	PID_B=$!
	python3 testproject/frontend/server.py 19556 &
	PID_F=$!
	sleep 2
	curl -sf http://localhost:19555/health
	curl -sf http://localhost:19555/
	curl -sf http://localhost:19556/
	kill $PID_B $PID_F
	wait $PID_B $PID_F 2>/dev/null \|\| true

	- name: Run full integration tests
	run: \|
	chmod +x tests/integration.sh
	sudo ./tests/integration.sh \
	--veld-bin "$(pwd)/target/debug/veld" \
	--project-dir "$(pwd)/testproject"

	- name: "Cleanup after integration tests"
	run: \|
	sudo "$(pwd)/target/debug/veld" uninstall 2>/dev/null \|\| true
	curl -s -X POST http://localhost:2019/stop 2>/dev/null \|\| true
	sudo rm -rf "$HOME/.veld" "$HOME/.local/lib/veld" /usr/local/lib/veld 2>/dev/null \|\| true
	sudo chown -R "$(whoami)" "$HOME/.local" 2>/dev/null \|\| true
	sudo launchctl bootout system/dev.veld.helper 2>/dev/null \|\| true
	launchctl bootout "gui/$(id -u)/dev.veld.helper" 2>/dev/null \|\| true
	launchctl bootout "gui/$(id -u)/dev.veld.daemon" 2>/dev/null \|\| true
	sleep 2
	# Restore the Caddy binary for subsequent tests.
	mkdir -p "$HOME/.local/lib/veld"
	cp /tmp/caddy-preserve "$HOME/.local/lib/veld/caddy"
	chmod +x "$HOME/.local/lib/veld/caddy"

	- name: "Test: unprivileged setup lifecycle"
	run: \|
	set -e
	VELD="$(pwd)/target/debug/veld"
	echo "=== Unprivileged setup ==="
	# Setup may fail on CA trust in CI (no keychain / Caddy timing) — allow partial success
	$VELD setup unprivileged \|\| echo "WARNING: setup returned non-zero (CA trust may have failed in CI)"

	echo ""
	echo "=== Run doctor (diagnostics) ==="
	$VELD doctor \|\| echo "WARNING: doctor reported issues (expected in CI — Caddy may not start)"
	$VELD doctor --json > /tmp/doctor-unpriv.json 2>/dev/null \|\| true
	cat /tmp/doctor-unpriv.json

	echo ""
	echo "=== Verify setup.json ==="
	cat ~/.veld/setup.json
	grep -q '"unprivileged"' ~/.veld/setup.json

	echo ""
	echo "=== Verify helper is running on user socket ==="
	test -S ~/.veld/helper.sock
	echo "Helper socket OK"

	echo ""
	echo "=== Check if Caddy is running ==="
	CADDY_OK=0
	for i in 1 2 3 4 5; do
	if curl -sf http://localhost:2019/id/veld-sentinel 2>/dev/null \| grep -q "veld"; then
	CADDY_OK=1
	break
	fi
	sleep 2
	done

	if [ "$CADDY_OK" = "1" ]; then
	echo "Caddy is running (sentinel verified)"
	curl -sf http://localhost:2019/config/apps/http/servers/veld/listen \| grep -q "18443"
	echo "Caddy listening on port 18443"

	echo ""
	echo "=== Start a service in unprivileged mode ==="
	cd testproject
	$VELD start frontend:local --name unpriv-test
	sleep 2

	echo ""
	echo "=== Verify URLs have :18443 ==="
	$VELD urls --name unpriv-test --json \| grep -q "18443"

	echo ""
	echo "=== Stop ==="
	$VELD stop --name unpriv-test
	cd ..
	else
	echo "WARNING: Caddy did not start in CI (timing issue), skipping HTTPS tests"
	fi

	echo ""
	echo "=== Uninstall (unprivileged) ==="
	$VELD uninstall \|\| echo "WARNING: uninstall returned non-zero"

	echo ""
	echo "=== Verify cleanup ==="
	sleep 2
	if ! curl -sf http://localhost:2019/config/ >/dev/null 2>&1; then
	echo "Caddy stopped OK"
	else
	echo "WARNING: Caddy still running (may need manual cleanup)"
	fi

	if [ ! -f ~/.veld/setup.json ]; then
	echo "setup.json removed OK"
	else
	echo "WARNING: setup.json still exists"
	fi

	echo ""
	echo "=== Unprivileged lifecycle PASSED ==="

	- name: "Cleanup between lifecycle tests (unpriv → priv)"
	if: always()
	run: \|
	curl -s -X POST http://localhost:2019/stop 2>/dev/null \|\| true
	launchctl bootout "gui/$(id -u)/dev.veld.helper" 2>/dev/null \|\| true
	launchctl bootout "gui/$(id -u)/dev.veld.daemon" 2>/dev/null \|\| true
	rm -rf "$HOME/.veld" 2>/dev/null \|\| true
	rm -f "$HOME/.local/lib/veld/caddy-data" 2>/dev/null \|\| true
	# Privileged setup runs as root. sudo on macOS may resolve
	# lib_dir() to /usr/local/lib/veld. Ensure Caddy is there too.
	sudo mkdir -p /usr/local/lib/veld
	sudo cp /tmp/caddy-preserve /usr/local/lib/veld/caddy
	sudo chmod +x /usr/local/lib/veld/caddy
	sleep 1

	- name: "Test: privileged setup lifecycle"
	run: \|
	set -e
	VELD="$(pwd)/target/debug/veld"
	echo "=== Privileged setup ==="
	sudo $VELD setup privileged
	echo ""

	echo "=== Run doctor (diagnostics) ==="
	$VELD doctor \|\| echo "WARNING: doctor reported issues (expected in CI)"
	$VELD doctor --json > /tmp/doctor-priv.json 2>/dev/null \|\| true
	cat /tmp/doctor-priv.json

	echo ""
	echo "=== Verify setup.json ==="
	cat ~/.veld/setup.json
	grep -q '"privileged"' ~/.veld/setup.json

	echo ""
	echo "=== Verify helper is running on system socket ==="
	test -S /var/run/veld-helper.sock
	echo "System socket OK"

	echo ""
	echo "=== Check Caddy ==="
	CADDY_OK=0
	for i in 1 2 3 4 5; do
	if curl -sf http://localhost:2019/id/veld-sentinel 2>/dev/null \| grep -q "veld"; then
	CADDY_OK=1
	break
	fi
	sleep 2
	done

	if [ "$CADDY_OK" = "1" ]; then
	echo "Caddy running (sentinel verified)"
	curl -sf http://localhost:2019/config/apps/http/servers/veld/listen \| grep -q '":443"'
	echo "Caddy on port 443"

	echo ""
	echo "=== Start a service in privileged mode ==="
	cd testproject
	if $VELD start frontend:local --name priv-test 2>&1; then
	sleep 2
	URLS=$($VELD urls --name priv-test --json 2>/dev/null \|\| true)
	if echo "$URLS" \| grep -q "18443"; then
	echo "FAIL: URLs contain :18443 in privileged mode"
	exit 1
	fi
	echo "Clean URLs OK"
	$VELD stop --name priv-test 2>/dev/null \|\| true
	else
	echo "WARNING: veld start failed in CI (non-fatal)"
	fi
	cd ..
	else
	echo "WARNING: Caddy did not start, skipping service tests"
	fi

	echo ""
	echo "=== Uninstall (privileged) ==="
	sudo $VELD uninstall \|\| echo "WARNING: uninstall returned non-zero"

	echo ""
	echo "=== Verify cleanup ==="
	sleep 2
	if ! curl -sf http://localhost:2019/config/ >/dev/null 2>&1; then
	echo "Caddy stopped OK"
	else
	echo "WARNING: Caddy still running"
	fi
	echo "=== Privileged lifecycle PASSED ==="

	- name: "Cleanup between lifecycle tests (priv → auto)"
	if: always()
	run: \|
	sudo curl -s -X POST http://localhost:2019/stop 2>/dev/null \|\| true
	sudo launchctl bootout system/dev.veld.helper 2>/dev/null \|\| true
	sudo rm -rf "$HOME/.veld" "$HOME/.local/lib/veld" /usr/local/lib/veld 2>/dev/null \|\| true
	sudo chown -R "$(whoami)" "$HOME/.local" 2>/dev/null \|\| true
	sleep 1
	# Restore the Caddy binary for subsequent tests.
	mkdir -p "$HOME/.local/lib/veld"
	cp /tmp/caddy-preserve "$HOME/.local/lib/veld/caddy"
	chmod +x "$HOME/.local/lib/veld/caddy"

	- name: "Test: auto-bootstrap (no setup)"
	run: \|
	set -e
	VELD="$(pwd)/target/debug/veld"
	echo "=== Auto-bootstrap: start without any setup ==="
	cd testproject
	# Auto-bootstrap may fail in CI due to Caddy startup issues — allow soft failure
	if $VELD start frontend:local --name auto-test 2>&1; then
	sleep 2

	echo ""
	echo "=== Verify it auto-bootstrapped ==="
	cat ~/.veld/setup.json
	grep -q '"auto"' ~/.veld/setup.json

	echo ""
	echo "=== Verify URLs have :18443 ==="
	$VELD urls --name auto-test --json \| grep -q "18443"

	echo ""
	echo "=== Stop ==="
	$VELD stop --name auto-test
	else
	echo "WARNING: auto-bootstrap veld start failed in CI (Caddy may not start in this environment)"
	# Still verify the setup.json was written
	if [ -f ~/.veld/setup.json ]; then
	grep -q '"auto"' ~/.veld/setup.json && echo "setup.json mode=auto OK"
	fi
	fi
	cd ..

	echo ""
	echo "=== Cleanup auto-bootstrap state ==="
	curl -s http://localhost:2019/stop >/dev/null 2>&1 \|\| true
	rm -rf "$HOME/.veld"
	sleep 1
	echo "=== Auto-bootstrap PASSED ==="

	# ── Release build matrix — catch build failures before merge ──────
	release-build:
	name: Release Build (${{ matrix.suffix }})
	runs-on: ${{ matrix.os }}
	needs: check
	strategy:
	fail-fast: false
	matrix:
	include:
	- target: aarch64-apple-darwin
	os: macos-latest
	suffix: macos-arm64
	goos: darwin
	goarch: arm64
	- target: x86_64-apple-darwin
	os: macos-latest
	suffix: macos-amd64
	goos: darwin
	goarch: amd64
	- target: x86_64-unknown-linux-gnu
	os: ubuntu-latest
	suffix: linux-amd64
	goos: linux
	goarch: amd64
	- target: aarch64-unknown-linux-gnu
	os: ubuntu-latest
	suffix: linux-arm64
	cross: true
	goos: linux
	goarch: arm64
	steps:
	- uses: actions/checkout@v4
	- uses: dtolnay/rust-toolchain@stable
	with:
	targets: ${{ matrix.target }}
	- uses: Swatinem/rust-cache@v2
	- uses: actions/setup-node@v4
	with:
	node-version: '22'
	cache: 'npm'
	cache-dependency-path: crates/veld-daemon/frontend/package-lock.json
	- name: Install frontend dependencies
	run: cd crates/veld-daemon/frontend && npm ci
	- uses: actions/setup-go@v5
	with:
	go-version: '1.22'

	- name: Install cross-compilation tools
	if: matrix.cross
	run: \|
	sudo apt-get update
	sudo apt-get install -y gcc-aarch64-linux-gnu
	echo '[target.aarch64-unknown-linux-gnu]' >> ~/.cargo/config.toml
	echo 'linker = "aarch64-linux-gnu-gcc"' >> ~/.cargo/config.toml

	- name: Build release binaries
	run: cargo build --release --target ${{ matrix.target }}

	- name: Build Caddy with veld_inject
	run: \|
	export PATH="$(go env GOPATH)/bin:$PATH"
	XCADDY_VERSION=$(cat .xcaddy-version)
	go install "github.com/caddyserver/xcaddy/cmd/xcaddy@v${XCADDY_VERSION}"
	GOOS=${{ matrix.goos }} GOARCH=${{ matrix.goarch }} xcaddy build \
	--with github.com/prosperity-solutions/veld/caddy/inject=./caddy/inject \
	--output ./caddy-build-test
	rm -f ./caddy-build-test

	injection-test:
	name: Injection E2E (macOS)
	runs-on: macos-latest
	needs: check
	steps:
	- uses: actions/checkout@v4
	- uses: dtolnay/rust-toolchain@stable
	- uses: Swatinem/rust-cache@v2
	- uses: actions/setup-node@v4
	with:
	node-version: '22'
	cache: 'npm'
	cache-dependency-path: crates/veld-daemon/frontend/package-lock.json
	- name: Install frontend dependencies
	run: cd crates/veld-daemon/frontend && npm ci
	- uses: actions/setup-go@v5
	with:
	go-version: '1.22'

	- name: Build veld
	run: cargo build

	- name: Install Next.js deps
	working-directory: testproject-nextjs
	run: npm install

	- name: Build Caddy with local veld_inject
	run: \|
	export PATH="$(go env GOPATH)/bin:$PATH"
	XCADDY_VERSION=$(cat .xcaddy-version)
	go install "github.com/caddyserver/xcaddy/cmd/xcaddy@v${XCADDY_VERSION}"
	mkdir -p "$HOME/.local/lib/veld"
	xcaddy build \
	--with github.com/prosperity-solutions/veld/caddy/inject=./caddy/inject \
	--output "$HOME/.local/lib/veld/caddy"
	chmod +x "$HOME/.local/lib/veld/caddy"

	- name: Setup veld (unprivileged)
	run: \|
	./target/debug/veld setup unprivileged \|\| echo "WARNING: setup returned non-zero (CA trust may have failed)"

	- name: Wait for Caddy
	run: \|
	for i in $(seq 1 15); do
	if curl -sf http://localhost:2019/id/veld-sentinel 2>/dev/null \| grep -q "veld"; then
	echo "Caddy is running"
	break
	fi
	sleep 2
	done

	- name: Run injection E2E tests
	run: \|
	chmod +x tests/test-injection.sh
	./tests/test-injection.sh \
	--veld-bin "$(pwd)/target/debug/veld" \
	--project-dir "$(pwd)/testproject-nextjs"

	- name: Cleanup
	if: always()
	run: \|
	./target/debug/veld uninstall 2>/dev/null \|\| true
	curl -s -X POST http://localhost:2019/stop 2>/dev/null \|\| true
	rm -rf "$HOME/.veld" "$HOME/.local/lib/veld" 2>/dev/null \|\| true
	launchctl bootout "gui/$(id -u)/dev.veld.helper" 2>/dev/null \|\| true
	launchctl bootout "gui/$(id -u)/dev.veld.daemon" 2>/dev/null \|\| true

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: liveness probes, recovery strategies, and skip_if rename #492

Workflow file

feat: liveness probes, recovery strategies, and skip_if rename #492

Uh oh!

Workflow file for this run