Add a way to bring up instances on ec2, benchmark them, and get profiles #71

Workflow file for this run

	name: CI

	on:
	push:
	branches: [main]
	pull_request:

	env:
	CARGO_TERM_COLOR: always
	RUST_BACKTRACE: 1
	# wal-rs is a git submodule at walshadow/wal-rs (see .gitmodules);
	# the path dep in Cargo.toml points there. Submodule URL is SSH; the
	# `Init wal-rs submodule` step in each job feeds a read-only deploy
	# key (secrets.WAL_RS_DEPLOY_KEY) via GIT_SSH_COMMAND so the clone
	# auths without disturbing the primary checkout's GITHUB_TOKEN path.

	# Cancel superseded runs on the same ref.
	concurrency:
	group: ${{ github.workflow }}-${{ github.ref }}
	cancel-in-progress: true

	jobs:
	test:
	name: build + e2e (pg${{ matrix.pg-major }})
	runs-on: ubuntu-latest
	timeout-minutes: 60

	# walshadow operationally supports PG 16+ (see PLAN.md "Supported
	# PostgreSQL versions"). Matrix pins the on-PATH PG binaries to one
	# specific major; tests + fixture-capture scripts pick up whichever
	# `initdb` / `pg_ctl` / `postgres` lands first.
	strategy:
	fail-fast: false
	matrix:
	pg-major: [16, 17, 18]

	steps:
	- name: Checkout walshadow
	uses: actions/checkout@v6
	with:
	path: walshadow

	- name: Init wal-rs submodule
	working-directory: walshadow
	env:
	WAL_RS_DEPLOY_KEY: ${{ secrets.WAL_RS_DEPLOY_KEY }}
	run: \|
	install -d -m 700 ~/.ssh
	printf '%s\n' "$WAL_RS_DEPLOY_KEY" > ~/.ssh/wal_rs_deploy
	chmod 600 ~/.ssh/wal_rs_deploy
	ssh-keyscan -H github.com >> ~/.ssh/known_hosts 2>/dev/null
	GIT_SSH_COMMAND='ssh -i ~/.ssh/wal_rs_deploy -o IdentitiesOnly=yes' \
	git submodule update --init --recursive

	- name: Install Rust toolchain
	uses: dtolnay/rust-toolchain@stable
	with:
	components: clippy, rustfmt

	- name: Cargo cache
	uses: Swatinem/rust-cache@v2
	with:
	workspaces: walshadow
	# Key on PG major so a feature/test-config change for one
	# major doesn't poison the others' caches.
	key: pg${{ matrix.pg-major }}

	- name: Install C deps (lz4 + zstd for clickhouse-c-rs)
	run: \|
	sudo apt-get update
	sudo apt-get install -y --no-install-recommends \
	build-essential liblz4-dev libzstd-dev curl gnupg lsb-release

	# The runner's preinstalled postgresql-common ships only one major
	# at a time; pull the requested major straight from PGDG so the
	# matrix can pin any of 16 / 17 / 18 independently of the runner
	# image's choice. `systemctl stop postgresql` parks any auto-
	# started instance so tests own the unix-socket dirs cleanly.
	- name: Install PostgreSQL ${{ matrix.pg-major }} (PGDG)
	run: \|
	sudo systemctl stop postgresql 2>/dev/null \|\| true
	curl -fsSL https://www.postgresql.org/media/keys/ACCC4CF8.asc \
	\| sudo gpg --dearmor -o /usr/share/keyrings/postgres-archive-keyring.gpg
	echo "deb [signed-by=/usr/share/keyrings/postgres-archive-keyring.gpg] https://apt.postgresql.org/pub/repos/apt/ $(lsb_release -cs)-pgdg main" \
	\| sudo tee /etc/apt/sources.list.d/pgdg.list
	sudo apt-get update
	sudo apt-get install -y --no-install-recommends \
	postgresql-${{ matrix.pg-major }} \
	postgresql-client-${{ matrix.pg-major }} \
	postgresql-server-dev-${{ matrix.pg-major }}
	sudo systemctl stop postgresql 2>/dev/null \|\| true
	PG_BIN=/usr/lib/postgresql/${{ matrix.pg-major }}/bin
	"$PG_BIN/postgres" -V
	"$PG_BIN/initdb" --version
	echo "$PG_BIN" >> "$GITHUB_PATH"

	# Official deb repo gives us both the `clickhouse` multitool (for
	# `clickhouse local` tests) and `clickhouse-server` (for the
	# readme_quickstart that connects to localhost:9000).
	- name: Install ClickHouse (server + client)
	run: \|
	sudo apt-get install -y apt-transport-https ca-certificates
	curl -fsSL https://packages.clickhouse.com/rpm/lts/repodata/repomd.xml.key \
	\| sudo gpg --dearmor -o /usr/share/keyrings/clickhouse-keyring.gpg
	echo "deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb stable main" \
	\| sudo tee /etc/apt/sources.list.d/clickhouse.list
	sudo apt-get update
	DEBIAN_FRONTEND=noninteractive sudo -E apt-get install -y \
	-o Dpkg::Options::="--force-confdef" \
	-o Dpkg::Options::="--force-confold" \
	clickhouse-server clickhouse-client clickhouse-common-static
	clickhouse --version

	- name: Start ClickHouse server
	run: \|
	sudo service clickhouse-server start
	# Wait for native protocol on :9000 (readme_quickstart probes it).
	for i in $(seq 1 60); do
	if clickhouse-client --query "SELECT 1" >/dev/null 2>&1; then
	echo "clickhouse-server up after ${i}s"
	exit 0
	fi
	sleep 1
	done
	echo "clickhouse-server failed to come up" >&2
	sudo tail -n 200 /var/log/clickhouse-server/clickhouse-server.err.log \|\| true
	exit 1

	# classify / filter / xlog_switch / vacuum_full_pg_depend tests
	# silently skip when no fixture bytes are present. Regenerate
	# them against the matrix PG major so each job's
	# filter_round_trip + classify_fixture + wal_stream_chunk_boundary
	# actually runs against that major's WAL layout.
	- name: Regenerate WAL fixtures (PG ${{ matrix.pg-major }})
	working-directory: walshadow
	env:
	WALSHADOW_USE_LOCAL: "1"
	run: \|
	fixtures/wal/classify/capture.sh
	fixtures/wal/filter/capture.sh
	fixtures/wal/xlog_switch/capture.sh
	fixtures/wal/vacuum_full_pg_depend/capture.sh

	- name: Build (workspace + tests)
	working-directory: walshadow
	run: cargo build --workspace --all-targets --locked

	# All e2e tests run: shadow_lifecycle, shadow_catalog, catalog_seed,
	# wal_stream_e2e (need initdb), classify_fixture + filter_round_trip
	# + wal_stream_chunk_boundary (use the freshly regenerated fixtures),
	# clickhouse-c-rs clickhouse_local + readme_quickstarts (need
	# clickhouse binary + server on :9000).
	- name: Test (workspace, single-threaded — each test owns a PG cluster)
	working-directory: walshadow
	env:
	# Tests pick non-overlapping ports but PG cluster startup is heavy;
	# avoid thrashing the runner by serializing.
	RUST_TEST_THREADS: "1"
	# Opts tests with `fx::dump_artifacts` into copying their tempdir
	# here on failure. Empty on success; uploaded as a job artifact
	# by the next step when present.
	WALSHADOW_ARTIFACT_DIR: ${{ github.workspace }}/test-artifacts
	run: cargo test --workspace --all-targets --locked -- --nocapture

	- name: Upload test artifacts on failure
	if: failure()
	uses: actions/upload-artifact@v7
	with:
	name: test-artifacts-pg${{ matrix.pg-major }}
	path: ${{ github.workspace }}/test-artifacts
	if-no-files-found: ignore
	retention-days: 7

	# walshadow extension is a separate PGXS module (shipped to shadow PG).
	# Build it against the matrix PG, install into that major's $libdir /
	# share/extension, then run pg_regress under a private --temp-instance
	# so we don't need an external server. Covers the C bridge function's
	# varlena, fixed-width by-val, by-ref, NULL, and ereport paths.
	- name: walshadow extension (build + pg_regress)
	working-directory: walshadow/pgext
	run: \|
	PG_BIN=/usr/lib/postgresql/${{ matrix.pg-major }}/bin
	PG_CONFIG=$PG_BIN/pg_config
	make PG_CONFIG=$PG_CONFIG
	sudo make PG_CONFIG=$PG_CONFIG install
	PG_REGRESS=/usr/lib/postgresql/${{ matrix.pg-major }}/lib/pgxs/src/test/regress/pg_regress
	rm -rf /tmp/walshadow_pgext_tmpinst
	"$PG_REGRESS" \
	--inputdir=./ --outputdir=./ \
	--bindir="$PG_BIN" \
	--temp-instance=/tmp/walshadow_pgext_tmpinst \
	--port=54329 \
	walshadow

	- name: walshadow extension regression diffs on failure
	if: failure()
	working-directory: walshadow/pgext
	run: \|
	test -f regression.diffs && cat regression.diffs \|\| true
	test -d /tmp/walshadow_pgext_tmpinst/log && \
	sudo tail -n 200 /tmp/walshadow_pgext_tmpinst/log/postmaster.log \|\| true

	- name: ClickHouse server logs on failure
	if: failure()
	run: sudo tail -n 500 /var/log/clickhouse-server/clickhouse-server.err.log \|\| true

	# Line-coverage baseline so future work can track movement. Runs the
	# full e2e suite under `cargo llvm-cov` against the middle supported
	# major (PG 17) — one major is enough for a coverage baseline; the
	# matrix above proves cross-version behaviour separately. Uploads
	# `lcov.info` + an HTML report as artifacts; summary lands in the
	# job log.
	coverage:
	name: line coverage (pg17)
	runs-on: ubuntu-latest
	timeout-minutes: 60

	steps:
	- name: Checkout walshadow
	uses: actions/checkout@v6
	with:
	path: walshadow

	- name: Init wal-rs submodule
	working-directory: walshadow
	env:
	WAL_RS_DEPLOY_KEY: ${{ secrets.WAL_RS_DEPLOY_KEY }}
	run: \|
	install -d -m 700 ~/.ssh
	printf '%s\n' "$WAL_RS_DEPLOY_KEY" > ~/.ssh/wal_rs_deploy
	chmod 600 ~/.ssh/wal_rs_deploy
	ssh-keyscan -H github.com >> ~/.ssh/known_hosts 2>/dev/null
	GIT_SSH_COMMAND='ssh -i ~/.ssh/wal_rs_deploy -o IdentitiesOnly=yes' \
	git submodule update --init --recursive

	- name: Install Rust toolchain
	uses: dtolnay/rust-toolchain@stable
	with:
	# cargo-llvm-cov needs `llvm-profdata` + `llvm-cov` shipped by
	# this component.
	components: llvm-tools-preview

	- name: Install cargo-llvm-cov
	uses: taiki-e/install-action@v2
	with:
	tool: cargo-llvm-cov

	- name: Cargo cache
	uses: Swatinem/rust-cache@v2
	with:
	workspaces: walshadow
	# Instrumented builds differ from the test job's; pin a
	# separate cache key so they don't collide.
	key: coverage

	- name: Install C deps
	run: \|
	sudo apt-get update
	sudo apt-get install -y --no-install-recommends \
	build-essential liblz4-dev libzstd-dev curl gnupg lsb-release

	- name: Install PostgreSQL 17 (PGDG)
	run: \|
	sudo systemctl stop postgresql 2>/dev/null \|\| true
	curl -fsSL https://www.postgresql.org/media/keys/ACCC4CF8.asc \
	\| sudo gpg --dearmor -o /usr/share/keyrings/postgres-archive-keyring.gpg
	echo "deb [signed-by=/usr/share/keyrings/postgres-archive-keyring.gpg] https://apt.postgresql.org/pub/repos/apt/ $(lsb_release -cs)-pgdg main" \
	\| sudo tee /etc/apt/sources.list.d/pgdg.list
	sudo apt-get update
	sudo apt-get install -y --no-install-recommends \
	postgresql-17 postgresql-client-17
	sudo systemctl stop postgresql 2>/dev/null \|\| true
	echo "/usr/lib/postgresql/17/bin" >> "$GITHUB_PATH"

	- name: Install ClickHouse (server + client)
	run: \|
	sudo apt-get install -y apt-transport-https ca-certificates
	curl -fsSL https://packages.clickhouse.com/rpm/lts/repodata/repomd.xml.key \
	\| sudo gpg --dearmor -o /usr/share/keyrings/clickhouse-keyring.gpg
	echo "deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb stable main" \
	\| sudo tee /etc/apt/sources.list.d/clickhouse.list
	sudo apt-get update
	DEBIAN_FRONTEND=noninteractive sudo -E apt-get install -y \
	-o Dpkg::Options::="--force-confdef" \
	-o Dpkg::Options::="--force-confold" \
	clickhouse-server clickhouse-client clickhouse-common-static

	- name: Start ClickHouse server
	run: \|
	sudo service clickhouse-server start
	for i in $(seq 1 60); do
	if clickhouse-client --query "SELECT 1" >/dev/null 2>&1; then
	echo "clickhouse-server up after ${i}s"
	exit 0
	fi
	sleep 1
	done
	sudo tail -n 200 /var/log/clickhouse-server/clickhouse-server.err.log \|\| true
	exit 1

	- name: Regenerate WAL fixtures (PG 17)
	working-directory: walshadow
	env:
	WALSHADOW_USE_LOCAL: "1"
	run: \|
	fixtures/wal/classify/capture.sh
	fixtures/wal/filter/capture.sh
	fixtures/wal/xlog_switch/capture.sh
	fixtures/wal/vacuum_full_pg_depend/capture.sh

	# Single `--no-report` test pass so cargo only instruments &
	# executes the suite once; subsequent `report` invocations re-read
	# the cached profdata to emit lcov + html + summary.
	- name: Run instrumented test suite (cargo llvm-cov)
	working-directory: walshadow
	env:
	RUST_TEST_THREADS: "1"
	WALSHADOW_ARTIFACT_DIR: ${{ github.workspace }}/test-artifacts
	run: \|
	cargo llvm-cov clean --workspace
	cargo llvm-cov --workspace --all-targets --locked --no-report \
	-- --nocapture

	- name: Upload test artifacts on failure
	if: failure()
	uses: actions/upload-artifact@v7
	with:
	name: test-artifacts-coverage
	path: ${{ github.workspace }}/test-artifacts
	if-no-files-found: ignore
	retention-days: 7

	- name: Coverage summary (job log)
	working-directory: walshadow
	run: cargo llvm-cov report --summary-only

	- name: Coverage lcov.info
	working-directory: walshadow
	run: cargo llvm-cov report --lcov --output-path lcov.info

	- name: Coverage HTML
	working-directory: walshadow
	run: cargo llvm-cov report --html

	- name: Upload coverage artifacts
	uses: actions/upload-artifact@v7
	with:
	name: coverage
	path: \|
	walshadow/lcov.info
	walshadow/target/llvm-cov/html
	if-no-files-found: error
	retention-days: 30

	- name: Fail on any 0%-coverage file
	working-directory: walshadow
	run: \|
	zero=$(awk '
	/^SF:/ { f = substr($0, 4) }
	/^LF:/ { lf = substr($0, 4) + 0 }
	/^LH:/ {
	if (substr($0, 4) + 0 == 0 && lf > 0) {
	i = index(f, "/src/"); if (i) f = substr(f, i + 1)
	print f
	}
	}
	' lcov.info)
	if [ -n "$zero" ]; then
	count=$(printf '%s\n' "$zero" \| grep -c .)
	echo "Files with 0% line coverage (every source file must be exercised by a test):"
	printf '%s\n' "$zero" \| sed 's/^/ /'
	echo "::error::${count} file(s) have 0% line coverage"
	exit 1
	fi
	echo "OK: all instrumented files have >0% line coverage."

	- name: ClickHouse server logs on failure
	if: failure()
	run: sudo tail -n 500 /var/log/clickhouse-server/clickhouse-server.err.log \|\| true

	lint:
	name: clippy + fmt
	runs-on: ubuntu-latest
	timeout-minutes: 15

	steps:
	- name: Checkout walshadow
	uses: actions/checkout@v6
	with:
	path: walshadow

	- name: Init wal-rs submodule
	working-directory: walshadow
	env:
	WAL_RS_DEPLOY_KEY: ${{ secrets.WAL_RS_DEPLOY_KEY }}
	run: \|
	install -d -m 700 ~/.ssh
	printf '%s\n' "$WAL_RS_DEPLOY_KEY" > ~/.ssh/wal_rs_deploy
	chmod 600 ~/.ssh/wal_rs_deploy
	ssh-keyscan -H github.com >> ~/.ssh/known_hosts 2>/dev/null
	GIT_SSH_COMMAND='ssh -i ~/.ssh/wal_rs_deploy -o IdentitiesOnly=yes' \
	git submodule update --init --recursive

	- name: Install Rust toolchain
	uses: dtolnay/rust-toolchain@stable
	with:
	components: clippy, rustfmt

	- name: Cargo cache
	uses: Swatinem/rust-cache@v2
	with:
	workspaces: walshadow
	key: lint

	- name: Install C deps
	run: \|
	sudo apt-get update
	sudo apt-get install -y --no-install-recommends \
	build-essential liblz4-dev libzstd-dev

	- name: cargo fmt
	working-directory: walshadow
	run: cargo fmt --all -- --check

	- name: cargo clippy
	working-directory: walshadow
	run: cargo clippy --workspace --all-targets -- -D warnings

Navigation Menu

Uh oh!

Add a way to bring up instances on ec2, benchmark them, and get profiles #71

Workflow file

Add a way to bring up instances on ec2, benchmark them, and get profiles #71

Uh oh!

Workflow file for this run