fix(frontend): keep admin page height consistent #1547

Workflow file for this run

.github/workflows/benchmarks.yml at db2333f

	# Licensed to the Apache Software Foundation (ASF) under one
	# or more contributor license agreements. See the NOTICE file
	# distributed with this work for additional information
	# regarding copyright ownership. The ASF licenses this file
	# to you under the Apache License, Version 2.0 (the
	# "License"); you may not use this file except in compliance
	# with the License. You may obtain a copy of the License at
	#
	# http://www.apache.org/licenses/LICENSE-2.0
	#
	# Unless required by applicable law or agreed to in writing,
	# software distributed under the License is distributed on an
	# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
	# KIND, either express or implied. See the License for the
	# specific language governing permissions and limitations
	# under the License.

	# Texera benchmarks — bench-agnostic umbrella workflow.
	#
	# This file is the single CI entry for ALL Texera performance benchmarks
	# (currently Arrow Flight E2E; JMH and others land here as well). The
	# workflow knows nothing about specific benches — bin/run-benchmarks.sh
	# is the opaque entry point that owns which benches run and where their
	# outputs land under bench-results/. Adding a new bench is:
	# 1. Append the run command to bin/run-benchmarks.sh.
	# 2. Add a `Publish <chart-name>` step block below pointing at the
	# bench's JSON output file with the right `tool:` setting.
	# This workflow file otherwise stays unchanged.
	#
	# Triggering — mirrors amber-integration's label gate (NOT file paths):
	# - PR: runs only when one of the labels mapped to the amber-integration
	# stack in required-checks.yml's LABEL_STACKS is present on the PR.
	# Labels are applied by the .github/labeler.yml workflow on opened /
	# synchronize, so we wait for that workflow to complete before
	# deciding (same pattern required-checks.yml uses). A PR run benches
	# the PR head, then re-runs the IDENTICAL trimmed grid against the
	# base-branch (main) commit it targets, in the SAME runner (see the
	# "Benchmark main baseline in the same runner" step). The delta between
	# those two cancels cross-runner hardware variance (the dominant source
	# of CI bench noise), so the PR comment's main-vs-branch comparison is
	# apples-to-apples rather than PR-here vs a stored baseline captured on
	# some other runner. PRs never publish to gh-pages.
	# - push to main: always runs (same trimmed grid as PR for quick post-
	# merge signal) but does NOT publish to gh-pages; it only emits the
	# job summary plus uploaded artifact. Publishing on every merge spammed
	# the repo's Pulse / all-branches commit count with bot commits, so
	# only the scheduled (daily) run persists the baseline now.
	# - schedule (daily): runs the full 27-config sweep and is the sole
	# writer that publishes to gh-pages (the authoritative long-term
	# baseline).
	# - workflow_dispatch: manual full-grid run (no publish; bring-your-own
	# trigger for ad-hoc exploration).
	#
	# Two modes via BENCH_MODE env (read by the bench Scala main):
	# pr — 3 configs × 20 batches, ~5 min (PR + push-to-main)
	# full — 27 configs × 200 batches, ~40 min (schedule + dispatch)
	#
	# Non-blocking: this workflow is NOT included in required-checks.yml's
	# `required-checks` aggregator, so its result doesn't gate merges even
	# when it fails. Adding it to branch protection later is a deliberate
	# .asf.yaml change.
	#
	# Permissions:
	# contents: write — needed by benchmark-action's auto-push to gh-pages.
	# PR runs (which GitHub auto-downgrades to read-only on forks) gate
	# auto-push off via the event check, so the missing write is never
	# exercised.

	name: Benchmarks

	on:
	push:
	branches: [main]
	pull_request:
	types: [opened, reopened, synchronize, labeled, unlabeled]
	schedule:
	# Daily full-grid baseline refresh, 12:00 UTC (05:00 PDT). PR and
	# post-merge runs use a trimmed 3-config grid to stay around 5 min; the
	# scheduled run covers the full 27-config sweep that the gh-pages
	# dashboard tracks long-term. Daily (rather than weekly) keeps the
	# baseline fresh and accumulates enough data points to average out CI
	# noise; the extra bot commits on gh-pages are intentionally tolerated.
	# 12:00 UTC lands in the early-morning PDT lull when GitHub runners are
	# less contended (and thus less noisy) than during late-night dev hours.
	# Bump to several times a day by adding more cron entries if denser
	# sampling is wanted.
	- cron: "0 12 * * *"
	workflow_dispatch:

	permissions:
	contents: write

	concurrency:
	group: benchmarks-${{ github.ref }}
	# On main: never cancel an in-flight baseline run; on PRs: supersede.
	cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}

	jobs:
	precheck:
	# Decide whether to run based on PR labels (push / dispatch always
	# run). Lifted from required-checks.yml's precheck so the trigger
	# surface matches amber-integration exactly.
	name: Precheck
	runs-on: ubuntu-latest
	outputs:
	run_bench: ${{ steps.decide.outputs.run_bench }}
	steps:
	- name: Wait for Pull Request Labeler
	if: github.event_name == 'pull_request'
	uses: actions/github-script@v8
	with:
	script: \|
	const ref = context.payload.pull_request.head.sha;
	const maxAttempts = 30;
	for (let i = 0; i < maxAttempts; i++) {
	const { data } = await github.rest.checks.listForRef({
	owner: context.repo.owner,
	repo: context.repo.repo,
	ref,
	check_name: "labeler",
	});
	const check = data.check_runs[0];
	if (check && check.status === "completed") {
	core.info(`labeler ${check.conclusion}`);
	return;
	}
	core.info(`labeler not ready (attempt ${i + 1}/${maxAttempts})`);
	await new Promise((r) => setTimeout(r, 10000));
	}
	core.warning("labeler did not complete within 5 minutes; proceeding with current labels.");

	- name: Decide whether to run bench
	id: decide
	uses: actions/github-script@v8
	with:
	script: \|
	const eventName = context.eventName;
	if (eventName !== "pull_request") {
	// push to main / workflow_dispatch always run.
	core.info(`event=${eventName} — running unconditionally`);
	core.setOutput("run_bench", "true");
	return;
	}
	// Re-fetch labels: the labeler may have just added some.
	const { data: pr } = await github.rest.pulls.get({
	owner: context.repo.owner,
	repo: context.repo.repo,
	pull_number: context.payload.pull_request.number,
	});
	const labels = pr.labels.map((l) => l.name);
	core.info(`PR labels: ${labels.join(", ") \|\| "(none)"}`);
	// Mirrors LABEL_STACKS in required-checks.yml: every label
	// whose stack list contains "amber-integration" triggers this
	// bench. Keep in sync if LABEL_STACKS there changes.
	const TRIGGER_LABELS = new Set([
	"pyamber",
	"engine",
	"amber-integration",
	"common",
	"ddl-change",
	"ci",
	]);
	const matched = labels.filter((l) => TRIGGER_LABELS.has(l));
	const shouldRun = matched.length > 0;
	core.info(
	shouldRun
	? `Triggering on labels: ${matched.join(", ")}`
	: "No trigger label present; skipping bench."
	);
	core.setOutput("run_bench", shouldRun ? "true" : "false");

	bench:
	name: Bench
	needs: precheck
	if: ${{ needs.precheck.outputs.run_bench == 'true' }}
	runs-on: ubuntu-latest
	env:
	JAVA_OPTS: -Xms2048M -Xmx2048M -Xss6M -XX:ReservedCodeCacheSize=256M -Dfile.encoding=UTF-8
	JVM_OPTS: -Xms2048M -Xmx2048M -Xss6M -XX:ReservedCodeCacheSize=256M -Dfile.encoding=UTF-8
	# `pr` mode = 3-config trimmed sweep (~5 min) for PR + post-merge.
	# `full` mode = 27-config sweep (~40 min) for schedule + manual.
	# Read by the bench Scala main (see GridSpec switch); workflow only
	# decides which mode to pass.
	BENCH_MODE: ${{ (github.event_name == 'schedule' \|\| github.event_name == 'workflow_dispatch') && 'full' \|\| 'pr' }}
	services:
	# The bench itself doesn't touch the DB, but sbt's transitive compile
	# chain reaches `common/auth` which imports JOOQ-generated classes
	# from `org.apache.texera.dao.jooq.generated.*`. JOOQ codegen at
	# sbt compile time requires a live Postgres to introspect against;
	# without it the auth module's `User` / `UserRoleEnum` symbols fail
	# to resolve and the whole bench compile aborts. Mirrors the same
	# service block from amber-integration in build.yml.
	postgres:
	image: postgres
	env:
	POSTGRES_PASSWORD: postgres
	ports:
	- 5432:5432
	options: >-
	--health-cmd="pg_isready -U postgres"
	--health-interval=10s
	--health-timeout=5s
	--health-retries=5
	steps:
	- name: Checkout
	uses: actions/checkout@v5
	with:
	fetch-depth: 0
	- name: Setup JDK
	uses: actions/setup-java@v5
	with:
	distribution: "temurin"
	java-version: 17
	- name: Setup Python
	uses: actions/setup-python@v6
	with:
	python-version: "3.12"
	- name: Install Python dependencies
	# Mirrors amber-integration's installer in build.yml so the bench
	# subprocess imports resolve identically (pytorch CPU index +
	# betterproto plugin via dev-requirements).
	run: \|
	python -m pip install uv
	if [ -f amber/requirements.txt ]; then uv pip install --system --index-strategy unsafe-best-match -r amber/requirements.txt; fi
	if [ -f amber/operator-requirements.txt ]; then uv pip install --system --index-strategy unsafe-best-match -r amber/operator-requirements.txt; fi
	if [ -f amber/dev-requirements.txt ]; then uv pip install --system --index-strategy unsafe-best-match -r amber/dev-requirements.txt; fi
	- name: Install protoc
	run: \|
	PROTOC_VERSION=$(cat bin/protoc-version.txt)
	curl -fsSL -o /tmp/protoc.zip "https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOC_VERSION}/protoc-${PROTOC_VERSION}-linux-x86_64.zip"
	sudo unzip -o /tmp/protoc.zip -d /usr/local
	sudo chmod +x /usr/local/bin/protoc
	sudo chmod -R a+rX /usr/local/include/google
	- name: Create Database for JOOQ codegen
	# Minimal subset of amber-integration's "Create Databases" step —
	# JOOQ only introspects against texera_db, not iceberg/lakefs/
	# lakekeeper schemas which the bench never touches.
	run: psql -h localhost -U postgres -f sql/texera_ddl.sql
	env:
	PGPASSWORD: postgres
	- name: Generate Python proto bindings
	run: bash bin/python-proto-gen.sh
	- name: Setup sbt launcher
	uses: sbt/setup-sbt@508b753e53cb6095967669e0911487d2b9bc9f41 # v1.1.22
	- uses: coursier/cache-action@90c37294538be80a558fd665531fcdc2b467b475 # v8.1.0
	with:
	extraSbtFiles: '[".sbt", "project/*.{scala,sbt}", "project/build.properties" ]'

	- name: Run benchmarks
	# Single opaque entry point — this workflow doesn't know which
	# benches exist. Adding a JMH suite later = appending one line
	# inside bin/run-benchmarks.sh and adding a publish step below.
	run: bash bin/run-benchmarks.sh

	- name: Benchmark main baseline in the same runner
	# PR only: re-run the IDENTICAL trimmed grid against the base-branch
	# (main) commit this PR targets, in THIS runner, right after the PR
	# run above. Comparing two runs from the same machine cancels the
	# cross-runner hardware variance that otherwise dominates CI bench
	# deltas, so benchmarks-pr-comment.yml can show a trustworthy
	# main-vs-branch comparison instead of PR-here vs a stored baseline
	# captured on some other runner.
	#
	# The output convention is preserved: the PR's own outputs stay in
	# bench-results/ untouched; we only ADD main's CSV as
	# arrow-flight-e2e-main.csv (plus the base SHA in a sidecar file).
	# The PR-mode grid is deterministic (see GridSpec in
	# ArrowFlightActorBench.scala), so main's rows key 1:1 against the
	# PR's rows for the comparison.
	#
	# Fail-soft by construction: no `set -e`, and a trap restores the
	# PR's results plus the original checkout no matter where the main
	# re-run dies (broken main, compile error, etc). On failure we emit
	# no main CSV, and the comment workflow falls back to the stored
	# gh-pages baseline. We also skip entirely if the PR run produced no
	# CSV (e.g. the bench itself failed upstream).
	if: ${{ github.event_name == 'pull_request' && !cancelled() }}
	env:
	BASE_SHA: ${{ github.event.pull_request.base.sha }}
	run: \|
	set -uo pipefail
	if [ ! -f bench-results/arrow-flight-e2e.csv ]; then
	echo "::warning::no PR bench CSV; skipping same-runner main baseline."
	exit 0
	fi
	ORIG_REF=$(git rev-parse HEAD)
	# Park the PR's outputs; main's re-run writes a fresh bench-results/.
	mv bench-results bench-results-pr
	restore() {
	rm -rf bench-results
	mv bench-results-pr bench-results 2>/dev/null \|\| true
	git checkout --force "$ORIG_REF" 2>/dev/null \|\| true
	}
	trap restore EXIT
	if ! git checkout --force "$BASE_SHA"; then
	echo "::warning::could not check out base SHA $BASE_SHA; skipping main baseline."
	exit 0
	fi
	# Re-sync Python deps to main's requirements: the deps installed
	# earlier are the PR's, and the bench subprocess imports must match
	# the main code we're about to compile and run. Without this the
	# "main" baseline would run main's Scala/Python sources against the
	# PR's pinned Python packages, which is not a clean main measurement.
	# (sbt recompiles main's Scala automatically when run-benchmarks.sh
	# invokes it below; only the pip deps need an explicit re-sync.)
	if [ -f amber/requirements.txt ]; then uv pip install --system --index-strategy unsafe-best-match -r amber/requirements.txt \|\| { echo "::warning::main requirements install failed; skipping main baseline."; exit 0; }; fi
	if [ -f amber/operator-requirements.txt ]; then uv pip install --system --index-strategy unsafe-best-match -r amber/operator-requirements.txt \|\| { echo "::warning::main operator-requirements install failed; skipping main baseline."; exit 0; }; fi
	if [ -f amber/dev-requirements.txt ]; then uv pip install --system --index-strategy unsafe-best-match -r amber/dev-requirements.txt \|\| { echo "::warning::main dev-requirements install failed; skipping main baseline."; exit 0; }; fi
	# Regenerate proto bindings against main's protos, then re-bench.
	bash bin/python-proto-gen.sh \|\| { echo "::warning::main proto-gen failed; skipping main baseline."; exit 0; }
	if bash bin/run-benchmarks.sh && [ -f bench-results/arrow-flight-e2e.csv ]; then
	cp bench-results/arrow-flight-e2e.csv bench-results-pr/arrow-flight-e2e-main.csv
	printf '%s' "$BASE_SHA" > bench-results-pr/arrow-flight-e2e-main.commit.txt
	echo "captured same-runner main baseline at $BASE_SHA"
	else
	echo "::warning::main baseline re-run failed; PR comment falls back to the gh-pages baseline."
	fi
	# trap restores the PR outputs (now incl. main CSV) plus original ref.

	- name: Stash PR number for downstream comment workflow
	# PR fork workflows can't comment (GitHub forces read-only token);
	# benchmarks-pr-comment.yml runs separately via workflow_run with
	# proper write access, and needs the PR number to find the target.
	# github.event.workflow_run.pull_requests is empty for fork PRs,
	# so we ferry the number via artifact.
	if: ${{ github.event_name == 'pull_request' && !cancelled() }}
	env:
	PR_NUMBER: ${{ github.event.pull_request.number }}
	run: echo "$PR_NUMBER" > bench-results/pr-number.txt

	- name: Render bench summary
	# Render the bench CSV into a markdown table on the workflow run
	# page. Visible without further clicks — and doesn't need any
	# extra permissions (writes to $GITHUB_STEP_SUMMARY only).
	if: ${{ !cancelled() }}
	run: \|
	{
	echo "## Bench results (\`$BENCH_MODE\` mode)"
	echo
	if [ -f bench-results/arrow-flight-e2e.csv ]; then
	echo '```csv'
	cat bench-results/arrow-flight-e2e.csv
	echo '```'
	else
	echo "_(no bench-results/arrow-flight-e2e.csv produced)_"
	fi
	} >> "$GITHUB_STEP_SUMMARY"

	- name: Upload bench artifacts
	if: ${{ !cancelled() }}
	uses: actions/upload-artifact@v4
	with:
	name: bench-results-${{ github.run_id }}
	path: bench-results/
	retention-days: 14

	# Publish to the gh-pages dashboard. auto-push + save-data-file are
	# gated on `schedule` ONLY: the daily full-grid run is the single
	# authoritative baseline writer. PR and push-to-main runs only emit
	# the job summary and the uploaded artifact, never touching the
	# tracked baseline. This is deliberate: each gh-pages write is a bot
	# commit (one per chart, so two per run), and persisting on every
	# merge to main flooded the repo's Pulse / all-branches commit count
	# with `github-action-benchmark` commits. The post-merge run still
	# gives quick signal via the rendered summary + artifact; only the
	# daily sweep persists. Adding a new benchmark = adding one publish
	# block below matching the JSON filename convention in
	# bin/run-benchmarks.sh.
	#
	# `skip-fetch-gh-pages: true` intentionally keeps baseline comparison
	# OFF for now. When flipped to false, the action will fetch the stored
	# gh-pages baseline, compare each run against main, post an alert
	# comment when a result regresses past `alert-threshold`, and (with
	# comment-on-alert / fail-on-alert) can block merge. We're deferring
	# that until the baseline has accumulated enough daily data points to
	# be trustworthy; turning it on is a deliberate follow-up to evaluate
	# later. auto-push on the daily schedule still appends to the branch.
	#
	# `continue-on-error: true` keeps any other gh-pages-side surprise
	# (permission glitches, transient git failures) from failing the
	# bench job overall — the bench data itself is already in the
	# uploaded artifact above.
	- name: Publish throughput
	if: ${{ !cancelled() }}
	continue-on-error: true
	uses: benchmark-action/github-action-benchmark@52576c92bccf6ac60c8223ec7eb2565637cae9ba # v1.22.1
	with:
	name: Arrow Flight E2E Throughput
	tool: customBiggerIsBetter
	output-file-path: bench-results/arrow-flight-e2e-throughput.json
	github-token: ${{ secrets.GITHUB_TOKEN }}
	auto-push: ${{ github.event_name == 'schedule' }}
	save-data-file: ${{ github.event_name == 'schedule' }}
	skip-fetch-gh-pages: true
	gh-pages-branch: gh-pages
	benchmark-data-dir-path: dev/bench
	alert-threshold: "150%"
	# comment-on-alert needs pull-requests:write; skip and let
	# results show up via summary-always instead.
	comment-on-alert: false
	summary-always: true
	- name: Publish latency
	if: ${{ !cancelled() }}
	continue-on-error: true
	uses: benchmark-action/github-action-benchmark@52576c92bccf6ac60c8223ec7eb2565637cae9ba # v1.22.1
	with:
	name: Arrow Flight E2E Latency
	tool: customSmallerIsBetter
	output-file-path: bench-results/arrow-flight-e2e-latency.json
	github-token: ${{ secrets.GITHUB_TOKEN }}
	auto-push: ${{ github.event_name == 'schedule' }}
	save-data-file: ${{ github.event_name == 'schedule' }}
	skip-fetch-gh-pages: true
	gh-pages-branch: gh-pages
	benchmark-data-dir-path: dev/bench
	alert-threshold: "150%"
	comment-on-alert: false
	summary-always: true

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(frontend): keep admin page height consistent #1547

Workflow file

fix(frontend): keep admin page height consistent #1547

Uh oh!

Workflow file for this run