merge(release/v3.8.51): refresh onto 3b752f9d4c

Bring inherited Fast Quality Gates / ESLint / unit-test / docs-sync fixes from #11940/#11955/#11975.

Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
This commit is contained in:
dependabot[bot]
2026-08-29 04:05:55 -03:00
committed by diegosouzapw
1382 changed files with 9000 additions and 5651 deletions

View File

@@ -1,9 +1,16 @@
name: Build App
# Manual-only since #11946. The hosted 7 GB runner can no longer build this tree — 19 of
# the last 30 runs died with "The runner has received a shutdown signal" (VM out of
# memory) ~8 min into `next build`, release/v3.8.51 itself included, even with the 10 GB
# swapfile below. Triggered on `push: branches: ["**"]` it painted every branch and every
# PR red while producing an artefact nothing downloads. The bundle is validated where a
# build actually fits:
# - main: ci.yml `Build` (self-hosted omni-build pool) on every merge
# - release/**: nightly-release-green.yml (same pool, continuous)
# Dispatch this workflow by hand when a hosted build artefact is genuinely needed.
on:
workflow_dispatch:
push:
branches: ["**"]
permissions:
contents: read

View File

@@ -609,13 +609,24 @@ jobs:
# Dynamic runner: when the release captain flips the USE_VPS_RUNNER repo var to
# 'true' (scripts/vps/release-runner-up.sh does it after the self-hosted VM is
# online), the heavy jobs run on the dedicated 32-core VPS runners (label
# omni-release) instead of queueing on the 20-concurrent-job hosted pool.
# omni-build) instead of queueing on the 20-concurrent-job hosted pool.
# Safety: fork PRs NEVER reach the self-hosted runner — the expression falls
# back to ubuntu-latest unless the PR head repo is this repository (push /
# dispatch events are own-origin by definition). Any failure path (VM down,
# var unset/false) also falls back to ubuntu-latest.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && (github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository)) && fromJSON('["self-hosted","omni-release"]') || 'ubuntu-latest' }}
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && (github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository)) && fromJSON('["self-hosted","omni-build"]') || 'ubuntu-latest' }}
needs: changes
# The .113 pool runs ONE next-build with room to spare and two at the edge: the
# box has 31 GB and a single next-build peaks at 1416 GB RSS. On 2026-08-28
# 13:50Z the kernel OOM-killed main's build while a PR build ran beside it
# (five Build jobs had been queued by a burst of PRs). Two lanes: main keeps
# its own so a release is never queued behind PR traffic; PR builds serialize
# among themselves. GitHub keeps one running + one pending per group and
# CANCELS older pendings — a cancelled PR build is re-runnable; a dead main
# build costs the publish its artefact and a 40-minute rebuild that OOMs.
concurrency:
group: heavy-build-${{ github.ref == 'refs/heads/main' && 'main' || 'pr' }}
cancel-in-progress: false
if: ${{ github.event_name != 'pull_request' || (needs.changes.outputs.code == 'true' && github.event.pull_request.draft == false) }}
steps:
- uses: actions/checkout@v7
@@ -649,14 +660,14 @@ jobs:
# Keep standalone/node_modules intact: package/electron jobs consume the
# Next-traced standalone tree and must not replace it with root node_modules.
run: |
tar -czf /tmp/e2e-build.tar.gz \
tar -czf "$RUNNER_TEMP/e2e-build.tar.gz" \
--exclude='.build/next/cache' \
.build/next
- name: Upload Next.js build for downstream jobs
uses: actions/upload-artifact@v7
with:
name: next-build
path: /tmp/e2e-build.tar.gz
path: ${{ runner.temp }}/e2e-build.tar.gz
retention-days: 1
package-artifact:
@@ -679,10 +690,14 @@ jobs:
uses: actions/download-artifact@v8
with:
name: next-build
path: /tmp/
# Workspace-relative on purpose: the matrix below includes windows-latest, whose
# default shell is pwsh, where $RUNNER_TEMP is empty (it is $env:RUNNER_TEMP) —
# #11896's first cut broke the Electron smoke on exactly that. A relative path
# works in bash and pwsh alike; hosted workspaces are ephemeral.
path: next-build-artifact
- name: Extract Next.js build artifact
run: |
tar -xzf /tmp/e2e-build.tar.gz
tar -xzf next-build-artifact/e2e-build.tar.gz
# build:cli consumes the downloaded .build/next standalone artifact and assembles dist/;
# it only rebuilds if the downloaded standalone artifact is missing.
- run: npm run build:cli
@@ -770,10 +785,14 @@ jobs:
uses: actions/download-artifact@v8
with:
name: next-build
path: /tmp/
# Workspace-relative on purpose: the matrix below includes windows-latest, whose
# default shell is pwsh, where $RUNNER_TEMP is empty (it is $env:RUNNER_TEMP) —
# #11896's first cut broke the Electron smoke on exactly that. A relative path
# works in bash and pwsh alike; hosted workspaces are ephemeral.
path: next-build-artifact
- name: Extract Next.js build artifact
run: |
tar -xzf /tmp/e2e-build.tar.gz
tar -xzf next-build-artifact/e2e-build.tar.gz
- name: Install Electron dependencies
working-directory: electron
run: npm install --no-audit --no-fund
@@ -960,7 +979,11 @@ jobs:
# 10min was sized before #7114 added the lcov reporter (Codecov/Sonar need it);
# merging 8 shard JSONs + text+json+lcov now takes ~10-12min — three consecutive
# release-tip runs died at exactly 10m as job-timeout "cancelled" (2026-07-15/16).
timeout-minutes: 20
# 30, not 20 (2026-08-29): the informational Codecov upload below hung for the rest of
# the budget on two consecutive main runs (33207760653, 33215115341); the job ended
# `cancelled` and dragged the whole run's conclusion to `cancelled` although every
# blocking job was green. The upload step now has its own ceiling; this is headroom.
timeout-minutes: 30
needs: test-unit
if: ${{ !cancelled() && needs.test-unit.result == 'success' && !contains(github.event.pull_request.labels.*.name, 'hotfix') }}
env:
@@ -1039,6 +1062,10 @@ jobs:
# (if-no-files-found: warn) — Sonar consumes the same file.
- name: Upload coverage to Codecov (informational)
if: always()
# Informational means informational: its own ceiling and continue-on-error, so a
# stalled upload can neither eat the job's budget nor turn a green job cancelled.
timeout-minutes: 5
continue-on-error: true
uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v7.0.0
with:
files: coverage/lcov.info
@@ -1233,10 +1260,14 @@ jobs:
uses: actions/download-artifact@v8
with:
name: next-build
path: /tmp/
# Workspace-relative on purpose: the matrix below includes windows-latest, whose
# default shell is pwsh, where $RUNNER_TEMP is empty (it is $env:RUNNER_TEMP) —
# #11896's first cut broke the Electron smoke on exactly that. A relative path
# works in bash and pwsh alike; hosted workspaces are ephemeral.
path: next-build-artifact
- name: Extract Next.js build artifact
run: |
tar -xzf /tmp/e2e-build.tar.gz
tar -xzf next-build-artifact/e2e-build.tar.gz
# WS4.1: duration-balanced shards (LPT over config/quality/e2e-timings.json).
# Measured skew of plain --shard was 14× (24m47s vs 1m47s) — E2E was the CI
# critical path. The balancer self-verifies completeness and exits non-zero on

View File

@@ -1,7 +1,15 @@
name: DAST smoke (PR)
# PRs into main only since #11946. The job's "Build CLI bundle" step is a backend-only
# `next build`; on the hosted 7 GB runner it fits main's tree (~5.5 min) but dies on
# release/v3.8.51 (VM shutdown ~7 min in, before the server even starts), and because the
# job is continue-on-error the result was a permanently red advisory check on every
# release PR — noise, not signal. DAST coverage for release/** lives on the nightly rail
# (nightly-schemathesis.yml, nightly-llm-security.yml); dispatch this workflow by hand
# to smoke a release branch on demand.
on:
workflow_dispatch:
pull_request:
branches: ["main", "release/**"]
branches: ["main"]
# Runner-cost guard (#8084): the CLI-bundle build alone is 6-11min; a docs-only PR
# cannot change DAST behavior, so skip the whole workflow for pure docs/markdown
# changes. Any code path in the diff still runs the full smoke.

View File

@@ -10,6 +10,11 @@ on:
description: "Release version (e.g., v1.6.8)"
required: true
type: string
publish_npm:
description: "Also run the npm publish leg (turn off when re-attaching desktop assets to a release whose npm package already shipped)"
required: false
default: true
type: boolean
# Least-privilege default: read-only at the top level; each job grants the writes it
# needs (build/release upload assets, publish-npm forwards npm provenance / packages
@@ -76,6 +81,9 @@ jobs:
- uses: actions/checkout@v7
with:
persist-credentials: false
# workflow_dispatch: build the tag being (re)built, not the dispatching branch. On a
# tag push this resolves to the same commit.
ref: ${{ needs.validate.outputs.version }}
- name: Setup Node
uses: actions/setup-node@v7
with:
@@ -161,6 +169,9 @@ jobs:
- uses: actions/checkout@v7
with:
persist-credentials: false
# workflow_dispatch: build the tag being (re)built, not the dispatching branch. On a
# tag push this resolves to the same commit.
ref: ${{ needs.validate.outputs.version }}
- name: Setup Node
uses: actions/setup-node@v7
with:
@@ -347,6 +358,8 @@ jobs:
with:
persist-credentials: false
fetch-depth: 0
# Source archives + SBOM come from the tag being released, not the dispatching branch.
ref: ${{ needs.validate.outputs.version }}
# `merge-multiple` is deliberately OFF. It resolves same-name collisions by ARRIVAL
# ORDER, and the two macOS jobs each emit their own `latest-mac.yml` listing only their
@@ -462,11 +475,20 @@ jobs:
publish-npm:
name: Publish to npm
needs: [validate, release]
# A re-dispatch that only re-attaches desktop assets must not publish the npm package again.
if: ${{ github.event_name != 'workflow_dispatch' || inputs.publish_npm }}
permissions:
# Must be `write`, not `read`: this job calls the reusable npm-publish.yml whose
# `publish` job needs `contents: write` (gh release upload — attach the SBOM, #3874).
# A reusable workflow's job cannot request more permission than the caller grants,
# so a `read` here makes GitHub reject the run at startup (startup_failure).
#
# `actions: read` for the same reason: the called `publish` job downloads the next-build
# artefact and requests it. v3.8.50 (run 33005490476) died at startup with "The nested
# job 'publish' is requesting 'actions: read', but is only allowed 'actions: none'" — and
# because `release` lives in this same workflow, the tag shipped with ZERO assets. Keep
# this block a superset of every job's permissions in npm-publish.yml.
actions: read
contents: write
id-token: write # npm provenance (forwarded to the reusable workflow)
packages: write # publish to npm.pkg.github.com

View File

@@ -10,7 +10,10 @@ permissions:
jobs:
promptfoo-guard:
name: promptfoo — injection guard (block mode, no secret)
runs-on: ubuntu-latest
# #11965: this job runs a backend-only `next build`; the hosted 7 GB runner cannot build
# release/v3.8.51 (VM shutdown ~7 min in), so it targets the box's light pool (`omni-light`:
# two listeners, jobs ≤ ~6 GB). Falls back to hosted when USE_VPS_RUNNER is off.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-light"]')) || 'ubuntu-latest' }}
steps:
- uses: actions/checkout@v7
with:
@@ -46,7 +49,10 @@ jobs:
garak:
name: garak probes (skip without provider secret)
runs-on: ubuntu-latest
# #11965: this job runs a backend-only `next build`; the hosted 7 GB runner cannot build
# release/v3.8.51 (VM shutdown ~7 min in), so it targets the box's light pool (`omni-light`:
# two listeners, jobs ≤ ~6 GB). Falls back to hosted when USE_VPS_RUNNER is off.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-light"]')) || 'ubuntu-latest' }}
# NOTE: the `secrets` context is NOT available in a job-level `if:` — referencing
# it there makes GitHub reject the file on push (startup_failure on every push).
# Map the secret into a job-level env and gate each step on a presence check, so

View File

@@ -68,7 +68,7 @@ jobs:
# this runs on the dedicated VPS runner — clean env (no operator OMNIROUTE_API_KEY,
# no local noauth CLIs => zero machine-specific false positives) and no contention.
# Nightly cron normally finds the var false (VM off) and falls back to hosted.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-release"]')) || 'ubuntu-latest' }}
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-build"]')) || 'ubuntu-latest' }}
env:
JWT_SECRET: ci-nightly-secret-with-sufficient-length-for-validation
API_KEY_SECRET: ci-nightly-api-key-secret-long
@@ -217,7 +217,7 @@ jobs:
# On a push, only run for a push to main — a push to release/* is handled by
# release-green above. Schedule/dispatch always run (they also sweep main).
if: ${{ github.event_name != 'push' || github.ref_name == 'main' }}
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-release"]')) || 'ubuntu-latest' }}
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-build"]')) || 'ubuntu-latest' }}
env:
JWT_SECRET: ci-nightly-secret-with-sufficient-length-for-validation
API_KEY_SECRET: ci-nightly-api-key-secret-long

View File

@@ -78,7 +78,10 @@ jobs:
a11y:
name: A11y axe (nightly, freeze-and-alert)
runs-on: ubuntu-latest
# #11965: this job runs a backend-only `next build`; the hosted 7 GB runner cannot build
# release/v3.8.51 (VM shutdown ~7 min in), so it targets the box's light pool (`omni-light`:
# two listeners, jobs ≤ ~6 GB). Falls back to hosted when USE_VPS_RUNNER is off.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-light"]')) || 'ubuntu-latest' }}
# The Playwright webServer (`start` mode) builds Next via build-next-isolated.mjs and
# boots the standalone server itself (waits on /api/monitoring/health, 15min webServer
# timeout). Unlike the per-PR test-e2e job, this nightly job has no pre-built artifact,

View File

@@ -10,7 +10,10 @@ permissions:
jobs:
schemathesis:
name: Schemathesis — OpenAPI contract fuzz (advisory)
runs-on: ubuntu-latest
# #11965: this job runs a backend-only `next build`; the hosted 7 GB runner cannot build
# release/v3.8.51 (VM shutdown ~7 min in), so it targets the box's light pool (`omni-light`:
# two listeners, jobs ≤ ~6 GB). Falls back to hosted when USE_VPS_RUNNER is off.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && fromJSON('["self-hosted","omni-light"]')) || 'ubuntu-latest' }}
timeout-minutes: 30
steps:
- uses: actions/checkout@v7

View File

@@ -23,11 +23,12 @@ on:
- next
- historic
publish_mode:
description: "staged = npm stage publish (owner approves with 2FA after the staged boot-verify); direct = legacy immediate publish (emergency fallback only)"
description: "auto = publish through npm Trusted Publishing (OIDC, no token, no 2FA prompt — the default); staged = npm stage publish (owner approves with 2FA); direct = legacy token publish (emergency fallback only)"
required: false
default: "staged"
default: "auto"
type: choice
options:
- auto
- staged
- direct
workflow_call:
@@ -62,7 +63,7 @@ jobs:
# mid-"Creating an optimized production build" while v3.8.48 had still fit in 16min.
# This job never runs on `pull_request`, so the fork-safety clause is always true here;
# it is kept verbatim so the expression stays greppable against ci.yml.
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && (github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository)) && fromJSON('["self-hosted","omni-release"]') || 'ubuntu-latest' }}
runs-on: ${{ (vars.USE_VPS_RUNNER == 'true' && (github.event_name != 'pull_request' || github.event.pull_request.head.repo.full_name == github.repository)) && fromJSON('["self-hosted","omni-build"]') || 'ubuntu-latest' }}
outputs:
version: ${{ steps.resolve.outputs.version }}
tag: ${{ steps.resolve.outputs.tag }}
@@ -204,8 +205,11 @@ jobs:
exit 0
fi
RUN=""
# $RUNNER_TEMP, never /tmp: on the .113 pool /tmp is a 12 GB tmpfs (RAM). Parking
# this 1.3 GB artefact there took 2732 min of the 76-min publish job — the
# same bytes upload from disk in 2 min. RUNNER_TEMP is per-runner and on disk.
for candidate in $CANDIDATES; do
if gh run download "$candidate" --repo "$REPO" --name next-build --dir /tmp/next-build 2>/dev/null; then
if gh run download "$candidate" --repo "$REPO" --name next-build --dir "$RUNNER_TEMP/next-build" 2>/dev/null; then
RUN="$candidate"
break
fi
@@ -215,8 +219,8 @@ jobs:
echo "::notice::none of the candidate runs still carries next-build (1-day retention) — falling back to a full build"
exit 0
fi
tar -xzf /tmp/next-build/e2e-build.tar.gz -C .
rm -rf /tmp/next-build
tar -xzf "$RUNNER_TEMP/next-build/e2e-build.tar.gz" -C .
rm -rf "$RUNNER_TEMP/next-build"
if [ -f .build/next/standalone/server.js ]; then
echo "✅ standalone tree restored from CI run $RUN — build:cli will skip next build"
else
@@ -404,8 +408,34 @@ jobs:
fi
npm --version
# Trusted Publishing (OIDC): npm mints a short-lived credential for THIS run from
# GitHub's id-token — no NPM_TOKEN secret, no 2FA prompt, provenance included, and
# it is the bypass npm sanctions now that tokens which skip 2FA are being retired
# (gh.io/npm-gat-bypass2fa-deprecation). Requires the package's Trusted Publisher to
# be configured on npmjs.com (owner: diegosouzapw/OmniRoute, workflow
# npm-publish.yml) and a github-hosted runner — which is why this job exists.
# Without that configuration `npm publish` fails with ENEEDAUTH: re-dispatch with
# publish_mode=staged or direct. Automatic publishing was the flow up to v3.8.48;
# v3.8.49 moved to staged (WS1.3) to keep a leaked token from publishing alone —
# OIDC gives the same guarantee without the manual approve.
- name: Publish to npm (Trusted Publishing / OIDC — automatic)
if: github.event_name != 'workflow_dispatch' || inputs.publish_mode == 'auto'
env:
VERSION: ${{ needs.publish.outputs.version }}
TAG: ${{ needs.publish.outputs.tag }}
run: |
set -euo pipefail
TARBALL="omniroute-${VERSION}.tgz"
test -f "$TARBALL" || { echo "tarball $TARBALL did not arrive from the publish job" >&2; ls -la; exit 1; }
# Deliberately NO NODE_AUTH_TOKEN in this step: npm >= 11.5 detects the GitHub
# OIDC token itself. Always pass --tag explicitly (defense in depth: an older
# VERSION can never claim `@latest`).
npm publish "$TARBALL" --provenance --access public --tag "$TAG" --ignore-scripts
echo "✅ Published omniroute@$VERSION (dist-tag=$TAG) via Trusted Publishing"
- name: Publish to npm (staged — owner approves with 2FA)
if: github.event_name != 'workflow_dispatch' || inputs.publish_mode != 'direct'
# Only on an explicit request now: Trusted Publishing below is the default.
if: github.event_name == 'workflow_dispatch' && inputs.publish_mode == 'staged'
env:
VERSION: ${{ needs.publish.outputs.version }}
TAG: ${{ needs.publish.outputs.tag }}

View File

@@ -61,6 +61,9 @@ jobs:
name: Build (advisory)
needs: changes
# FORK PRs ONLY. build.yml's `Fast Production Build` triggers on `push: branches: ["**"]`
# (#11946, 2026-08-29: build.yml is now workflow_dispatch-only — the hosted runner cannot
# build this tree in any profile, 8/8 recent fork PRs included — so own-origin PRs rely on
# ci.yml `Build` after merge to main and on nightly-release-green for release/**.)
# and runs `build:release` — a superset of this job — so for an own-origin branch this job
# was building the same tree twice. A fork contributor pushes to THEIR repo, so that push
# never fires here, and this is the only pre-merge build signal they get. Measured
@@ -189,8 +192,11 @@ jobs:
.eslintcache
.eslintcache-complexity
key: eslint-${{ runner.os }}-${{ hashFiles('eslint.config.mjs', 'eslint.complexity-ratchets.config.mjs', 'config/quality/eslint-suppressions.json', 'package-lock.json') }}
restore-keys: |
eslint-${{ runner.os }}-
# No restore-keys fallback on purpose (#11600, P-II.1 of the v3.8.50 postmortem): a
# cache built under a different suppressions file / lint config / lockfile reports
# stale per-file verdicts, which is exactly how 215 pre-existing errors stayed
# invisible for a whole cycle. Exact key or a cold full lint (~13 min) — never a
# partial cache from another configuration.
# Security scanners — same hardened install as ci.yml quality-extended
# (gh release download = authenticated, 5000 req/hr; curl to api.github.com
# is rate-limited to 60/hr and silently no-ops when throttled). The blocking
@@ -460,6 +466,12 @@ jobs:
# cache restore saturating the link), while the tests themselves tied, 2m54 vs 2m31. So
# self-hosted is strictly worse here and there is nothing to configure.
runs-on: ubuntu-latest
# A shard finishes in ~10 min. Without a ceiling a hung test process holds the PR for
# GitHub's 6 h default: on 2026-08-28 shard 1/4 sat 64 min without a line of output
# (twice, same spot — a timing race, gone on the third run) while the other three
# shards were long green. 30 min = 3x the normal wall-clock; a shard that needs more
# is a hang, not a slow run, and a fast red with a re-run beats a silent 6 h hold.
timeout-minutes: 30
strategy:
fail-fast: false
matrix:
@@ -524,8 +536,11 @@ jobs:
.eslintcache
.eslintcache-complexity
key: eslint-${{ runner.os }}-${{ hashFiles('eslint.config.mjs', 'eslint.complexity-ratchets.config.mjs', 'config/quality/eslint-suppressions.json', 'package-lock.json') }}
restore-keys: |
eslint-${{ runner.os }}-
# No restore-keys fallback on purpose (#11600, P-II.1 of the v3.8.50 postmortem): a
# cache built under a different suppressions file / lint config / lockfile reports
# stale per-file verdicts, which is exactly how 215 pre-existing errors stayed
# invisible for a whole cycle. Exact key or a cold full lint (~13 min) — never a
# partial cache from another configuration.
- name: ESLint (baseline congelado — warning novo = vermelho)
# lint:json writes the report; --max-warnings 0 keeps no-new-warnings policy.
run: npm run lint:json -- --max-warnings 0