Files
PSProxmoxVE/.github/workflows/package-currency.yml
T
goodolclint-claude[bot] b90791e2bf docs: migrate DECISIONS.md to house-format ADRs, and retire the review folder (#131)
D001-D021 become ADR 0001-0021 in docs/decisions/, one decision per file.
D017's PESTER_VERSION amendment was a second decision in one entry and becomes
ADR 0022. ADR 0023 records the migration and reverses the lane2-change-plan
ruling that deliberately kept DECISIONS.md until the CI lane work landed.

DECISIONS.md is reduced to a stub with a D-to-ADR redirect table, so the four
released CHANGELOG entries and older issue bodies that cite it degrade to a
redirect rather than a dead reference.

docs/review/ and docs/lane2-change-plan.md are deleted (ADR 0024). Of 91
findings, 83 were resolved and six of the seven still open were already GitHub
issues; F021 was the exception and is now #130.

CLAUDE.md's Key Conventions list gains the two rules it was missing and becomes
the checklist, with the ADRs carrying rationale.
2026-09-02 14:49:07 +00:00

299 lines
11 KiB
YAML

name: PVE Package Currency
# Lane 2 of the two-lane CI split (see docs/decisions/0017-ci-runs-two-lanes-a-pinned-gating-lane-and-a-report-only-currency-lane.md).
#
# Lane 1 (integration-tests.yml) pins the nested nodes to what the ISO ships
# and never upgrades — that pin is what makes it a stable merge gate.
# This lane does the opposite: it dist-upgrades the nested nodes to current
# PVE, reboots onto the new kernel, records the package set, and runs the full
# suite against it.
#
# REPORT-ONLY. Test failures here do NOT fail the job — this lane never gates
# a merge, and a red weekly cron nobody can action becomes noise. The signal is
# the recorded package set and (from commit 3) the rolling issue, not the check
# colour. A failure of the lane's own machinery — provisioning, upgrade, reboot
# — still fails the job, because that means the lane learned nothing.
#
# Shares the `integration-tests` concurrency group: both lanes drive the same
# nested VMIDs on the same parent node, so they must never run at once.
concurrency:
group: integration-tests
cancel-in-progress: false
on:
schedule:
# Mondays 06:00 UTC. Weekly is deliberate: this lane exists to notice
# upstream drift, and PVE's no-subscription repo does not move hourly.
- cron: '0 6 * * 1'
workflow_dispatch:
inputs:
keep_vms:
description: 'Skip cleanup so the nested PVE VMs survive for inspection'
required: false
type: boolean
default: false
# packages: write is granted per-job to container-image, the only job that
# pushes. The self-hosted jobs get read-only tokens: they SSH into freshly
# upgraded nodes and should not be able to overwrite published GHCR images.
permissions:
contents: read
packages: read
env:
SCRIPTS_DIR: tests/infrastructure/scripts
TEST_IMAGE: ghcr.io/goodolclint/psproxmoxve-integration
CACHE_DIR: /opt/pve-integration
PVE_VERSIONS: '9'
PVE9_ISO: proxmox-ve_9.2-1.iso
jobs:
# ── Build module artifact (GitHub-hosted) ────────────────────────
build:
runs-on: ubuntu-latest
timeout-minutes: 10
steps:
- uses: actions/checkout@v7
- name: Setup .NET
uses: actions/setup-dotnet@v6
with:
dotnet-version: '10.0.x'
- name: Build module
run: dotnet publish src/PSProxmoxVE/PSProxmoxVE.csproj --configuration Release --framework netstandard2.0 --output ./publish/netstandard2.0
- name: Clean publish output
run: rm -f ./publish/netstandard2.0/*.deps.json
- name: Upload module artifact
uses: actions/upload-artifact@v7
with:
name: module-currency
path: ./publish/netstandard2.0/
# ── Build test container image (GitHub-hosted) ───────────────────────
container-image:
runs-on: ubuntu-latest
timeout-minutes: 20
permissions:
contents: read
packages: write
steps:
- uses: actions/checkout@v7
- name: Log in to GHCR
uses: docker/login-action@v4
with:
registry: ghcr.io
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Build and push
uses: docker/build-push-action@v7
with:
context: .
file: tests/Dockerfile.test
target: dev-infra
push: true
tags: ${{ env.TEST_IMAGE }}:${{ github.sha }}
# ── Provision, dist-upgrade, reboot (self-hosted) ────────────────────
provision:
needs: [build, container-image]
runs-on: psproxmoxve
timeout-minutes: 60
container:
image: ghcr.io/goodolclint/psproxmoxve-integration:${{ github.sha }}
credentials:
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
volumes:
- /opt/pve-integration:/opt/pve-integration
steps:
# persist-credentials: false on the self-hosted jobs — nothing downstream
# uses git, and the token would otherwise be written into .git/config on
# a runner that outlives the job's container.
- uses: actions/checkout@v7
with:
persist-credentials: false
- name: Provision PVE instances and dist-upgrade
shell: bash
env:
PVE_ENDPOINT: ${{ secrets.PVE_ENDPOINT }}
PVE_API_TOKEN: ${{ secrets.PVE_API_TOKEN }}
PVE_TARGET_NODE: ${{ vars.PVE_TARGET_NODE }}
PVE_PASSWORD: ${{ secrets.PVE_TEST_PASSWORD }}
STORAGE_VM_FQDN: ${{ vars.STORAGE_VM_FQDN }}
TF_VAR_disk_storage: ${{ vars.DISK_STORAGE }}
TF_VAR_iso_storage: ${{ vars.ISO_STORAGE }}
TF_VAR_network_bridge: ${{ vars.NETWORK_BRIDGE }}
TF_VAR_pool_id: ${{ vars.POOL_ID }}
# The one line that separates this lane from lane 1.
PVE_DIST_UPGRADE: '1'
run: bash ${SCRIPTS_DIR}/run-integration.sh provision
# success() not always(): when provisioning succeeded the package file
# must exist, and if-no-files-found: error enforces that invariant. When
# provisioning failed the job is already red, and a missing-artifact
# error would only bury the real cause.
- name: Upload recorded package set
if: success()
uses: actions/upload-artifact@v7
with:
name: pve-package-set
path: ${{ env.CACHE_DIR }}/work/*-packages.txt
if-no-files-found: error
# ── Integration tests against the upgraded nodes ─────────────────────
test:
needs: [build, provision]
runs-on: psproxmoxve
timeout-minutes: 20
container:
image: ghcr.io/goodolclint/psproxmoxve-integration:${{ github.sha }}
credentials:
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
volumes:
- /opt/pve-integration:/opt/pve-integration
outputs:
suite_rc: ${{ steps.suite.outputs.suite_rc }}
strategy:
fail-fast: false
matrix:
pve_version: ['9']
steps:
- uses: actions/checkout@v7
with:
persist-credentials: false
- name: Download module artifact
uses: actions/download-artifact@v8
with:
name: module-currency
path: ./publish/netstandard2.0/
# Report-only applies to TEST failures, not to the lane's own machinery.
# cmd_test returns 3 for genuine Pester failures and 4 when it cannot
# reach or authenticate to a node. A blanket continue-on-error would
# swallow 4 as well — and an unreachable node is exactly the symptom of a
# reboot gone wrong, so the lane would go green having learned nothing.
- name: Run integration tests (PVE ${{ matrix.pve_version }})
id: suite
shell: bash
env:
PVE_PASSWORD: ${{ secrets.PVE_TEST_PASSWORD }}
MODULE_ARTIFACT: ./publish/netstandard2.0
run: |
set +e
bash ${SCRIPTS_DIR}/run-integration.sh test ${{ matrix.pve_version }}
rc=$?
echo "suite_rc=${rc}" >> "$GITHUB_OUTPUT"
if [ "${rc}" -eq 0 ]; then
echo "Suite passed against current PVE."
elif [ "${rc}" -eq 3 ]; then
echo "::warning title=Suite failed against current PVE::Report-only - tests failed (rc=3). See the uploaded results."
else
echo "::error title=Currency lane machinery failed::run-integration.sh exited ${rc}, which is not a test failure. The lane learned nothing about package currency."
fi
[ "${rc}" -eq 0 ] || [ "${rc}" -eq 3 ]
- name: Diagnose cluster state
if: always() && steps.suite.outputs.suite_rc != '0'
shell: bash
env:
PVE_PASSWORD: ${{ secrets.PVE_TEST_PASSWORD }}
run: bash ${SCRIPTS_DIR}/diagnose-cluster.sh ${{ matrix.pve_version }}
- name: Upload test results
if: always()
uses: actions/upload-artifact@v7
with:
name: currency-test-results-pve${{ matrix.pve_version }}
path: TestResults/
# ── Report: diff the package set, keep one rolling issue ─────────────
report:
needs: [provision, test]
# !cancelled() not always(): always() is true on cancellation too, and a
# run the operator aborted must not publish a baseline or edit the issue.
if: ${{ !cancelled() && needs.provision.result == 'success' }}
runs-on: ubuntu-latest
timeout-minutes: 10
permissions:
contents: write # push the ci/package-baseline data branch
issues: write # maintain the rolling issue
steps:
# Credentials persist deliberately: this job pushes the data branch.
# No pull-requests: write — the baseline goes to an unprotected data
# branch, not a PR, so nothing here can reach main or the review agent.
- uses: actions/checkout@v7
with:
fetch-depth: 1
- name: Download recorded package set
uses: actions/download-artifact@v8
with:
name: pve-package-set
path: packages/
- name: Report package currency
shell: bash
env:
GH_TOKEN: ${{ github.token }}
SUITE_RC: ${{ needs.test.outputs.suite_rc }}
run: |
# Empty means the suite job was skipped or cancelled — that is not the
# same claim as "failed against current PVE".
# 3 is the only rc that means "tests ran and failed". Anything else
# non-zero means the suite never produced a verdict, and the report
# must not claim the module failed against current PVE.
if [ -z "${SUITE_RC:-}" ]; then
outcome=not-run
elif [ "${SUITE_RC}" = "0" ]; then
outcome=success
elif [ "${SUITE_RC}" = "3" ]; then
outcome=failure
else
outcome=inconclusive
fi
bash "${SCRIPTS_DIR}/report-package-currency.sh" packages "${outcome}"
# ── Cleanup: destroy all VMs (always runs) ──────────────────────
cleanup:
needs: [provision, test]
if: always() && needs.provision.result != 'skipped' && !inputs.keep_vms
runs-on: psproxmoxve
timeout-minutes: 15
container:
image: ghcr.io/goodolclint/psproxmoxve-integration:${{ github.sha }}
credentials:
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
volumes:
- /opt/pve-integration:/opt/pve-integration
steps:
- uses: actions/checkout@v7
with:
persist-credentials: false
- name: Cleanup PVE instances
shell: bash
env:
PVE_ENDPOINT: ${{ secrets.PVE_ENDPOINT }}
PVE_API_TOKEN: ${{ secrets.PVE_API_TOKEN }}
PVE_TARGET_NODE: ${{ vars.PVE_TARGET_NODE }}
PVE_PASSWORD: ${{ secrets.PVE_TEST_PASSWORD }}
STORAGE_VM_FQDN: ${{ vars.STORAGE_VM_FQDN }}
TF_VAR_disk_storage: ${{ vars.DISK_STORAGE }}
TF_VAR_iso_storage: ${{ vars.ISO_STORAGE }}
TF_VAR_network_bridge: ${{ vars.NETWORK_BRIDGE }}
TF_VAR_pool_id: ${{ vars.POOL_ID }}
run: bash ${SCRIPTS_DIR}/run-integration.sh force-cleanup