Files
Niklas Ye a8ee742533
CI / test (push) Successful in 4s
CI / chart (push) Successful in 1s
CI / security (push) Successful in 11s
Give release.yaml's test job the database ci.yaml already has
v0.11.0 was tagged and published nothing: release.yaml runs the same
`make fmt lint test` as ci.yaml, and the Postgres service the suite now
needs was added to ci.yaml alone. Its test job failed on the missing
TERDUT_TEST_DSN, and binaries, image, chart and scan-image are all
downstream of it, so they skipped. The tag stays -- a published tag is
immutable and moving one is how this repo ran v0.4.0 for ten days while
every artifact said v0.3.0 -- so the fix is the next version.

The two workflows carry their own copies of this block because a service
container cannot be factored into the Makefile the way the checks are.
That is the second copy, and the reason this failed: the gate is one
target, but what the gate needs to run is declared per workflow.
2026-09-20 11:12:41 +02:00

245 lines
10 KiB
YAML

name: Release
# Checkout, interpolation and caching conventions match ci.yaml -- see the header there
# for why there are no JS actions and why every `${{ }}` goes through `env:`.
#
# There is no upload-artifact/download-artifact equivalent here (both are JS actions, and
# this Gitea has no artifact store wired up), so the job that builds the binaries is also
# the job that publishes them. Nothing is handed between jobs at all.
on:
push:
tags:
- 'v*'
workflow_dispatch:
# A tag is not normally re-pushed, so this mostly matters when one is force-moved during
# a botched release -- the superseded run stops holding runner slots.
concurrency:
group: release-${{ github.ref }}
cancel-in-progress: true
env:
REPO_URL: https://git.ryuvia.com/niklas/terdut-server.git
API: https://git.ryuvia.com/api/v1/repos/niklas/terdut-server
REGISTRY: git.ryuvia.com
IMAGE: git.ryuvia.com/niklas/terdut-server
jobs:
# Gates every publishing job below. A tag that fails here publishes nothing: the
# binaries, the image and the chart are all downstream of it.
test:
runs-on: ubuntu-latest
container:
image: golang:1.26.6-bookworm
volumes:
- go-mod-cache:/go/pkg/mod
- go-build-cache:/root/.cache/go-build
- gobin-cache:/go/bin
# The same database ci.yaml's test job gets, for the same reason: `make test` needs a
# real Postgres and fails without TERDUT_TEST_DSN rather than skipping. This job is
# the gate every publishing job below hangs off, so it has to be able to run the
# suite -- v0.11.0 was tagged with the service here missing and published nothing.
services:
postgres:
image: postgres:17-alpine
env:
POSTGRES_USER: terdut
POSTGRES_PASSWORD: terdut
POSTGRES_DB: terdut_test
options: >-
--health-cmd "pg_isready -U terdut -d terdut_test"
--health-interval 5s
--health-timeout 5s
--health-retries 12
env:
TERDUT_TEST_DSN: postgres://terdut:terdut@postgres:5432/terdut_test?sslmode=disable
steps:
- name: Checkout
env:
REF_NAME: ${{ github.ref_name }}
run: git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
# Same gate as ci.yaml, and the same one a developer runs. See the Makefile for why
# each check is there; restating it here is how the two drift apart.
- name: Format, vet and test
run: make fmt lint test
binaries:
needs: test
runs-on: ubuntu-latest
container:
image: golang:1.26.6-bookworm
volumes:
- go-mod-cache:/go/pkg/mod
- go-build-cache:/root/.cache/go-build
- gobin-cache:/go/bin
steps:
- name: Checkout
env:
REF_NAME: ${{ github.ref_name }}
run: git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
# Compiling is the Makefile's; uploading is not. `make binaries` is runnable on a
# laptop, while the step below needs a token and the Gitea release API, which is
# this workflow's business and nothing a developer wants a target for.
- name: Build every target
env:
REF_NAME: ${{ github.ref_name }}
run: make binaries VERSION="$REF_NAME"
# Creating the release is made idempotent rather than assumed-new: a re-run of a
# failed release must not die on the release that already exists. Assets are
# replaced the same way, so a re-run repairs a partial upload.
- name: Publish the release
env:
REF_NAME: ${{ github.ref_name }}
TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
auth="Authorization: token $TOKEN"
body=$(curl -sf -H "$auth" "$API/releases/tags/$REF_NAME" || true)
if [ -z "$body" ]; then
body=$(curl -sf -X POST -H "$auth" -H 'Content-Type: application/json' \
-d "{\"tag_name\":\"$REF_NAME\",\"name\":\"$REF_NAME\"}" \
"$API/releases")
fi
# The release object serialises `id` first, so the first match is the release's
# own id and not one of the nested author/asset ids.
release_id=$(printf '%s' "$body" | grep -o '"id":[0-9]*' | head -1 | cut -d: -f2)
[ -n "$release_id" ] || { echo "::error::could not determine release id"; exit 1; }
echo "release id $release_id"
for f in dist/*; do
name=$(basename "$f")
# Drop an existing asset of the same name first: Gitea happily stores two
# attachments with one name, and the updater matches by name.
old=$(curl -sf -H "$auth" "$API/releases/$release_id/assets" \
| tr '}' '\n' | grep "\"name\":\"$name\"" \
| grep -o '"id":[0-9]*' | head -1 | cut -d: -f2 || true)
if [ -n "$old" ]; then
curl -sf -X DELETE -H "$auth" "$API/releases/$release_id/assets/$old" || true
fi
echo "uploading $name"
curl -sf -X POST -H "$auth" -F "attachment=@$f" \
"$API/releases/$release_id/assets?name=$name" > /dev/null
done
# Host mode on purpose (no `container:`): this is the only context with a Docker CLI
# pointed at the dind daemon. A `container:` job would sit on the dind bridge with no
# docker socket at all.
image:
needs: test
runs-on: ubuntu-latest
steps:
- name: Checkout
env:
REF_NAME: ${{ github.ref_name }}
run: git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
- name: Log in to the registry
env:
TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: echo "$TOKEN" | docker login "$REGISTRY" -u niklas --password-stdin
# buildx setup, the platform list and why there is no QEMU all live on the `push`
# target now, so the same command publishes from a laptop and from here.
- name: Build and push
env:
REF_NAME: ${{ github.ref_name }}
run: make push VERSION="$REF_NAME"
# Also host mode: helm is baked into the runner image, and a `container:` job could not
# install it -- get.helm.sh is unreachable from the dind bridge.
chart:
needs: test
runs-on: ubuntu-latest
steps:
- name: Checkout
env:
REF_NAME: ${{ github.ref_name }}
run: git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
# This job is the only thing that publishes the chart, which is what keeps the
# published metadata honest. There used to be a second publisher on every charts/**
# push to main, and the two raced for the same chart version with different answers:
# this one stamps version and appVersion from the tag, that one took Chart.yaml
# verbatim, where appVersion is the hardcoded "latest". Whichever landed first won,
# so the metadata of a release depended on which runner was quicker -- chart 0.9.0
# went out on 2026-08-08 reading appVersion "latest" that way.
#
# It could not be fixed by making both agree: the tag is pushed after the branch, so
# a workflow triggered by the main push cannot know the version it is about to be
# tagged with. One publisher, triggered by the tag.
#
# The cost is that the chart only ships with an app release. That is no real loss --
# `make helm-package` derives the chart version from the tag, so a chart-only change
# has no version of its own to be released under anyway. Chart fixes ride the next
# tag.
- name: Refuse a non-version tag
env:
REF_NAME: ${{ github.ref_name }}
run: |
set -eu
if ! echo "$REF_NAME" | grep -qE '^v[0-9]'; then
echo "::error::refusing to publish a chart for non-version tag ${REF_NAME}"
exit 1
fi
# Render before publishing. Until 2026-09-01 this job packaged and pushed without
# linting, so a template that did not compile reached the registry and was found by
# Flux instead.
- name: Lint and render the chart
run: make helm-lint
# The version and appVersion are no longer sed'd into Chart.yaml before packaging:
# `helm package --version --app-version` sets both from the tag without mutating the
# tree mid-build, which is what the rest of the release process already assumed
# happened. The isolated helm repo list moved onto the targets with them.
- name: Package and push
env:
REF_NAME: ${{ github.ref_name }}
TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: |
set -eu
echo "$TOKEN" | helm registry login "$REGISTRY" -u niklas --password-stdin
make helm-package helm-push VERSION="$REF_NAME"
# Host mode, like image and chart: this needs a docker daemon to run trivy in, and a
# `container:` job would sit on the dind bridge with none.
#
# It scans the pushed image rather than a locally built one, because trivy cannot read a
# local image on this runner -- Talos has no docker socket and the dind sidecar shares no
# filesystem with the job -- so it pulls from the registry. That is also why this runs
# after `image` rather than gating it: a red scan does not unpublish anything.
#
# What a red scan means is therefore not "the release failed" but "do not bump the wrapper
# chart in Ryuvia/charts to this version". The image and chart are already published by
# the time this runs, and deliberately so -- this pipeline does not deploy.
#
# riksdata and rd-web have had this since they were set up; terdut-server went without any
# image scanning until 2026-09-02, so every release before v0.9.4 was published with no
# CVE check at all.
scan-image:
needs: image
runs-on: ubuntu-latest
steps:
- name: Checkout
env:
REF_NAME: ${{ github.ref_name }}
run: git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
# Credentials are passed even though these packages are anonymously pullable -- that
# is a property of the personal namespace this publishes to, not something a release
# should depend on staying true.
- name: Scan the pushed image (trivy)
env:
TRIVY_USERNAME: niklas
TRIVY_PASSWORD: ${{ secrets.REGISTRY_TOKEN }}
REF_NAME: ${{ github.ref_name }}
run: make security-image VERSION="$REF_NAME"