Files
terdut-operator/.gitea/workflows/ci.yaml
T
Niklas Ye b4ccdb09d5
Release / test (push) Successful in 2m48s
CI / chart (push) Successful in 1s
CI / security (push) Successful in 1m3s
CI / test (push) Successful in 2m5s
Release / chart (push) Successful in 4s
Release / image (push) Successful in 7m6s
Release / scan-image (push) Failing after 33s
Stage 5: installer chart + release infra, kind e2e pass through the chart
Chart (charts/terdut-operator) generated via kubebuilder's own helm/v2-alpha
plugin from config/'s kustomize output -- CRDs + manager Deployment/RBAC
come from the same markers every other stage already generates, one source
of truth. Hand-added on top: the optional terdutServer values block
(DESIGN.md §10's "helm install and get a server" path, off by default) and
the release-skill plumbing -- .release.conf, release-vars/helm-lint/push/
helm-package/helm-push/release Makefile targets, .gitea/workflows/release.yaml
(test -> image/chart -> scan-image) -- mirroring terdut-server's own shape
(registry/namespace convention, multi-arch buildx push, trivy/govulncheck/
gitleaks scans). ci.yaml gains security and chart jobs to match.

Two real issues caught while wiring this, fixed before either shipped:
- Dockerfile's builder stage didn't pin --platform=$BUILDPLATFORM, which
  would have made a multi-arch release build fail outright on this org's
  runners (no binfmt registration) -- same fix terdut-server's own
  Dockerfile already needed for the same reason.
- govulncheck found one real, reachable finding: google.golang.org/grpc
  v1.82.1 (transitive via controller-runtime's otel exporter), fixed by
  bumping to v1.83.1.

Full golden-path kind e2e pass, this time through `helm install` rather than
raw kustomize: TerdutServer (real terdut-server v0.33.0 image) -> TerdutTeam
-> one of each child kind, each confirmed Ready and then independently
confirmed against terdut-server's own API from inside the cluster (not just
the operator's own status). Deleted every CR in reverse order and confirmed
server-side cleanup the same independent way for all three child kinds, the
team, and the server. No new bugs found -- Stage 1's own kind pass already
caught what a real cluster catches that envtest can't.

Also dropped the kubebuilder helm plugin's default .github/workflows/
scaffold, same as Stage 0 already did for the main scaffold: this org runs
on Gitea, not GitHub.

Not done here, deliberately: an actual tagged release. release-preflight
found no terdut-operator/ entry under Ryuvia/charts yet to bump -- that
one-time wrapper bootstrap is a decision about deploying this operator for
real, not a side effect of finishing this stage.

make fmt lint test helm-lint build all clean.
2026-10-01 14:47:10 +02:00

132 lines
5.2 KiB
YAML

name: CI
# Same shape as terdut-server's ci.yaml: the release workflow gates a tag, which is
# late, so this runs the same checks on the way in instead.
#
# push is scoped to main rather than all branches so a branch pushed as part of a pull
# request is not checked twice.
#
# No actions/checkout, deliberately -- same reason as terdut-server: the runner image's
# `nodejs` package predates ES2022 static initialiser blocks, so actions/checkout@v4
# dies with `SyntaxError: Unexpected token '{'` before running. Cloning with git
# directly avoids JS actions entirely. This repo is public, so the clone needs no
# credential.
#
# `${{ }}` values are passed through `env:` and referenced as quoted shell variables --
# a ref name is attacker-influenced by anyone who can push a branch or open a PR.
on:
push:
branches: [main]
pull_request:
concurrency:
group: ci-${{ github.ref }}
cancel-in-progress: true
env:
REPO_URL: https://git.ryuvia.com/niklas/terdut-operator.git
jobs:
# `make fmt lint test` is exactly what a developer runs locally, so a green job here
# and a green working copy mean the same thing by construction. `test` also drives
# controller-gen/setup-envtest (via the Makefile's own `manifests generate ... test`
# chain), which fetches the envtest kube-apiserver/etcd binaries from GitHub Releases
# (setup-envtest v0.25's only source -- the legacy GCS kubebuilder-tools bucket 403s
# now for every version tried, confirmed 2026-09-30, no fallback there).
#
# This used to fail in CI: TLS handshake timeout reaching github.com from inside
# this container (same symptom terdut-server's ci.yaml documents for
# get.helm.sh/github.com), fetching the envtest kube-apiserver/etcd binaries from
# GitHub Releases (setup-envtest v0.25's only source -- the legacy GCS
# kubebuilder-tools bucket 403s now for every version tried, no fallback there).
# History, in case it recurs:
# - Ryuvia/charts' act-runner/templates/networkpolicy.yaml is the only
# NetworkPolicy in the cluster and is deny-ingress only -- ruled out, no
# in-repo egress rule governs this.
# - Running this job on the bare runner host instead of in a container fails
# earlier and differently ("go: command not found", no Go there) -- ruled out.
# - The act-runner dind sidecar's MTU mismatch (1450 pod vs. 1500 Docker-bridge
# default) was real but an initial fix for it (Ryuvia/charts#272) did not
# resolve this specific failure when tested in isolation -- see Ryuvia/charts#271
# for that round's write-up.
# - A second attempt at the MTU fix, 2026-09-30, did resolve it: `setup-envtest`
# now fetches envtest-v1.37.0-linux-amd64.tar.gz from github.com in ~4s and
# `test` passes. Confirmed via the actual job log, not just the exit code.
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 }}
HEAD_SHA: ${{ github.event.pull_request.head.sha }}
run: |
if [ -n "$HEAD_SHA" ]; then
# A pull_request ref_name is "<n>/merge", which is not a fetchable branch.
git clone "$REPO_URL" .
git checkout -q "$HEAD_SHA"
else
git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
fi
- name: Format, lint and test
run: make fmt lint test
# Runs on every push and pull request, unlike the image scan, which needs something
# published to scan and so lives in release.yaml -- same split as terdut-server's.
# govulncheck reads the source and its module graph, gitleaks reads the working
# tree; neither sees what the other does.
security:
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 }}
HEAD_SHA: ${{ github.event.pull_request.head.sha }}
run: |
if [ -n "$HEAD_SHA" ]; then
git clone "$REPO_URL" .
git checkout -q "$HEAD_SHA"
else
git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
fi
- name: Go vulnerability scan (govulncheck)
run: make security-go
- name: Secret scan (gitleaks)
run: make security-secrets
# Host mode, no `container:`: helm is baked into the runner image, and a container
# job could not install it -- get.helm.sh is unreachable from the dind bridge, same
# reason terdut-server's own chart job runs on the host.
chart:
runs-on: ubuntu-latest
steps:
- name: Checkout
env:
REF_NAME: ${{ github.ref_name }}
HEAD_SHA: ${{ github.event.pull_request.head.sha }}
run: |
if [ -n "$HEAD_SHA" ]; then
git clone "$REPO_URL" .
git checkout -q "$HEAD_SHA"
else
git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
fi
- name: Lint and render the chart
run: make helm-lint