G2: reusable lint workflows for Ansible roles, Helm charts and Dockerfiles

Three workflow_call workflows that run in the shared CI tools image (G1):
- reusable-ansible-role-lint: ansible-lint (which includes yamllint's
  rules) on a copy of the role named as the playbooks use it, with the
  repo name also resolvable for galaxy's skeleton tests/test.yml.
  role-name is a warning; .ansible-lint-ignore baselines are honoured.
- reusable-helm-lint: helm lint, then helm template | kubeconform for
  Kubernetes 1.31.2, with optional extra values files.
- reusable-dockerfile-lint: hadolint on every Dockerfile, threshold
  warning by default.

docs/ci-cd/adding-ci.md has caller templates, inputs, the baseline
command, and what was and wasn't tested.

Tested locally by running each workflow's steps verbatim in a local build
of the tools image: ansible-role-gitea fails as-is (48 findings), passes
with a baseline and fails on a new bad task; helm-jamulus passes, a
missing-value template and an invalid Service fail; the tools Dockerfile
passes and an unpinned one fails. Not yet run on the runner, and
cross-repo uses: is untested. No callers added, so no runs start.

Refs DevOps/software-infrastructure#50

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
2026-11-14 18:49:25 -05:00
co-authored by Claude Opus 5.5
parent ef693ffd97
commit 1dab77645e
4 changed files with 354 additions and 0 deletions
@@ -0,0 +1,71 @@
# Reusable lint for an Ansible role repo. DevOps/software-infrastructure#50 (G2).
# Usage: docs/ci-cd/adding-ci.md.
#
# Runs ansible-lint in the shared CI tools image (G1, #49), so nothing is
# installed at run time. ansible-lint also applies yamllint's rules (yaml[*]),
# so one baseline covers both. Repo files are honoured if present:
# .yamllint yamllint settings ansible-lint uses for yaml[*]
# .ansible-lint ansible-lint config
# .ansible-lint-ignore baseline of known issues, so the check passes today
# and fails on new problems. Generate it with
# `ansible-lint --generate-ignore`.
name: reusable-ansible-role-lint
on:
workflow_call:
inputs:
tools-image:
description: CI tools image, pinned by version.
type: string
default: gitea.eom.dev/devops/ci-tools:1.0.0
role-name:
description: >-
Name the playbooks use for this role (e.g. ericomeehan.gitea).
Default: "ericomeehan." plus the repo name without "ansible-role-".
type: string
default: ""
ansible-lint-args:
description: Extra ansible-lint arguments, e.g. "--profile production".
type: string
default: ""
permissions:
contents: read
jobs:
lint:
runs-on: ubuntu-latest
timeout-minutes: 15
container:
image: ${{ inputs.tools-image }}
steps:
- uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0
- name: ansible-lint
env:
ROLE_NAME: ${{ inputs.role-name }}
REPO: ${{ github.repository }}
EXTRA_ARGS: ${{ inputs.ansible-lint-args }}
run: |
set -euo pipefail
name="${ROLE_NAME:-ericomeehan.${REPO##*/ansible-role-}}"
# ansible-lint takes the role's name from its directory (symlinks are
# resolved), so lint a copy in a directory with the name the
# playbooks use. Role repos are small.
roles="${RUNNER_TEMP:-/tmp}/roles"
rm -rf "${roles:?}/$name"
mkdir -p "$roles/$name"
cp -a . "$roles/$name/"
# The galaxy skeleton's tests/test.yml names the role by its repo
# name, so make that resolve too (syntax-check can't be skipped).
repo_name="${REPO##*/}"
if [ "$repo_name" != "$name" ]; then ln -sfn "$roles/$name" "$roles/$repo_name"; fi
export ANSIBLE_ROLES_PATH="$roles"
cd "$roles/$name"
echo "Linting role as ${name} with $(ansible-lint --version | head -1)"
# role-name is a warning: the roles keep galaxy's skeleton meta and
# are named ericomeehan.<name> by the playbooks, which galaxy's
# naming rule rejects. Everything else counts.
# shellcheck disable=SC2086 # EXTRA_ARGS is a list of flags
ansible-lint --nocolor -f pep8 --warn-list role-name $EXTRA_ARGS
@@ -0,0 +1,52 @@
# Reusable Dockerfile lint (hadolint). DevOps/software-infrastructure#50 (G2).
# Usage: docs/ci-cd/adding-ci.md. A repo's .hadolint.yaml is honoured.
name: reusable-dockerfile-lint
on:
workflow_call:
inputs:
tools-image:
description: CI tools image, pinned by version.
type: string
default: gitea.eom.dev/devops/ci-tools:1.0.0
dockerfiles:
description: "Space-separated Dockerfile paths. Default: every Dockerfile* and *.Dockerfile in the repo."
type: string
default: ""
failure-threshold:
description: Lowest hadolint severity that fails the job (error, warning, info, style).
type: string
default: warning
permissions:
contents: read
jobs:
lint:
runs-on: ubuntu-latest
timeout-minutes: 10
container:
image: ${{ inputs.tools-image }}
steps:
- uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0
- name: hadolint
env:
DOCKERFILES: ${{ inputs.dockerfiles }}
THRESHOLD: ${{ inputs.failure-threshold }}
run: |
set -euo pipefail
files=()
if [ -n "$DOCKERFILES" ]; then
for f in $DOCKERFILES; do files+=("$f"); done
else
while IFS= read -r f; do files+=("$f"); done < <(
find . -path ./.git -prune -o -type f \( -name 'Dockerfile*' -o -name '*.Dockerfile' \) -print | sort)
fi
if [ "${#files[@]}" -eq 0 ]; then
echo "::error::No Dockerfiles found. Set the dockerfiles input."
exit 1
fi
printf 'Linting %s\n' "${files[@]}"
hadolint --no-color --failure-threshold "$THRESHOLD" "${files[@]}"
+72
View File
@@ -0,0 +1,72 @@
# Reusable lint and render check for a Helm chart. DevOps/software-infrastructure#50 (G2).
# Usage: docs/ci-cd/adding-ci.md.
#
# helm lint, then helm template | kubeconform against the cluster's Kubernetes
# version, in the shared CI tools image (G1, #49).
name: reusable-helm-lint
on:
workflow_call:
inputs:
tools-image:
description: CI tools image, pinned by version.
type: string
default: gitea.eom.dev/devops/ci-tools:1.0.0
chart-path:
description: Path to the chart directory (the one with Chart.yaml).
type: string
default: "."
kube-version:
description: Kubernetes version to render and validate for. Keep it at the cluster's.
type: string
default: "1.31.2"
values-files:
description: Space-separated values files, relative to the chart, rendered in addition to the defaults.
type: string
default: ""
strict:
description: Run helm lint with --strict (warnings fail).
type: boolean
default: false
permissions:
contents: read
jobs:
lint:
runs-on: ubuntu-latest
timeout-minutes: 15
container:
image: ${{ inputs.tools-image }}
steps:
- uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0
- name: helm lint and kubeconform
env:
CHART: ${{ inputs.chart-path }}
KUBE_VERSION: ${{ inputs.kube-version }}
VALUES_FILES: ${{ inputs.values-files }}
STRICT: ${{ inputs.strict }}
run: |
set -euo pipefail
cd "$CHART"
if grep -q '^dependencies:' Chart.yaml; then
helm dependency build --skip-refresh . || helm dependency build .
fi
lint=(helm lint . --kube-version "$KUBE_VERSION")
if [ "$STRICT" = "true" ]; then lint+=(--strict); fi
"${lint[@]}"
sets=("")
for f in $VALUES_FILES; do sets+=("$f"); done
for f in "${sets[@]}"; do
extra=()
label="default values"
if [ -n "$f" ]; then extra=(-f "$f"); label="$f"; fi
echo "== render and validate: ${label}"
helm template ci-render . --kube-version "$KUBE_VERSION" "${extra[@]}" \
| kubeconform -strict -summary -ignore-missing-schemas \
-kubernetes-version "${KUBE_VERSION#v}" -
done
+159
View File
@@ -0,0 +1,159 @@
# Adding CI to a repo (G2)
Refs: DevOps/software-infrastructure#50.
Reusable workflows live in this repo under `.gitea/workflows/reusable-*.yaml`.
A repo adds a small caller file that points at one of them. Every lint job runs
in the shared CI tools image (`gitea.eom.dev/devops/ci-tools`, G1 #49), so
nothing is installed at run time and versions are the same everywhere.
> **Status (2026-11-14):** written and tested locally, not yet run on the
> runner. The tools image isn't published yet (G1), and the runner is still
> being stabilised (S8 #41). Don't add callers to repos until both are done.
> G5 (#53) and G6 (#54) do the rollout.
## The workflows
| Workflow | For | Runs |
|---|---|---|
| `reusable-ansible-role-lint.yaml` | `ansible-role-*` repos | `ansible-lint`, which also applies yamllint's rules |
| `reusable-helm-lint.yaml` | Helm chart repos | `helm lint`, then `helm template` piped to `kubeconform` |
| `reusable-dockerfile-lint.yaml` | Repos with a Dockerfile | `hadolint` |
| `reusable-secret-scan.yaml` | Every repo | `gitleaks`, see [secret-scanning.md](secret-scanning.md) |
## Caller templates
Put one file per check in the repo's `.gitea/workflows/`. Pin the reusable
workflow to `@main` for now. Once these have releases (P8), pin a tag.
Every caller should have a `concurrency` group and, where it helps, path
filters, so a burst of pushes doesn't queue many runs on the shared runner.
### Ansible role
```yaml
name: lint
on:
push:
branches: [main]
pull_request:
concurrency:
group: lint-${{ github.ref }}
cancel-in-progress: true
jobs:
ansible:
uses: DevOps/software-infrastructure/.gitea/workflows/reusable-ansible-role-lint.yaml@main
```
| Input | Default | Notes |
|---|---|---|
| `tools-image` | `gitea.eom.dev/devops/ci-tools:1.0.0` | |
| `role-name` | `ericomeehan.` + repo name without `ansible-role-` | The name the playbooks use, e.g. `ericomeehan.gitea`. |
| `ansible-lint-args` | empty | Extra flags, e.g. `--profile production`. |
How it works and why:
- ansible-lint reads a role's name from its directory, so the job lints a copy
named like the playbooks name it. The repo's own name also resolves, because
galaxy's skeleton `tests/test.yml` uses it and `syntax-check` can't be
skipped.
- `role-name` is a warning, not a failure. Our roles keep galaxy's skeleton
`meta/main.yml` and are called `ericomeehan.<name>`, which galaxy's naming
rule rejects.
- The repo's `.ansible-lint`, `.yamllint` and `.ansible-lint-ignore` are used
if present.
**Baseline for existing problems.** Most roles have many existing findings
(ansible-role-gitea has 48, mostly `name[casing]` and `fqcn`). To pass today
and fail only on new problems, commit a baseline:
```bash
docker run --rm -v "$PWD:/src" -w /tmp gitea.eom.dev/devops/ci-tools:1.0.0 bash -c '
mkdir -p /tmp/roles && cp -a /src /tmp/roles/ericomeehan.NAME &&
ln -s /tmp/roles/ericomeehan.NAME /tmp/roles/ansible-role-NAME &&
cd /tmp/roles/ericomeehan.NAME &&
ANSIBLE_ROLES_PATH=/tmp/roles ansible-lint --warn-list role-name --generate-ignore;
cp .ansible-lint-ignore /src/'
```
Replace `NAME`. Baselined findings still print, as warnings. Fixing them later
means deleting lines from `.ansible-lint-ignore`.
### Helm chart
```yaml
name: lint
on:
push:
branches: [main]
pull_request:
concurrency:
group: lint-${{ github.ref }}
cancel-in-progress: true
jobs:
helm:
uses: DevOps/software-infrastructure/.gitea/workflows/reusable-helm-lint.yaml@main
with:
chart-path: .
```
| Input | Default | Notes |
|---|---|---|
| `tools-image` | `gitea.eom.dev/devops/ci-tools:1.0.0` | |
| `chart-path` | `.` | Directory with `Chart.yaml`. |
| `kube-version` | `1.31.2` | The cluster's version on 2026-11-14. Bump it when the cluster is upgraded. |
| `values-files` | empty | Extra values files to render as well, space-separated. The defaults are always rendered. |
| `strict` | `false` | `helm lint --strict`. |
Chart dependencies are built if `Chart.yaml` has any. kubeconform skips kinds
it has no schema for (CRDs) rather than failing on them. Schemas are fetched
from GitHub at run time.
### Dockerfile
```yaml
name: lint
on:
push:
branches: [main]
paths: ["**/Dockerfile*", "**/*.Dockerfile", ".hadolint.yaml"]
pull_request:
paths: ["**/Dockerfile*", "**/*.Dockerfile", ".hadolint.yaml"]
concurrency:
group: dockerfile-lint-${{ github.ref }}
cancel-in-progress: true
jobs:
hadolint:
uses: DevOps/software-infrastructure/.gitea/workflows/reusable-dockerfile-lint.yaml@main
```
| Input | Default | Notes |
|---|---|---|
| `tools-image` | `gitea.eom.dev/devops/ci-tools:1.0.0` | |
| `dockerfiles` | every `Dockerfile*` and `*.Dockerfile` | Space-separated paths. The job fails if it finds none. |
| `failure-threshold` | `warning` | `error`, `warning`, `info` or `style`. Use a `.hadolint.yaml` for per-rule ignores. |
## How these were tested (2026-11-14)
Each workflow's `run:` steps were taken from the YAML as written and run inside
a local build of the tools image (1.0.0), against:
| Workflow | Case | Result |
|---|---|---|
| ansible-role-lint | DevOps/ansible-role-gitea as it is | fails, 48 findings |
| | the same with a generated baseline | passes |
| | baseline plus a new `shell:` task | fails on the new task only |
| helm-lint | DevOps/helm-jamulus | passes |
| | a template referencing a missing value | fails in `helm lint` |
| | a Service with a string port | fails in kubeconform |
| dockerfile-lint | `images/ci-tools/Dockerfile` | passes |
| | `FROM debian:latest` with unpinned apt | fails (DL3007, DL3008) |
| | repo with no Dockerfile | fails with a clear message |
`actionlint` 1.7.12 and `yamllint` pass on all the workflow files.
**Not tested yet:**
- Running on the Gitea runner at all, including `container.image` from an input.
- Calling a workflow from another repo with `uses: owner/repo/path@ref`, and
from a private repo calling this public one. This is the G2 acceptance check.
If the runner can't resolve cross-repo `uses:`, the fallback is to copy the
caller with the steps inlined, generated from these files.