Compare commits

...

9 Commits

Author SHA1 Message Date
Niklas Ye 9669b8f477 Set the chart's placeholder version to 0.9.4
CI / chart (push) Successful in 0s
CI / security (push) Successful in 24s
CI / test (push) Successful in 28s
Release / test (push) Successful in 28s
Release / chart (push) Successful in 1s
Release / binaries (push) Successful in 28s
Release / image (push) Successful in 1m11s
Release / scan-image (push) Successful in 23s
Cosmetic, as in f46e5f5. `make helm-package` passes --version and
--app-version from the tag, so neither field decides anything about what
release.yaml publishes.

Done anyway because a tree heading for v0.9.4 that still says 0.9.3 tells
a reader something false, and the tree is what gets read before the tag
exists. appVersion keeps the v, per APPVERSION_PREFIX in .release.conf.

Claude-Session: https://claude.ai/code/session_014m2pJdpCTv3mvvUUuBM54Y
2026-09-04 18:08:21 +02:00
Niklas Ye a7871ed7c6 Stop the bootstrap hook installing curl at run time
The hook's container was alpine:3 and its first line was
`apk add --no-cache curl`. That writes the binary into the container's
writable upper layer, and every exec of it afterwards is, correctly, a
dropped binary: Falco's `Drop and execute new binary in container`
(PCI_DSS_11.5.1, MITRE TA0003) fired twice at Critical on the upgrade to
chart 0.9.3, 65ms after the container started, with
evt.arg.flags=EXE_WRITABLE|EXE_UPPER_LAYER. Ryuvia/charts#100 has the
event lines.

A true positive of the rule and a false positive of intent, and it is not
a one-off: the hook is post-install,post-upgrade, so it recurred on every
release. The cluster is still in the Falco burn-in with detections routed
to a null receiver, which is the only reason nobody was paged for it.

Fixed here rather than with a Falco exception on purpose. An exception
would have to name this container and would then stay in the rule set
forever, blinding it for the one workload that already runs as root with
create-secret RBAC, and it would leave the second problem untouched: this
runs as a post-upgrade hook, a failed hook fails the release, so every
`helm upgrade` of terdut-server depended on dl-cdn.alpinelinux.org
answering. That dependency is now gone.

alpine/curl is still a full Alpine, so sh, cat, sleep, grep, cut, head and
tail are all present -- verified in-cluster before the swap rather than
assumed, since a missing utility would surface as a failed post-upgrade
hook and not as anything visible here. Digest-pinned, as the wrapper
chart's own sidecar images are. The image declares an ENTRYPOINT, which
the Job's `command:` overrides; a comment says so, because rewriting that
to `args:` would silently run curl's entrypoint instead of the script.

No change to the script's logic, to the RBAC, or to when the hook runs.
Nothing on the terdut-tui side of the API moves, and no terdut-tui version
is required or excluded by this.

Worth recording while it is in view, and deliberately not acted on here:
there is no terdut-server-admin-key secret in the namespace, so the POST
returns 403, the hook logs "Server already bootstrapped, nothing to do"
and exits before the secret-creating branch. On an upgrade this hook
currently achieves nothing at all. Narrowing it to post-install would
remove the detection outright, but that changes what the hook is for and
belongs in its own change.

Claude-Session: https://claude.ai/code/session_014m2pJdpCTv3mvvUUuBM54Y
2026-09-04 18:08:05 +02:00
Niklas Ye 79f5db2636 Scan the source and the working tree too, not just the image
CI / chart (push) Successful in 0s
CI / security (push) Successful in 19s
CI / test (push) Successful in 25s
The image scan added yesterday reads the built artifact. It cannot see a
vulnerable dependency the binary never calls into, and it cannot see a
credential in a file that never reaches the image — this one is FROM scratch
and contains a single binary, so almost nothing in the repo is in it. Those
are two different questions and they need two different tools, which is why
riksdata and rd-web have run govulncheck and gitleaks all along.

Both run on every push and pull request rather than only on a tag, since
neither needs anything published.

Checked by hand before wiring in, as with the image scan. govulncheck
reports no vulnerabilities the code can reach, and gitleaks finds nothing in
the tree.

What govulncheck does report is worth writing down, because it is the
argument for having it. It found three advisories in chi and reports none of
them, all three being IP spoofing in middleware.RealIP, which router.go does
not use — it uses Logger and Recoverer. The analysis is symbol-level rather
than dependency-level, so adding middleware.RealIP would turn this red on
the next push. That is precisely when someone should be made to look, and it
is a plausible thing to reach for here, since the API sits behind a gateway
and real client addresses are exactly what RealIP is for. The fourth finding
is an integer overflow in golang.org/x/sys/windows, which a linux/scratch
image will not be calling.

Both gates were checked for the failure direction as well. gitleaks exits 1
on a private key block. Worth knowing when testing it: it allowlists
well-known example credentials, so the AWS key from Amazon's own
documentation does not trip it and proves nothing.

Neither reads git history. gitleaks runs with --no-git, which scans the
working tree, so it stops a secret on the way in and says nothing about what
is already committed.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-02 10:11:13 +02:00
Niklas Ye 84146fc903 Scan the published image for known vulnerabilities
CI / chart (push) Successful in 1s
CI / test (push) Successful in 25s
terdut-server was the only one of the three release-managed repos with no
image scanning at all. riksdata and rd-web have had a scan-image job since
they were set up; everything published here up to and including v0.9.3 went
out without a CVE check.

It scans the pushed image rather than a locally built one, for the same
reason the siblings do: trivy cannot read a local image on this runner,
since Talos has no docker socket and the dind sidecar shares no filesystem
with the job. So it runs after image rather than gating it, and a red scan
unpublishes nothing. What it means is narrower and worth stating plainly: do
not bump the wrapper chart in Ryuvia/charts to that version.

Checked before wiring it in rather than after. v0.9.3 scans clean at
HIGH,CRITICAL with unfixed findings ignored, so this does not turn the
pipeline red on arrival, and the same command exits 1 on an image that does
have findings — a gate that cannot fail is not a gate.

One platform is scanned, not both. The image is FROM scratch, so there are
no OS packages and trivy sees a single target: the Go binary and its module
graph. linux/amd64 and linux/arm64 are that same module set built for a
different GOARCH, so a finding in one is a finding in both. On an image with
a base layer that reasoning would not hold.

Still no govulncheck and no gitleaks here, which riksdata and rd-web run in
a separate CI job. This is the only security scanning terdut-server has.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-02 10:04:51 +02:00
Niklas Ye c6f1fe317e Correct the docs that said this repo has no publishing targets
CI / chart (push) Successful in 0s
CI / test (push) Successful in 25s
Both CLAUDE.md and README.md claimed there were deliberately no build or
push targets because the workflow owned publishing. That stopped being true
in 69fcc24, which moved publishing onto the Makefile so release.yaml could
call it — the docs described the arrangement that change replaced.

The claim was wrong in its reasoning too, not just out of date. It was
written on the assumption that riksdata and rd-web duplicated their
pipelines by having those targets. They never did: their workflows call
make and always have, which is what makes a green gate locally and a green
pipeline the same code instead of two descriptions of it. This repo was the
exception, for the single day it had a Makefile that nothing called.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-01 22:44:22 +02:00
Niklas Ye f46e5f5729 Set the chart's placeholder version to 0.9.3
CI / chart (push) Successful in 0s
CI / test (push) Successful in 24s
Release / test (push) Successful in 24s
Release / chart (push) Successful in 1s
Release / binaries (push) Successful in 9s
Release / image (push) Successful in 57s
Cosmetic, and done anyway, for the same reason as 477454e: these two lines
decide nothing about what is published — `helm package --version
--app-version` now sets both from the tag, so they are not even read during
a release — but a tree heading for v0.9.3 that says 0.9.2 tells its reader
something false.

The comment above them was rewritten to stop naming a specific version. It
described the tree as "heading for v0.9.2", which is exactly the kind of
line that has to be edited every release and is therefore the kind that goes
stale. It also said nothing that publishes looks at these fields, which only
became literally true in 69fcc24 when the sed was replaced by helm's own
flags.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-01 22:38:41 +02:00
Niklas Ye 69fcc24a4d Drive the pipeline through make, the way riksdata and rd-web do
CI / chart (push) Successful in 0s
CI / test (push) Successful in 57s
Both workflows restated the build in YAML: gofmt, go vet and go test inline
in two places, buildx inline in a third, and the chart's version sed'd into
Chart.yaml before packaging. The Makefile added in 1081260 then described
the same checks a second time for local use, which made "green locally means
green in CI" a promise about keeping two files in step rather than a
property of the setup.

riksdata and rd-web never had that problem — their workflows call make and
have done all along. This repo was the odd one out, and only because it had
no Makefile until today. Now ci.yaml runs `make fmt lint test`, release.yaml
runs the same plus `make binaries`, `make push`, `make helm-package` and
`make helm-push`, and the reasoning behind each check lives on the target
rather than in whichever YAML file was edited last.

Three things change rather than just move:

The chart is linted before it is published. release.yaml packaged and pushed
without ever rendering the templates, so a chart that did not compile would
have reached the registry and been found by Flux. ci.yaml gained a chart job
for the same reason.

helm package --version --app-version replaces the sed. The published
metadata is identical, but the tree is no longer mutated mid-build, and it
is what the rest of the release process already assumed happened.

`make push` refuses VERSION=dev. Publishing is one command now, so it is
also one command to run by accident; dev is not a version anyone releases.

Deliberately not moved: uploading the release assets. Compiling them is
`make binaries` and runs anywhere, but the upload needs a token and the
Gitea release API, which is the workflow's business and not something worth
a target.

`push` builds and pushes in one step, unlike riksdata's separate build and
push, because buildx cannot load a multi-platform image into the local store
— it can only push it. `build` stays single-platform and local-only.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-01 22:36:54 +02:00
Niklas Ye 6a4f902e38 Declare that this repo's release prose is English
CI / test (push) Successful in 5s
The release skill's house style opened with "everything here is written in
Swedish", stated as a house default. It is not one: Swedish belongs to
riksdata and rd-web because their interface copy and the riksdag data they
present are inseparable from it. Nothing about an on-call tool is — the
labels, the API and the data here are English, and so are the code comments
and the docs.

Stated as a rule that was remembered rather than read, it produced the whole
of v0.9.2 in Swedish: four commits, the tag body and both charts PRs. The
PRs were rewritten; the commits on main and the published tag could not be,
so v0.9.2 stays as the record of the mistake.

PROSE_LANG makes it a declaration each repo carries, which release-preflight
prints on the prose line before any of the four texts get written. See
reference/house-style.md in the skill.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-01 22:24:41 +02:00
Niklas Ye 5f9c202d65 Beskriv hur ett släpp går till, och vad som är särskilt här
CI / test (push) Successful in 6s
Repot kom in under den gemensamma släppprocessen i 1081260 utan att någon
dokumentation sa det. README hade Kubernetes och Development men ingenting om
vägen från en commit till något som rullar i klustret, och till skillnad från
riksdata och rd-web fanns ingen CLAUDE.md alls.

Två saker om just det här repot är värda att stå skrivna, eftersom båda
avviker från syskonrepona och båda upptäcktes mitt i ett släpp:

Pipelinen har ingen bildskanning. riksdata och rd-web kör trivy efter
publiceringen; .gitea/workflows/release.yaml här har test, binaries, image och
chart och inget mer. Ett grönt släpp är alltså inget belägg för att bilden är
fri från kända sårbarheter, och en släppnot får inte antyda det.

Wrapperchartets values.yaml har två tag:-rader, appbilden och
python-sidovagnen för säkerhetskopiering, så chart-bump behöver --image för
att veta vilken som flyttas. Utan den vägrar den, vilket är rätt.

Samtidigt: varför det inte finns några build- eller push-mål, varför chartet
publiceras enbart från taggen (766f439), och varför make test kör -race när CI
inte gör det. Helm-exemplet pekade fortfarande på 0.9.0 och går till 0.9.2.

Claude-Session: https://claude.ai/code/session_01S7R4gWTz5wh5xCY4nCSJjN
2026-09-01 22:09:15 +02:00
9 changed files with 391 additions and 103 deletions
+62 -24
View File
@@ -59,32 +59,70 @@ jobs:
git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
fi
# This exists because `go vet` does not look at import order: the move to
# git.ryuvia.com rewrote every import path without re-sorting, the new path sorts
# before github.com/..., and both repos sat unformatted through a green CI run and
# a release before anyone noticed.
# The gate is the Makefile's rather than a second copy of it here, the way riksdata
# and rd-web already do it. `make fmt lint test` is exactly what a developer runs, so
# a green pipeline and a green working copy mean the same thing by construction
# instead of by remembering to update two files together.
#
# Both of gofmt's failure modes need handling, and they are not alike. A file that
# is merely misformatted is listed on stdout with exit 0 -- so the failure has to
# be raised by hand. A file that does not parse is the opposite: nothing on stdout
# and exit 2, which a naive `[ -n "$unformatted" ]` reads as success. The first
# draft of this step had exactly that hole.
- name: Format
# The reasoning that used to live here moved with the targets: why gofmt is checked
# at all (import order survives `go vet`, and both repos sat unformatted through a
# green run and a release -- 9046f6e), why both of gofmt's failure modes need
# handling, and why `test` adds -race when this job does not have to.
- name: Format, vet 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. Both are needed: govulncheck reads the
# source and its module graph, trivy reads the built artifact, and 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: ${{ github.event.pull_request.head.sha }}
run: |
if ! unformatted=$(gofmt -l .); then
echo "::error::gofmt could not parse the tree"
gofmt -l . # re-run unredirected so the parse errors reach the log
exit 1
fi
if [ -n "$unformatted" ]; then
echo "::error::not gofmt'd:"
echo "$unformatted"
gofmt -d .
exit 1
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: Vet
run: go vet ./...
- name: Go vulnerability scan (govulncheck)
run: make security-go
- name: Test
run: go test ./...
- 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
# release.yaml's chart job runs on the host.
#
# The chart had no lint step in any workflow until 2026-09-01: release.yaml packaged and
# pushed it without rendering it first, so a template that did not compile would have
# been found by Flux rather than here.
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
+58 -70
View File
@@ -41,35 +41,10 @@ jobs:
REF_NAME: ${{ github.ref_name }}
run: git clone --depth=1 --branch "$REF_NAME" "$REPO_URL" .
# This exists because `go vet` does not look at import order: the move to
# git.ryuvia.com rewrote every import path without re-sorting, the new path sorts
# before github.com/..., and both repos sat unformatted through a green CI run and
# a release before anyone noticed.
#
# Both of gofmt's failure modes need handling, and they are not alike. A file that
# is merely misformatted is listed on stdout with exit 0 -- so the failure has to
# be raised by hand. A file that does not parse is the opposite: nothing on stdout
# and exit 2, which a naive `[ -n "$unformatted" ]` reads as success. The first
# draft of this step had exactly that hole.
- name: Format
run: |
if ! unformatted=$(gofmt -l .); then
echo "::error::gofmt could not parse the tree"
gofmt -l . # re-run unredirected so the parse errors reach the log
exit 1
fi
if [ -n "$unformatted" ]; then
echo "::error::not gofmt'd:"
echo "$unformatted"
gofmt -d .
exit 1
fi
- name: Vet
run: go vet ./...
- name: Test
run: go test ./...
# 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
@@ -86,21 +61,13 @@ jobs:
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: ${{ github.ref_name }}
run: |
set -eu
mkdir -p dist
for target in linux/amd64 linux/arm64 darwin/amd64 darwin/arm64; do
GOOS="${target%/*}"
GOARCH="${target#*/}"
out="dist/terdut-${REF_NAME}-${GOOS}-${GOARCH}"
echo "building $out"
GOOS="$GOOS" GOARCH="$GOARCH" go build \
-ldflags "-w -s -X main.version=${REF_NAME}" \
-o "$out" ./cmd/terdut
done
run: make binaries VERSION="$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
@@ -158,23 +125,12 @@ jobs:
TOKEN: ${{ secrets.REGISTRY_TOKEN }}
run: echo "$TOKEN" | docker login "$REGISTRY" -u niklas --password-stdin
# The default "docker" driver cannot build more than one platform at a time; the
# docker-container driver can. Reused across runs if it survived the last one.
- name: Prepare buildx
run: docker buildx create --name terdut --use 2>/dev/null || docker buildx use terdut
# No QEMU: the Dockerfile's builder stage runs on $BUILDPLATFORM and cross-compiles
# from TARGETARCH, so both platforms build natively. See the comment in Dockerfile.
# 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: ${{ github.ref_name }}
run: |
docker buildx build \
--platform linux/amd64,linux/arm64 \
--build-arg "VERSION=${REF_NAME}" \
--tag "${IMAGE}:latest" \
--tag "${IMAGE}:${REF_NAME}" \
--push .
run: make push VERSION="$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.
@@ -200,10 +156,10 @@ jobs:
# 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 --
# the sed below ties the chart version to the app version, so a chart-only change
# `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: Stamp the chart version from the tag
- name: Refuse a non-version tag
env:
REF_NAME: ${{ github.ref_name }}
run: |
@@ -212,11 +168,17 @@ jobs:
echo "::error::refusing to publish a chart for non-version tag ${REF_NAME}"
exit 1
fi
CHART_VERSION="${REF_NAME#v}"
sed -i "s/^version:.*/version: ${CHART_VERSION}/" charts/terdut-server/Chart.yaml
sed -i "s/^appVersion:.*/appVersion: \"${REF_NAME}\"/" charts/terdut-server/Chart.yaml
cat charts/terdut-server/Chart.yaml
# 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: ${{ github.ref_name }}
@@ -224,12 +186,38 @@ jobs:
run: |
set -eu
echo "$TOKEN" | helm registry login "$REGISTRY" -u niklas --password-stdin
# Isolated repo config: the machine-wide helm repo list is not this job's
# business, and one unreachable entry in it aborts otherwise-fine commands.
# HELM_REPOSITORY_CACHE is deliberately NOT overridden alongside it -- helm
# writes a refreshed index to the default cache and then looks for it in the
# overridden one.
export HELM_REPOSITORY_CONFIG="$PWD/.helm-repos.yaml"
: > "$HELM_REPOSITORY_CONFIG"
helm package charts/terdut-server -d dist
helm push "dist/terdut-server-${REF_NAME#v}.tgz" "oci://${REGISTRY}/niklas"
make helm-package helm-push VERSION="$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"
+5
View File
@@ -1,6 +1,11 @@
# build output
/terdut
/terdut-server
# `make binaries` and `make helm-package` write here
/dist/
# isolated helm repo list written by the publishing targets, so the machine-wide
# one (which has an unreachable entry) cannot abort a release
/.helm-repos.yaml
# SQLite database files
*.db
+7
View File
@@ -6,6 +6,7 @@
# Defaults, set here only where this repo differs:
# CHARTS_REPO=$HOME/git/charts CHARTS_DIR=<image basename>
# GITEA_LOGIN=Ryuvia APPVERSION_PREFIX=
# PROSE_LANG=en
# Same as the image basename, so this is only stated to be read rather than derived.
CHARTS_DIR=terdut-server
@@ -15,3 +16,9 @@ CHARTS_DIR=terdut-server
# version and appVersion from the tag when it publishes -- but people read it, and until
# 2026-09-01 it said "latest" while the tree headed for a numbered release.
APPVERSION_PREFIX=v
# English. The Swedish in riksdata and rd-web follows from their subject matter, not from a
# house style: terdut-server is an on-call tool whose labels, API and data are English, and
# nothing about it is coupled to Swedish. Code comments and docs here were always English;
# from 2026-09-01 the release prose is too.
PROSE_LANG=en
+43
View File
@@ -0,0 +1,43 @@
## Release
Say **"Release"** (or "Release X.Y.Z") and the `release` skill runs it: commit, push, tag,
wait for the pipeline, then open the wrapper-chart PR against `Ryuvia/charts`. It stops
there — merging and the Flux reconcile stay manual, deliberately.
Preconditions and the plan, without side effects:
```sh
~/.claude/skills/release/scripts/release-preflight # state + suggested version
~/.claude/skills/release/scripts/release-preflight vX.Y.Z # validate that release
```
Config is `.release.conf` here plus `make release-vars`. The process itself lives in
`~/.claude/skills/release/`; why it is shaped this way is in README.md §Releasing.
Two things about this repo specifically:
- **The image is scanned after it is published, not before.** `scan-image` runs trivy
against the pushed image, because trivy cannot read a locally built one on this runner.
A red scan therefore unpublishes nothing — it means: do not bump the wrapper chart in
`Ryuvia/charts` to this version. Added 2026-09-02; every release up to and including
v0.9.3 was published with no CVE check at all.
- **The wrapper chart has two `tag:` lines** — the app image and the python backup sidecar —
so `chart-bump` needs `--image "$IMAGE"` to know which one moves.
## Checks
`make fmt lint test helm-lint` **is** what the pipeline runs — `ci.yaml` and `release.yaml`
call these targets rather than restating them, the way riksdata and rd-web do. A green gate
here and a green pipeline are the same code, not two descriptions of it. `test` adds `-race`,
which the workflows do not have to ask for since they call the target; see the comment on it
for why.
`make release` (build + push the multi-arch image, package + push the chart) is what
`release.yaml` invokes. Do not run it by hand — it refuses `VERSION=dev` for that reason, and
publishing happens by pushing a tag.
Three scans, and they see different things: `security-go` (govulncheck) reads the source and
its module graph and reports only vulnerabilities the code can actually reach;
`security-secrets` (gitleaks) reads the working tree, not the history, so it catches a secret
on the way in rather than auditing what is already committed; `security-image` (trivy) reads
the published artifact and therefore only runs on a tag. The first two gate every push.
+135
View File
@@ -82,3 +82,138 @@ release-vars: ## Print the variables the release process reads
# the multi-arch tag, which is both easy to do by accident and invisible afterwards
# — the tag would still resolve, just not on arm64. Publishing happens by pushing a
# tag; nothing else.
## --- publishing ---
#
# These exist so .gitea/workflows/release.yaml can call `make release` instead of
# restating the build in YAML, the way riksdata and rd-web already do. One definition
# of how this is built and published, runnable locally, reviewable in a diff.
#
# VERSION is the git tag, passed in by the workflow. The guard below is why a stray
# local `make release` cannot publish: dev is not a version anyone releases.
VERSION ?= dev
# Helm requires strict SemVer — strip a leading 'v' if present.
CHART_VERSION := $(shell echo "$(VERSION)" | sed 's/^v//')
PLATFORMS ?= linux/amd64,linux/arm64
BUILDX_BUILDER ?= terdut
TRIVY_VERSION := 0.73.0
GOVULNCHECK_VERSION := v1.1.4
GITLEAKS_VERSION := v8.30.0
# --pull, not --no-cache: refresh the base image without discarding the layer cache.
DOCKER_BUILD_FLAGS ?= --pull
# An isolated repo list. The machine-wide one is not this build's business, and one
# unreachable entry in it aborts otherwise-fine helm commands — there is a dead
# TrueCharts repo on this host that does exactly that. HELM_REPOSITORY_CACHE is
# deliberately NOT overridden alongside it: helm writes a refreshed index to the default
# cache and then looks for it in the overridden one.
HELM_ISOLATED = HELM_REPOSITORY_CONFIG=$(CURDIR)/.helm-repos.yaml
.PHONY: require-version
require-version:
@test "$(VERSION)" != "dev" || \
(echo "VERSION=dev names no release — pass VERSION=vX.Y.Z (the workflow passes the tag)" && exit 1)
.PHONY: build
build: ## Build the image for this host only, without pushing (local check / CI smoke)
docker build $(DOCKER_BUILD_FLAGS) \
--build-arg VERSION=$(VERSION) \
-t $(IMAGE):$(VERSION) .
# Multi-arch, so unlike riksdata and rd-web this cannot be a separate build then push:
# buildx cannot load a multi-platform result into the local image store, it can only
# push it. `build` above stays single-platform and local-only for that reason.
#
# No QEMU: the Dockerfile's builder stage runs on $$BUILDPLATFORM and cross-compiles from
# TARGETARCH, so both platforms build natively. The default "docker" driver cannot build
# more than one platform at a time; the docker-container driver can.
.PHONY: push
push: require-version ## Build and publish the multi-arch image
docker buildx create --name $(BUILDX_BUILDER) --use 2>/dev/null || docker buildx use $(BUILDX_BUILDER)
docker buildx build \
--platform $(PLATFORMS) \
--build-arg "VERSION=$(VERSION)" \
--tag "$(IMAGE):latest" \
--tag "$(IMAGE):$(VERSION)" \
--push .
# --version and --app-version come from the tag, so Chart.yaml's own fields decide nothing
# about what is published. They used to be rewritten in place with sed before packaging;
# the flags do the same job without mutating the tree mid-build.
.PHONY: helm-package
helm-package: require-version ## Package the chart, versioned from the tag
$(HELM_ISOLATED) helm package $(HELM_CHART) \
--version $(CHART_VERSION) \
--app-version $(VERSION) \
--destination dist
.PHONY: helm-push
helm-push: require-version ## Push the packaged chart to the OCI registry
$(HELM_ISOLATED) helm push dist/terdut-server-$(CHART_VERSION).tgz $(HELM_REPO)
.PHONY: binaries
binaries: require-version ## Cross-compile the release binaries into dist/
@mkdir -p dist
@set -eu; for target in linux/amd64 linux/arm64 darwin/amd64 darwin/arm64; do \
GOOS="$${target%/*}"; GOARCH="$${target#*/}"; \
out="dist/terdut-$(VERSION)-$${GOOS}-$${GOARCH}"; \
echo "building $$out"; \
GOOS="$$GOOS" GOARCH="$$GOARCH" go build \
-ldflags "-w -s -X main.version=$(VERSION)" \
-o "$$out" ./cmd/terdut; \
done
.PHONY: release
release: push helm-package helm-push ## Publish image + chart (the workflow's one call)
## --- security ---
# Symbol-level, not dependency-level: govulncheck reports a vulnerability only when the
# code can actually reach it. As of 2026-09-02 this repo imports three chi advisories and
# reports none of them, because all three are middleware.RealIP and router.go uses Logger
# and Recoverer. That is the useful property rather than a loophole -- adding
# middleware.RealIP would turn this red, which is exactly when someone should look.
.PHONY: security-go
security-go: ## Scan Go deps for known CVEs (govulncheck)
go run golang.org/x/vuln/cmd/govulncheck@$(GOVULNCHECK_VERSION) ./...
# --no-git scans the working tree rather than the history, so this catches a secret on the
# way in. It is not a history audit and finding nothing here says nothing about what is
# already committed. --redact because the finding is printed into a CI log.
#
# Note when testing it that gitleaks allowlists well-known example credentials -- the AWS
# key from their own documentation does not trip it. A private key block does.
.PHONY: security-secrets
security-secrets: ## Scan the working tree for committed secrets (gitleaks)
go run github.com/zricethezav/gitleaks/v8@$(GITLEAKS_VERSION) detect --no-git \
--source . --redact --no-banner --exit-code 1
# Scans the pushed image, not a local one: trivy cannot read a locally built image on the
# runner -- Talos has no docker socket, and the dind sidecar shares no filesystem with the
# job -- so it pulls from the registry. Same reason riksdata and rd-web scan after pushing.
#
# It cannot gate a deploy, because this pipeline does not deploy. A red scan means: do not
# bump the wrapper chart in Ryuvia/charts to this version.
#
# The image is FROM scratch, so there are no OS packages to scan and trivy sees exactly one
# target -- the Go binary and its module graph. That also makes scanning a single platform
# sufficient here: linux/amd64 and linux/arm64 are the same modules built for a different
# GOARCH, so a CVE in one is a CVE in both. On an image with a base layer that would not
# hold and both platforms would need scanning.
#
# This is the last of the three scans and the only one that needs a published artifact;
# security-go and security-secrets above run on every push.
.PHONY: security-image
security-image: require-version ## Scan the pushed image for CVEs (needs VERSION)
@# The named volume persists trivy's vulnerability DB between runs; without it every
@# scan re-downloads the whole database.
docker run --rm -e TRIVY_USERNAME -e TRIVY_PASSWORD \
-v trivy-cache:/root/.cache/trivy \
docker.io/aquasec/trivy:$(TRIVY_VERSION) image --severity HIGH,CRITICAL \
--ignore-unfixed --exit-code 1 $(IMAGE):$(VERSION)
+62 -1
View File
@@ -57,7 +57,7 @@ with the app — chart `x.y.z` is always app `vx.y.z`:
```bash
helm upgrade --install terdut-server oci://git.ryuvia.com/niklas/terdut-server \
--version 0.9.0 \
--version 0.9.2 \
--namespace terdut-server --create-namespace \
--set networking.hostname=terdut.example.com
```
@@ -661,3 +661,64 @@ go test ./... # run all tests
go build ./... # compile all packages
go run ./cmd/terdut # run locally
```
`make fmt lint test helm-lint` is the gate. It mirrors `.gitea/workflows/ci.yaml` step for
step, so a green run here means a green pipeline — with one deliberate exception: `make test`
adds `-race`, which CI does not. The sweeper, the notifier goroutine and the dead man's switch
sweep all touch the same single database connection, and a race between them would surface as
a flaky incident in production rather than as a red build.
## Releasing
```
push or PR → ci.yaml gofmt, go vet, go test -race
govulncheck, gitleaks
helm lint + render
push tag vX.Y.Z → release.yaml the same gate, then publish:
git.ryuvia.com/niklas/terdut-server:vX.Y.Z
oci://git.ryuvia.com/niklas/terdut-server X.Y.Z
then trivy-scan the pushed image
PR to Ryuvia/charts → bump the wrapper chart to X.Y.Z; on merge
Flux reconciles and the release rolls out
```
Both artifacts go to the **personal** Gitea namespace rather than `ryuvia`, because Gitea
scopes package visibility to the owner with no per-package override — so `ryuvia/*` is private
because the org is. Publishing to `niklas` keeps them anonymously pullable, which is why no
pull secret is needed in the cluster. Same reasoning, and the same choice, as riksdata and
rd-web.
Saying **"Release"** runs all three rows: the `release` skill commits, pushes, tags, waits for
the pipeline, and opens the `Ryuvia/charts` PR, stopping before the merge. See
`~/.claude/skills/release/`, or `.release.conf` here for this repo's part of it.
The chart is published **only** from the tag, by the `chart` job. 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 — chart 0.9.0 went out reading `appVersion: "latest"` that way. One
publisher, triggered by the tag (`766f439`). The cost is that a chart-only change has no
version of its own and rides the next app tag.
Both workflows are thin drivers over the Makefile: `ci.yaml` runs `make fmt lint test` and
`make helm-lint`, `release.yaml` adds `make binaries`, `make push`, `make helm-package` and
`make helm-push`. That is deliberate — it is what makes a green local gate and a green
pipeline the same code rather than two descriptions of it, and it is how riksdata and rd-web
have always worked.
`make push` builds and pushes in one step, unlike those two, because the image is
`linux/amd64,linux/arm64` and buildx cannot load a multi-platform result into the local image
store. `make build` stays single-platform and local-only. Both refuse `VERSION=dev`:
publishing is one command, so it is also one command to run by accident. Publishing happens
by pushing a tag.
Two things the release process needs to know about this repo:
- **The image scan runs after publishing**, like riksdata's and rd-web's: trivy cannot read
a locally built image on this runner, so it pulls the pushed one. A red `scan-image` means
do not bump the wrapper chart to that version — it cannot unpublish anything. The image is
`FROM scratch`, so trivy sees exactly one target, the Go binary and its module graph.
- **The wrapper chart's `values.yaml` has two `tag:` lines** — the app image and the python
backup sidecar — so `chart-bump` is given `--image` to say which one moves.
The wrapper chart must have **its own `version:` bumped in the same commit**. Flux reconciles
with `reconcileStrategy: ChartVersion`, so a chart whose version did not change produces no
new artifact and the change is never deployed — with no error anywhere.
+6 -5
View File
@@ -6,13 +6,14 @@ type: application
# released values. .gitea/workflows/release.yaml rewrites both from the git tag when it
# publishes, so the chart version always equals the app version.
#
# They are kept in step with the tag anyway. Being read is the only thing these two
# lines do -- nothing that publishes looks at them -- and a tree heading for v0.9.2 that
# says 0.9.0 tells its reader something false. That is what they said until 2026-09-01,
# They are kept in step with the tag anyway. Being read is the only thing these two lines
# do -- `helm package --version --app-version` sets the published values from the tag and
# never consults these -- and a tree heading for a numbered release that states an older
# number tells its reader something false. They said 0.9.0 and "latest" until 2026-09-01,
# through two releases.
#
# appVersion and image.tag in values.yaml no longer agree, and that is not an oversight:
# image.tag stays "latest", which is what a local install actually pulls. appVersion is
# metadata and drives nothing.
version: 0.9.2
appVersion: "v0.9.2"
version: 0.9.4
appVersion: "v0.9.4"
@@ -23,13 +23,23 @@ spec:
serviceAccountName: {{ include "terdut-server.fullname" . }}-bootstrap
containers:
- name: bootstrap
image: alpine:3
# alpine/curl, not alpine:3 + `apk add curl`. Installing the binary at run time
# writes it into the container's writable upper layer, which is exactly the
# signature Falco's `Drop and execute new binary in container` (MITRE TA0003)
# exists to catch -- this hook emitted two Critical events on every single
# upgrade. See Ryuvia/charts#100. It also made `helm upgrade` depend on the
# Alpine CDN answering, since this runs as a post-upgrade hook and a failed
# hook fails the release.
#
# Still a full Alpine underneath, so sh, cat, sleep, grep, cut, head and tail
# are all present (verified in-cluster 2026-09-04). The image declares
# ENTRYPOINT ["/entrypoint.sh"], which `command:` below overrides -- do not
# change `command:` to `args:`.
image: alpine/curl:8.21.0@sha256:a1c44bab54d88e18ea9a6a4ecefab7f2d230b968567b78960fcaff8d51b7f067
command:
- /bin/sh
- -c
- |
apk add --no-cache curl > /dev/null 2>&1
SERVICE_URL="http://{{ include "terdut-server.fullname" . }}:{{ .Values.service.port }}"
SECRET_NAME="{{ include "terdut-server.bootstrapSecretName" . }}"
K8S_API="https://kubernetes.default.svc"