Compare commits

...

2 Commits

Author SHA1 Message Date
joakimp a3ff601bf0 Bump opencode 1.14.42 -> 1.14.44; close v1.14.42 omos-with-pi gap
Validate / docs-check (push) Successful in 18s
Validate / validate-base (push) Successful in 11m46s
Validate / validate-omos (push) Successful in 13m32s
Validate / validate-with-pi (push) Successful in 13m20s
Validate / validate-omos-with-pi (push) Successful in 19m23s
Publish Docker Image / smoke-base (push) Successful in 11m46s
Publish Docker Image / smoke-omos (push) Successful in 13m45s
Publish Docker Image / smoke-with-pi (push) Successful in 13m13s
Publish Docker Image / smoke-omos-with-pi (push) Successful in 15m47s
Publish Docker Image / build-base (push) Successful in 42m18s
Publish Docker Image / build-omos (push) Successful in 52m1s
Publish Docker Image / build-with-pi (push) Successful in 46m28s
Publish Docker Image / build-omos-with-pi (push) Successful in 54m36s
Publish Docker Image / update-description (push) Successful in 14s
opencode-ai 1.14.44 published 20:26 UTC (1.14.43 skipped upstream).
Bumping to 1.14.44 instead of re-running the failed v1.14.42 build
gives us the same 3h CI cost and picks up upstream bug fixes.

Closes the v1.14.42 omos-with-pi gap. The v1.14.42 tag's
build-omos-with-pi job failed during publish: oh-my-opencode-slim@1.0.7
had been published with a dependency on @opencode-ai/sdk@1.14.44, and
our build hit the npm registry within ~2 minutes of that SDK version
landing -- before the tarball had propagated across npm's CDN. The
manifest's dist-tags.latest pointed at 1.14.44 but a tarball fetch on
/-/sdk-1.14.44.tgz returned 404. Tarball is now fully fetchable.

Result on Docker Hub once v1.14.44 publishes:
  v1.14.42 / latest                        -> stable (3 of 4 variants)
  v1.14.42-omos / latest-omos              -> stable
  v1.14.42-with-pi / latest-with-pi        -> stable
  v1.14.42-omos-with-pi                    -> NEVER PUBLISHED (404 if pulled)
  latest-omos-with-pi                      -> still v1.14.41b until v1.14.44
  v1.14.44 / latest                        -> NEW (replaces latest)
  v1.14.44-omos / latest-omos              -> NEW
  v1.14.44-with-pi / latest-with-pi        -> NEW
  v1.14.44-omos-with-pi / latest-omos-with-pi -> NEW (closes the gap)

CHANGELOG: v1.14.44 entry added with the propagation-race rationale,
v1.14.42 entry annotated with the known gap. Reverse-chrono preserved.
2026-05-09 22:33:16 +02:00
joakimp 6fde27c212 Document the build pipeline architecture in .gitea/README.md
Validate / docs-check (push) Successful in 16s
Validate / validate-base (push) Successful in 12m9s
Validate / validate-omos (push) Successful in 16m45s
Validate / validate-with-pi (push) Successful in 13m30s
Validate / validate-omos-with-pi (push) Successful in 15m15s
The split-base build architecture, the NPM_CONFIG_PREFIX gotcha, the
hash-driven base cache reuse mechanism, and the cutover plan from
docker-publish.yml to docker-publish-split.yml were previously
scattered across:
  - inline Dockerfile.base / Dockerfile.variant comments
  - CHANGELOG Unreleased entries
  - AGENTS.md mentions
  - docker-publish-split.yml header comment
  - my own session notes

Consolidate into .gitea/README.md as the canonical architectural doc.
Gitea (like GitHub) auto-renders this when navigating to .gitea/ in
the web UI, so anyone investigating 'why is CI shaped this way?'
finds it on the first click. Cross-referenced from AGENTS.md as the
first thing to read when touching CI.

Covers:
  - The two release pipelines and why both exist
  - Why split-base: cross-variant cache misses on layer-hash-divergence
  - The 6 phases of the split-base pipeline with an ASCII diagram
  - base-decide hash inputs and Docker Hub probe logic
  - NPM_CONFIG_PREFIX variant-override pattern (the volume-shadow trap)
  - Registry cache strategy (mode=max for cross-arch reuse)
  - Wall-clock estimates: version-bump vs base-touching releases
  - Validate workflow role
  - Runner expectations: catthehacker image, disk reclaim, concurrency,
    Gitea Actions @v4 artifact incompatibility
  - 4-step migration plan from docker-publish.yml to .split.yml
  - Cross-refs to related docs

Does not duplicate AGENTS.md content; links to it for domain facts and
release-day checklist.
2026-05-09 19:28:03 +02:00
5 changed files with 308 additions and 2 deletions
+291
View File
@@ -0,0 +1,291 @@
# CI / Build Pipeline
This directory contains the gitea Actions workflows and the supporting
documentation for opencode-devbox's CI. If you're investigating *why*
the build pipeline is shaped the way it is, you're in the right place.
## Workflows in this directory
| File | Trigger | Role |
|---|---|---|
| [`workflows/docker-publish.yml`](workflows/docker-publish.yml) | `push: tags: v*` | **Production release pipeline.** Multi-arch build of all four variants (`base`, `omos`, `with-pi`, `omos-with-pi`), publish to Docker Hub, update Hub description. ~165180 min wall clock. |
| [`workflows/docker-publish-split.yml`](workflows/docker-publish-split.yml) | `workflow_dispatch` (manual) | **Experimental split-base pipeline.** Two-phase build: shared `base-<hash>` published once, then four thin variant deltas. Estimated ~3040 min on cache hit, ~7090 min when base needs rebuilding. Not yet validated end-to-end; once 12 dispatch test runs prove it, this will take over `on: push: tags: v*` and `docker-publish.yml` will be retired. |
| [`workflows/validate.yml`](workflows/validate.yml) | `push: branches: main` + PR | **Lightweight gate.** amd64-only smoke test of all four variants + `DOCKER_HUB.md` sync check. ~30 min. Fires on every push to `main`. |
## Why two release pipelines exist
opencode-devbox publishes **four image variants** (`base`, `omos`, `with-pi`, `omos-with-pi`) × **two architectures** (amd64, arm64) = **eight image tags per release**. Today's runners are 2 self-hosted gitea Actions runners. arm64 builds are emulated under QEMU, which is the dominant cost (~35x slower than native).
The four variants share ~95% of their layers (Debian + apt + Node + AWS CLI + mempalace + dev tools + entrypoints). The original `Dockerfile` was a single multi-stage build with `INSTALL_*` build-args gating variant-specific RUNs. BuildKit's per-layer cache key is content-addressed, but as soon as a build-arg-gated `RUN` produces a different layer hash for variant A vs variant B, every subsequent layer also has a different parent → identical commands re-execute per variant. Result: minimal cross-variant cache reuse on a fresh build.
Two improvements were considered:
1. **Reorder the original Dockerfile** so all variant-gated RUNs land at the bottom — modest gain, ~1020% wall-clock reduction. *Not pursued.*
2. **Split into `Dockerfile.base` + `Dockerfile.variant`** with the base published as a long-lived shared image — significant gain, ~5070% wall-clock reduction with hash-driven cache reuse. *Pursued.*
The split-base architecture is what the `docker-publish-split.yml` workflow exercises.
## How the split-base pipeline works
```
┌──────────────────┐
│ base-decide │ compute base-<hash>;
│ │ probe Docker Hub.
│ hash inputs: │
│ Dockerfile.base│
│ rootfs/ │
│ entrypoint*.sh │
└────────┬─────────┘
┌─────────────┴─────────────┐
│ need_build = true? │
└─────────────┬─────────────┘
yes │ no
┌──────────────────┐
│ build-base │ multi-arch build,
│ │ push base-<hash>
└────────┬─────────┘ to Docker Hub.
┌───────────────────────┼───────────────────────┐
▼ ▼ ▼
┌──────────┐ ┌──────────┐ ┌──────────────┐
│smoke-base│ │smoke-omos│ ... │smoke-omos-pi │ amd64 only,
└────┬─────┘ └────┬─────┘ └──────┬───────┘ parallel.
│ │ │
▼ ▼ ▼
┌──────────┐ ┌──────────┐ ┌──────────────┐
│build- │ │build- │ │build- │ multi-arch,
│variant- │ │variant- │ ... │variant- │ parallel,
│base │ │omos │ │omos-with-pi │ tag push.
└────┬─────┘ └────┬─────┘ └──────┬───────┘
└───────────────────────┴──────────────────────┘
┌──────────────────────────┐
│ promote-base-latest │ crane copy
│ │ base-<hash>
│ │ → base-latest
└────────┬─────────────────┘
┌──────────────────────────┐
│ update-description │
└──────────────────────────┘
```
### Step 1: `base-decide`
Compute a SHA-256 hash over the inputs that determine the base image's
content:
```sh
{
cat Dockerfile.base
find rootfs -type f -print0 | sort -z | xargs -0 cat
cat entrypoint.sh entrypoint-user.sh
} | sha256sum | cut -c1-12
```
The 12-character truncated hash becomes `base-<hash>`. Probe Docker Hub
for this tag via `docker manifest inspect`:
- If it exists → set `need_build=false`. `build-base` is skipped entirely.
- If it doesn't → set `need_build=true`. `build-base` runs.
This is the core cache-reuse mechanism. Version-bump-only releases
(only `Dockerfile.variant` or build-args changed) hit the cache. Releases
that change anything in the base — apt packages, AWS CLI, Node version,
locale list, entrypoint scripts — pay the full base-build cost once.
### Step 2: `build-base` (conditional)
Only runs when `need_build=true`. Multi-arch (amd64 + arm64) build of
`Dockerfile.base`, pushed to `joakimp/opencode-devbox:base-<hash>`.
Registry cache via `--cache-from/--cache-to` reduces incremental rebuilds
when only one or two layers changed.
The base image is **not** tagged `base-latest` here — that promotion
happens at the very end after all variants succeed (see step 5).
### Step 3: `smoke-*` (×4, parallel)
For each variant: build amd64-only against the base tag, load into
local docker, run [`scripts/smoke-test.sh`](../scripts/smoke-test.sh).
Variant build-args:
| variant | INSTALL_OPENCODE | INSTALL_OMOS | INSTALL_PI |
|---|---|---|---|
| `base` | true | false | false |
| `omos` | true | true | false |
| `with-pi` | true | false | true |
| `omos-with-pi` | true | true | true |
Smoke runs `--variant <name>` to enable variant-specific assertions.
Gate the publish: a smoke failure for variant X blocks `build-variant-X`.
### Step 4: `build-variant-*` (×4, parallel)
For each variant that passed smoke: multi-arch (amd64 + arm64) build of
`Dockerfile.variant`, pushed to Docker Hub with the user-facing release
tags:
| Build job | Tags pushed |
|---|---|
| `build-variant-base` | `vX.Y.Z`, `latest` |
| `build-variant-omos` | `vX.Y.Z-omos`, `latest-omos` |
| `build-variant-with-pi` | `vX.Y.Z-with-pi`, `latest-with-pi` |
| `build-variant-omos-with-pi` | `vX.Y.Z-omos-with-pi`, `latest-omos-with-pi` |
The `latest*` aliases are only updated when `promote_latest=true` (the
manual dispatch input) — for test runs, `promote_latest=false` keeps the
production aliases pointing at the previous good release.
### Step 5: `promote-base-latest`
Once all four variants successfully publish, re-tag `base-<hash>` as
`base-latest` using `crane copy`. This is a **manifest-level re-tag, not
a rebuild** — it touches only Docker Hub's image index, takes seconds,
and is atomic.
The reason this happens *after* variants succeed (rather than alongside
`build-base`) is so a partial failure leaves `base-latest` pointing at
the previous known-good base. External consumers who pin to
`base-latest` (e.g. the planned pi-devbox repo) never see a broken base.
### Step 6: `update-description`
Push the generated `DOCKER_HUB.md` to the Hub repo's `full_description`
field via the Hub REST API. Same step as the production pipeline.
## NPM_CONFIG_PREFIX gotcha (variant override pattern)
The base sets
```
ENV NPM_CONFIG_PREFIX=/home/developer/.pi/npm-global
```
This is intentional — it makes `pi install npm:<pkg>` and `npm install -g`
land on the `devbox-pi-config` named volume at runtime, so user-installed
packages survive container recreate AND image rebuild.
But the *variant build* inherits this prefix at build time. If left as-is,
`npm install -g opencode-ai@$VERSION` in `Dockerfile.variant` would
install opencode into `/home/developer/.pi/npm-global/...`, which is then
**shadowed by the volume mount at runtime** → opencode disappears from
PATH on first start.
Fix: each `npm install -g` in `Dockerfile.variant` overrides the prefix
per-RUN:
```dockerfile
RUN NPM_CONFIG_PREFIX=/usr npm install -g opencode-ai@${OPENCODE_VERSION}
```
Baked binaries land on `/usr/bin/...` (system prefix), survive the volume
mount. Runtime-installed user packages still land on
`~/.pi/npm-global/...`. Both visible on PATH.
## Cache strategy
Two registry caches are configured:
```yaml
cache-from: type=registry,ref=joakimp/opencode-devbox:base-buildcache
cache-to: type=registry,ref=joakimp/opencode-devbox:base-buildcache,mode=max
cache-from: type=registry,ref=joakimp/opencode-devbox:base-variant-buildcache
cache-to: type=registry,ref=joakimp/opencode-devbox:base-variant-buildcache,mode=max
```
`mode=max` exports cache for *all* layers, not just the final image's
layers. Important for multi-arch builds where the cross-arch layer reuse
matters more.
## Wall-clock estimates
| Scenario | Production pipeline | Split-base pipeline |
|---|---|---|
| Version-bump-only release (only opencode/pi/omos version changed) | ~165180 min | **~3040 min** (base cache hit) |
| Base-touching release (apt/Node/Debian/entrypoint change) | ~165180 min | **~7090 min** (base rebuilds) |
The split-base pipeline pays its dues on base-touching releases (which are
infrequent — a few times a year for Debian / Node major version bumps).
Most releases are version-bumps and ride the cache.
## Validate workflow
[`validate.yml`](workflows/validate.yml) is the lightweight gate that runs
on every push to `main` and on PRs. It:
1. Runs `scripts/generate-dockerhub-md.py --check` to enforce
`DOCKER_HUB.md` is in sync with `HUB_TEMPLATE`.
2. Builds each of the four variants amd64-only (no multi-arch, no push)
and runs `scripts/smoke-test.sh`.
This catches regressions before they reach a tag push. Wall clock ~30 min.
## Runner expectations
- **Image:** `catthehacker/ubuntu:act-latest`. Each job runs inside a
fresh container of this image. Don't assume any pre-installed
toolchains beyond what catthehacker ships.
- **Disk pressure:** the runner host has ~40 GB of usable overlay space,
often 70%+ used at job start. Every job that does `load: true` (smoke)
starts with a `Reclaim runner disk` step that strips
catthehacker-resident toolchains (Android SDK, .NET, Swift, GHC, JVM,
Boost, Chromium, PowerShell) and prunes stale docker state. Don't
remove these steps without testing on a fresh runner.
- **Concurrency:** 2 runners. Jobs in the same workflow run can fan out to
both; jobs in *different* workflow runs are serialized by gitea's queue.
The `concurrency: { group: ${{ workflow }}-${{ ref }}, cancel-in-progress: false }`
setting keeps tag pushes from racing each other but allows
per-PR/per-branch parallelism.
- **Workflow visibility in UI:** gitea Actions only surfaces workflows
from the **default branch** in the web UI's workflow list, even for
`workflow_dispatch` triggers. Workflows on feature branches are
invisible until merged to `main`.
- **Disk reclaim quirk:** `actions/{upload,download}-artifact@v4+` does
not work on Gitea (depends on a GitHub-only Artifact API). Stick to
`@v3` if matrix-fanout-with-artifacts is ever needed. We avoided this
by using `docker/build-push-action@v7` with comma-separated
`platforms: linux/amd64,linux/arm64` — natively does multi-arch push
in a single job, no artifact dance.
## Migration plan: split-base → production
1. **Validate the split-base dispatch.** Trigger
`docker-publish-split.yml` manually with `release_tag=v0.0.0-split-test`
and `promote_latest=false`. Confirm all jobs go green, image sizes
match the production baseline within ~10%, and no unexpected layer
rebuilds appear in `build-variant-*` logs after the FROM line.
2. **Run a second dispatch** to confirm cache-hit behavior:
`base-decide` should set `need_build=false`, `build-base` should be
skipped entirely, total wall clock should drop to ~2540 min.
3. **Cut over.** In a single commit:
- Edit `docker-publish-split.yml`: change `on: workflow_dispatch:` to
`on: push: tags: v*` and wire `$GITHUB_REF` into the `release_tag`
input, set `promote_latest=true` for production runs.
- Delete `docker-publish.yml`.
- Delete the original `Dockerfile` (keep `Dockerfile.base` +
`Dockerfile.variant`).
- Update `CHANGELOG.md`: promote the "Build pipeline" Unreleased entry.
4. **Tag a release.** First production release on the new pipeline. Watch
it like a hawk for the first run.
## Related docs
- [`AGENTS.md`](../AGENTS.md) — domain facts, release-day checklist,
documentation coupling rules. Read first when modifying CI behavior.
- [`CHANGELOG.md`](../CHANGELOG.md) — the build pipeline rewrite is
recorded under `Unreleased` until the cutover lands.
- `Dockerfile`, `Dockerfile.base`, `Dockerfile.variant` — production
single-Dockerfile build and the split-base counterparts. Comments at
the top of each explain its role.
- [`scripts/smoke-test.sh`](../scripts/smoke-test.sh) — invoked by all
three workflows; this is the single source of truth for "what does a
built image have to satisfy".
- [`scripts/generate-dockerhub-md.py`](../scripts/generate-dockerhub-md.py)
— generates `DOCKER_HUB.md` from `HUB_TEMPLATE`. `--check` enforces
sync in `validate.yml`.
+1
View File
@@ -15,6 +15,7 @@ Docker image packaging [opencode](https://opencode.ai) into a production-ready d
- `scripts/generate-dockerhub-md.py` — generates `DOCKER_HUB.md` from a hand-maintained `HUB_TEMPLATE` constant. `--check` fails if the committed file is out of sync (enforced by the `validate` workflow).
- `DOCKER_HUB.md`**auto-generated** from `HUB_TEMPLATE` in `scripts/generate-dockerhub-md.py`. Do not edit directly. Pushed to Docker Hub description via CI API call. Must stay under 25 kB. Short description field must be ≤100 bytes.
- `README.md` — authoritative source documentation for everything in this repo. Independent of `DOCKER_HUB.md`: the Hub doc is hand-maintained in the generator's `HUB_TEMPLATE` and intentionally slim, linking back to the gitea README for depth.
- `.gitea/README.md`**read this first** if you're touching CI. Architectural overview of the build pipeline (production vs split-base), wall-clock estimates, NPM_CONFIG_PREFIX gotcha, runner expectations, migration plan.
- `.gitea/workflows/validate.yml` — lightweight amd64 build + smoke test on push to main and PRs. Also runs the DOCKER_HUB.md sync check.
- `.gitea/workflows/docker-publish.yml` — production CI pipeline on tag push: smoke-test each variant on amd64, then full multi-arch (amd64 + arm64) build-and-push, then update Docker Hub description.
- `.gitea/workflows/docker-publish-split.yml`**WIP, branch `feat/split-build` only.** Two-phase split-base pipeline. Triggers on `workflow_dispatch` only so it runs alongside the production pipeline without conflict. Pushes to user-supplied `release_tag` input (e.g. `v0.0.0-split-test`); `latest*` aliases only updated when `promote_latest: true`. Compute base hash, conditionally build base, then 4 variant deltas in parallel.
+14
View File
@@ -8,12 +8,26 @@ Tags follow `v{opencode_version}[letter]` — bare tag for the first build on a
## Unreleased
Docs:
- **New: `.gitea/README.md`** — architectural overview of the build pipeline. Documents the production single-Dockerfile path vs the merged-but-unvalidated split-base path, hash-driven base cache reuse, wall-clock estimates, the `NPM_CONFIG_PREFIX` variant-override pattern, runner expectations (catthehacker container, disk reclaim, concurrency, gitea-actions @v4 artifact gotcha), and the cutover plan. Auto-renders when navigating to `.gitea/` in the gitea web UI. Linked from `AGENTS.md` as the first thing to read when touching CI.
Build pipeline (merged to main as `Dockerfile.base` + `Dockerfile.variant` + `.gitea/workflows/docker-publish-split.yml`, NOT yet validated end-to-end — the `workflow_dispatch` test against `:base-<hash>` + `:v0.0.0-split-test*` aliases is still the gating step before this can take over `on: push: tags: v*`):
- **New: split-base build pipeline.** `Dockerfile.base` (variant-independent layers — apt, locales, AWS CLI, Node.js, mempalace, gitea-mcp, user setup, chromadb prewarm, ENVs, entrypoints) builds once and is published as `joakimp/opencode-devbox:base-<sha>`. `Dockerfile.variant` `FROM`s that base and adds only opencode/omos/pi installs (or skips them per build-args). Companion workflow `.gitea/workflows/docker-publish-split.yml` runs as a `workflow_dispatch`-only pipeline alongside the existing `docker-publish.yml` so they don't conflict. Hash-driven base reuse: a content hash of `Dockerfile.base + rootfs/ + entrypoint*.sh` becomes the base tag; if the tag already exists on Docker Hub, the base build is skipped entirely. Estimated wall clock: version-bump-only release ~3040 min (vs ~165180 min today); base-touching release ~6070 min. Trade-off: two Dockerfiles to maintain, and `npm install -g` in the variant must override `NPM_CONFIG_PREFIX=/usr` per-RUN to keep baked binaries off the volume-shadowed path. Once 12 successful workflow_dispatch runs validate the output against the existing pipeline, the new workflow takes over `on: push: tags: v*` and the original is retired.
## v1.14.44 — 2026-05-09
opencode 1.14.42 → 1.14.44 bump (1.14.43 skipped upstream). Also completes the matrix coverage that v1.14.42 missed: `build-omos-with-pi` failed mid-publish on v1.14.42 due to an upstream npm CDN propagation race — `oh-my-opencode-slim@1.0.7` had been published declaring a dependency on `@opencode-ai/sdk@1.14.44`, and our build hit the registry within ~2 minutes of that SDK version landing, before the tarball had propagated across npm's CDN. The build returned 404 on the SDK fetch even though the manifest's `dist-tags.latest` already pointed at 1.14.44. Tarball is now fully fetchable; v1.14.44 builds cleanly across all four variants.
- **Bump:** opencode 1.14.42 → 1.14.44 (`OPENCODE_VERSION` build-arg default in both `Dockerfile` and `Dockerfile.variant`).
Known gap: `joakimp/opencode-devbox:v1.14.42-omos-with-pi` and the corresponding `latest-omos-with-pi` alias were NOT published in the v1.14.42 release (`build-omos-with-pi` job failed for the reason above). `latest-omos-with-pi` continued pointing at v1.14.41b until v1.14.44 published. Users on the `latest-omos-with-pi` floating tag were unaffected; users pulling explicit `:v1.14.42-omos-with-pi` would get a 404 from Hub. Closed by v1.14.44.
## v1.14.42 — 2026-05-09
**Note:** Of the 4 multi-arch variants, 3 published cleanly (`v1.14.42`, `v1.14.42-omos`, `v1.14.42-with-pi`, plus their `latest*` aliases). `build-omos-with-pi` failed during the publish step due to an upstream npm CDN propagation race (see v1.14.44 entry above for detail). Re-running the failed job would have required another full ~3h matrix rerun in gitea Actions; we chose to bump opencode to 1.14.44 instead and let the next tag close the gap.
opencode 1.14.41 → 1.14.42 bump. Carries along all container-side changes accumulated since v1.14.41b: pi package rename to `@earendil-works/*`, npm-prefix-on-volume fix, Hub doc rewrite, README/AGENTS docs catchup.
Image changes:
+1 -1
View File
@@ -5,7 +5,7 @@ ARG DEBIAN_VERSION=trixie-slim
FROM debian:${DEBIAN_VERSION} AS base
ARG TARGETARCH
ARG OPENCODE_VERSION=1.14.42
ARG OPENCODE_VERSION=1.14.44
LABEL maintainer="joakimp"
LABEL description="Portable opencode developer container"
+1 -1
View File
@@ -32,7 +32,7 @@ ARG USER_NAME=developer
# ── Install opencode via npm ─────────────────────────────────────────
ARG INSTALL_OPENCODE=true
ARG OPENCODE_VERSION=1.14.42
ARG OPENCODE_VERSION=1.14.44
RUN if [ "${INSTALL_OPENCODE}" = "true" ]; then \
NPM_CONFIG_PREFIX=/usr npm install -g opencode-ai@${OPENCODE_VERSION} && \
opencode --version ; \