perf(build): keep test suites out of the image and cache CI builds durably
CI / check (push) Successful in 30s
CI / tests-unit (push) Successful in 1m53s
CI / tests-integration (push) Successful in 1m54s
CI / tests-imaging-browsers (push) Successful in 2m26s
CI / tests-quality (push) Successful in 2m19s
CI / tests-ui (push) Successful in 4m31s
CI / preflight (push) Skipped
CI / deploy (push) Failing after 3m9s

Three build-time improvements, all measured on this host.

The build context no longer ships the test suites. `next build` type-checks
the whole project, so `e2e/` and `integration/` in the context made the image
build fail on a type error in test code and pulled @playwright/test, vitest,
msw, testcontainers and drizzle-kit into the image. The CI runs those suites
on the host. Type-check time in the build drops from 2.2s to 1.3s, and a
contract test now fails if application code ever imports from those folders.

The CI now exports the build cache to a directory outside the runner
workspace: /var/cache/epicnext-cms-buildcache for the release and a separate
map for branch preflights, so a branch can never feed the release cache. The
workspace is wiped after every job and `docker builder prune` runs after
every release, so the next build started with an empty cache and downloaded
all 493 packages again (measured: reused 0, downloaded 493). With the
durable export the frozen dependency layer is reused (measured: reused 493,
downloaded 0) even after a full prune. A cache directory that is not
writable only produces a warning; the release is never blocked.

Investigated the remaining ~30s of Next.js build tracing: excluding the
dev-only packages with `outputFileTracingExcludes` made it slower (35s vs
30s), because the traced output was already minimal (5 packages). The cost
is Next's own page-config collection and file tracing over 329 routes, not
the dependency set, so no configuration was kept.
This commit is contained in:
openhands committed 2026-10-21 19:29:50 +02:00
1 parent 8ca538f99f
commit ecd8e5852c
8 files changed
+148 -2

No files matched your search

+18
View File
@@ -329,6 +329,24 @@ Two caches make repeated builds cheap:
daemon cache to 4 GB after every deploy. The directory is ignored by both
Git and the build context; delete it to force a cold build.
The CI has the same problem on a bigger scale: the runner workspace is wiped
after every job and `scripts/docker-prune.sh` trims the daemon cache after
every release, so the next build started with an empty cache and downloaded all
493 packages again (measured: `reused 0, downloaded 493`). `scripts/ci-deploy.sh`
and `scripts/ci-preflight.sh` therefore export the build cache to a directory
outside the workspace — `/var/cache/epicnext-cms-buildcache` for the release and
`/var/cache/epicnext-cms-preflight-cache` for branch preflights, so a branch can
never feed the release cache. With that export in place the same build re-uses
the frozen dependency layer (`reused 493, downloaded 0`) even after a full
`docker builder prune`. Set `CMS_BUILD_CACHE_DIR` to another path to move the
cache, or to an empty value to switch it off.
The build context excludes `e2e/` and `integration/`. `next build` type-checks
the whole project, so shipping the test suites inside the image made a type
error in test code break the image build and pulled `@playwright/test`, `vitest`,
`msw` and `testcontainers` into it. The CI runs those suites on the host
(`pnpm test:ui`, `pnpm test:integration`).
### Volumes
Only the CMS (and the optional avatar imager container, see [Avatar Imaging](#avatar-imaging)) run in Docker. The heavy client assets and gamedata stay on the host and are shared into the container so imports and uploads persist: