teqo-backup/Dockerfile
Francisco Solla cf95fbabc0 fix(deploy): assa o pnpm na imagem do migrator — destrava o deploy staging
O primeiro deploy staging desde a OPS103 falhou nas migrations (run 34765554681,
job deploy-staging): a imagem do migrator roda `pnpm migrate` no start do
container via corepack e o cache do corepack do build da stage `deps` não é
copiado para a imagem, então o runtime tenta baixar pnpm do registry.npmjs.org
e falhou com ETIMEDOUT/ENETUNREACH (o mesmo download funcionou no build minutos
antes e no migrator de produção). Produção ficou skipped e intocada.

`RUN corepack install` na stage `AS migrator` baixa o pnpm (versão do
packageManager) em build time e assa o cache na layer — o runtime fica offline.
Prova: `docker run --network none <imagem> pnpm --version` = 10.11.0 com o bake
e o erro exato do staging sem ele. Pin novo em deployScript.unit.spec.ts;
post-mortem e changelog.
2026-09-13 13:03:29 -03:00

99 lines
3.7 KiB
Docker

# syntax=docker/dockerfile:1
# From https://github.com/vercel/next.js/blob/canary/examples/with-docker/Dockerfile
FROM node:24.19.0-alpine AS base
RUN apk add --no-cache libc6-compat && corepack enable
# Install dependencies only when needed
FROM base AS deps
WORKDIR /app
# pnpm-workspace.yaml carries onlyBuiltDependencies — pnpm 10 no longer reads the "pnpm" field.
COPY package.json pnpm-lock.yaml pnpm-workspace.yaml ./
RUN pnpm install --frozen-lockfile
# One-shot image used to apply committed Payload migrations before a release.
FROM base AS migrator
WORKDIR /app
COPY --from=deps /app/node_modules ./node_modules
COPY . .
# Bake the package manager INTO the image: `pnpm migrate` runs at CONTAINER
# START, so without a corepack cache the runtime downloads pnpm from
# registry.npmjs.org — the maintenance container's outbound is not guaranteed
# (the staging deploy of 2026-09-13 failed exactly there, ETIMEDOUT to the
# registry, while the same download worked at build time). The install runs at
# BUILD time, where the network is available, and the cache ships in the layer.
RUN corepack install
CMD ["pnpm", "migrate"]
# Rebuild the source code only when needed
FROM base AS builder
WORKDIR /app
COPY --from=deps /app/node_modules ./node_modules
COPY . .
ARG NEXT_PUBLIC_SITE_URL
ENV NEXT_PUBLIC_SITE_URL=$NEXT_PUBLIC_SITE_URL
ENV NEXT_TELEMETRY_DISABLED=1
# Static generation reads Payload data. BuildKit mounts keep credentials out of
# layers and image history; migrations run separately through the migrator —
# the deploy script applies them BEFORE this build (OPS66: static routes read
# the new schema, so the build must see a migrated DB).
#
# OPS99: Generate importMap with dummy S3_* envs BEFORE next build.
# This ensures the production image always has the S3ClientUploadHandler entry,
# eliminating the class of bugs where the admin goes blank (OPS69/OPS72/OPS73).
# The dummy values are safe because Payload's generator doesn't connect to S3.
# Shell prefix assignments bind to a single command, so each side of the `&&`
# declares its own env: the generator runs through `pnpm generate:importmap`
# (which owns the `--conditions=react-server` the Payload CLI needs to import
# server-only modules), and `next build` carries the standalone flag itself.
RUN --mount=type=secret,id=database_url,env=DATABASE_URL \
--mount=type=secret,id=payload_secret,env=PAYLOAD_SECRET \
S3_BUCKET=build-dummy \
S3_ENDPOINT=http://127.0.0.1:3900 \
S3_ACCESS_KEY_ID=build-dummy \
S3_SECRET_ACCESS_KEY=build-dummy \
pnpm generate:importmap && \
NEXT_OUTPUT_STANDALONE=1 \
NODE_OPTIONS="--no-deprecation --max-old-space-size=8000" \
pnpm exec next build
# Production image, copy all the files and run next
FROM base AS runner
WORKDIR /app
ENV NODE_ENV=production
ENV NEXT_TELEMETRY_DISABLED=1
RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs
# Remove this line if you do not have this folder
COPY --from=builder /app/public ./public
# Set the correct permission for prerender cache
RUN mkdir .next
RUN chown nextjs:nodejs .next
# Automatically leverage output traces to reduce image size
# https://nextjs.org/docs/advanced-features/output-file-tracing
COPY --from=builder --chown=nextjs:nodejs /app/.next/standalone ./
COPY --from=builder --chown=nextjs:nodejs /app/.next/static ./.next/static
USER nextjs
EXPOSE 3000
ENV PORT=3000
ENV HOSTNAME=0.0.0.0
# server.js is created by next build from the standalone output
# https://nextjs.org/docs/pages/api-reference/next-config-js/output
HEALTHCHECK --interval=30s --timeout=5s --start-period=20s --retries=3 \
CMD node -e "fetch('http://127.0.0.1:3000/').then((response) => { if (!response.ok) process.exit(1) }).catch(() => process.exit(1))"
CMD ["node", "server.js"]