Files
jorgecuadros-platform/.gitea/workflows/deploy.yml
T
rmancinasandClaude Opus 5 7e3b530174
Build and Push Images / Build jorgecuadros-web (push) Successful in 1m46s
Build and Push Images / Build jorgecuadros-api (push) Successful in 1m58s
fix(deploy): pull images explicitly, and detect api/web drift by commit
The first successful galactus deploy came up all-green while the web tier was
running a build from two commits earlier. The registry held web:latest from
3ff56e6; the host still had a web:latest cached from 4ee7ec7; the deploy
reported success and served the old one. The API was only current because it
had been pulled by hand during earlier debugging.

Two independent failures, both fixed here.

1. Images are not pulled. The deploy action's `pull: true` does not reliably
   refresh an already-cached moving tag on a standalone endpoint. Added a
   Pull images step (deploy/scripts/pull-images.mjs) that pulls each image
   through Portainer's Docker API with registry credentials and fails the
   deploy if a pull fails — note the endpoint answers 200 even when the pull
   errored, so the stream body has to be inspected, not just the status.

2. The drift check could not see it. Both the verify step and the web footer
   compared APP_VERSION, but on a branch build BOTH tiers report "master", so
   equality proved nothing. They now compare gitSha, which is the only field
   that differs between two builds of the same branch. api and web come from
   one matrix run, so a difference can only mean an image was not replaced.

   This needed a /version on the web tier too — previously its build identity
   was only readable by scraping window.__APP_BUILD__ out of the HTML.

pull-images.mjs builds the X-Registry-Auth header as URL-safe base64 WITH
padding: Node's "base64url" omits the padding and Portainer's Go decoder
rejects it with "Illegal base64 data at input byte N".

Verified against galactus: pulls both images, and exits non-zero on a
nonexistent tag.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-07-30 14:57:44 -07:00

316 lines
14 KiB
YAML

# Manual PROD deploy to Portainer.
#
# This does NOT build — build.yml already builds + pushes the api/web images.
# This workflow (re)applies the deploy/*.stack.yml files to the Portainer Swarm.
# Trigger it by hand from the Actions tab ("Run workflow") and choose:
# - tag: which already-published image tag to ship (default: latest)
# - scope: how much to deploy
# app = web + api only (the usual app release) [default]
# full = db + minio + web + api (bring up / update the whole platform)
#
# The `tag` input carries NO leading `v`: metadata-action's {{version}} turns
# git tag v1.2.3 into image tag 1.2.3. Tag v1.2.3, dispatch 1.2.3.
#
# Order: db+minio (full only) -> pre-migrate backup -> prisma migrate deploy ->
# app -> verify the API reports the version you asked for. Rollback = dispatch
# an older tag; that rolls back CODE only, never the schema, which is why every
# schema change must be expand/contract. See docs/DEPLOY_AND_MIGRATIONS.md.
#
# galactus (the office server) is standalone Docker, not this Swarm — it has its
# own workflow, .gitea/workflows/deploy-galactus.yml.
#
# cssnr/portainer-stack-deploy-action creates each stack on first run and updates
# it on every run, so no manual stack pre-creation in the Portainer UI. On a
# `full` deploy the db + minio stacks are applied BEFORE the app (the API depends
# on them). db + minio are stateful + pinned to node label jorgecuadros_db=true
# (see their stack files) — re-applying them is idempotent and keeps their data.
#
# Prereqs (once):
# - one swarm node labelled jorgecuadros_db=true (db + minio + api pin there).
# - Gitea repo secrets set (Settings > Actions > Secrets):
# # Portainer
# PORTAINER_URL https://192.168.4.212:9443
# PORTAINER_API_KEY Portainer access token
# PORTAINER_ENDPOINT_ID 2 (the local Swarm endpoint)
# PORTAINER_APP_STACK_NAME e.g. jorgecuadros-prod-app
# PORTAINER_DB_STACK_NAME e.g. jorgecuadros-prod-db (full only)
# PORTAINER_MINIO_STACK_NAME e.g. jorgecuadros-prod-minio (full only)
# # App runtime
# DATABASE_URL mysql://jorgecuadros:<pass>@192.168.4.212:3306/jorgecuadros
# SESSION_SECRET 64-hex (openssl rand -hex 32)
# APP_API_ORIGIN http://192.168.4.212:3001 (browser-facing API URL)
# APP_WEB_ORIGIN http://192.168.4.212:3000 (web public origin, API CORS)
# APP_S3_ENDPOINT http://192.168.4.212:9000 (server-side minio URL)
# # Object storage (app + minio stack)
# MINIO_ROOT_USER minio access key
# MINIO_ROOT_PASSWORD minio secret key
# # Database stack (full only)
# MYSQL_PASSWORD app-user password (matches DATABASE_URL)
# MYSQL_ROOT_PASSWORD mysql root password
# - the runner must reach BOTH Portainer (9443) and MySQL (3306) — the
# migration step connects to the database directly. If it cannot reach 3306,
# migrate by hand and dispatch with skip_migrate=true.
# - ONE-TIME on a database built with `prisma db push` (i.e. every database
# that exists today): baseline it before the first run, or the migrate step
# fails with P3005 "database schema is not empty":
# npx prisma@5 migrate resolve --applied 0000_init \
# --schema packages/database/prisma/schema.prisma
name: Deploy to Portainer
on:
workflow_dispatch:
inputs:
tag:
description: "Image tag to deploy (latest, sha-<short>, or vX.Y.Z)"
required: true
default: "latest"
scope:
description: "What to deploy"
type: choice
required: true
default: "app"
options:
- app
- full
bootstrap:
description: "First-ever deploy: allow the pre-migrate backup to be skipped when no API container exists yet"
type: boolean
required: false
default: false
skip_migrate:
description: "Skip prisma migrate deploy (use when the runner cannot reach MySQL and you migrated by hand)"
type: boolean
required: false
default: false
env:
REGISTRY: git.mancinas.io
jobs:
deploy:
name: Deploy (${{ github.event.inputs.scope }})
runs-on: docker
container:
image: node:20-alpine
steps:
- name: Install tools
# openssl: prisma's migration engine picks its musl/openssl build at
# runtime and cannot resolve one without it.
run: apk add --no-cache openssl ca-certificates git
- uses: actions/checkout@v4
# An unset secret arrives as an empty string, and the deploy action then
# fails with "Input required and not supplied: token" — which names the
# action's input, not the secret you forgot.
- name: Preflight — required secrets
env:
PORTAINER_URL: ${{ secrets.PORTAINER_URL }}
PORTAINER_API_KEY: ${{ secrets.PORTAINER_API_KEY }}
PORTAINER_ENDPOINT_ID: ${{ secrets.PORTAINER_ENDPOINT_ID }}
PORTAINER_APP_STACK_NAME: ${{ secrets.PORTAINER_APP_STACK_NAME }}
PORTAINER_DB_STACK_NAME: ${{ secrets.PORTAINER_DB_STACK_NAME }}
PORTAINER_MINIO_STACK_NAME: ${{ secrets.PORTAINER_MINIO_STACK_NAME }}
DATABASE_URL: ${{ secrets.DATABASE_URL }}
SESSION_SECRET: ${{ secrets.SESSION_SECRET }}
APP_API_ORIGIN: ${{ secrets.APP_API_ORIGIN }}
APP_WEB_ORIGIN: ${{ secrets.APP_WEB_ORIGIN }}
APP_S3_ENDPOINT: ${{ secrets.APP_S3_ENDPOINT }}
MINIO_ROOT_USER: ${{ secrets.MINIO_ROOT_USER }}
MINIO_ROOT_PASSWORD: ${{ secrets.MINIO_ROOT_PASSWORD }}
MYSQL_PASSWORD: ${{ secrets.MYSQL_PASSWORD }}
MYSQL_ROOT_PASSWORD: ${{ secrets.MYSQL_ROOT_PASSWORD }}
SCOPE: ${{ github.event.inputs.scope }}
run: |
REQUIRED="PORTAINER_URL PORTAINER_API_KEY PORTAINER_ENDPOINT_ID
PORTAINER_APP_STACK_NAME DATABASE_URL SESSION_SECRET
APP_API_ORIGIN APP_WEB_ORIGIN APP_S3_ENDPOINT
MINIO_ROOT_USER MINIO_ROOT_PASSWORD"
if [ "$SCOPE" = "full" ]; then
REQUIRED="$REQUIRED PORTAINER_DB_STACK_NAME PORTAINER_MINIO_STACK_NAME
MYSQL_PASSWORD MYSQL_ROOT_PASSWORD"
fi
missing=""
for name in $REQUIRED; do
eval "value=\${$name}"
[ -z "$value" ] && missing="$missing $name"
done
if [ -n "$missing" ]; then
echo "::error::missing repo secrets:$missing"
echo "::error::set them under Settings > Actions > Secrets"
exit 1
fi
echo "all required secrets present for scope=$SCOPE"
# --- full only: database ---------------------------------------------
- name: Deploy database stack
if: ${{ github.event.inputs.scope == 'full' }}
uses: cssnr/portainer-stack-deploy-action@v1
with:
url: ${{ secrets.PORTAINER_URL }}
token: ${{ secrets.PORTAINER_API_KEY }}
name: ${{ secrets.PORTAINER_DB_STACK_NAME }}
file: deploy/jorgecuadros-db.stack.yml
type: file
endpoint: ${{ secrets.PORTAINER_ENDPOINT_ID }}
env_data: |
{
"MYSQL_SERVER_ID": "1",
"MYSQL_PORT": "3306",
"MYSQL_DATABASE": "jorgecuadros",
"MYSQL_USER": "jorgecuadros",
"MYSQL_PASSWORD": "${{ secrets.MYSQL_PASSWORD }}",
"MYSQL_ROOT_PASSWORD": "${{ secrets.MYSQL_ROOT_PASSWORD }}"
}
# --- full only: object storage ---------------------------------------
- name: Deploy minio stack
if: ${{ github.event.inputs.scope == 'full' }}
uses: cssnr/portainer-stack-deploy-action@v1
with:
url: ${{ secrets.PORTAINER_URL }}
token: ${{ secrets.PORTAINER_API_KEY }}
name: ${{ secrets.PORTAINER_MINIO_STACK_NAME }}
file: deploy/jorgecuadros-minio.stack.yml
type: file
endpoint: ${{ secrets.PORTAINER_ENDPOINT_ID }}
env_data: |
{
"MINIO_API_PORT": "9000",
"MINIO_CONSOLE_PORT": "9001",
"MINIO_ROOT_USER": "${{ secrets.MINIO_ROOT_USER }}",
"MINIO_ROOT_PASSWORD": "${{ secrets.MINIO_ROOT_PASSWORD }}"
}
# --- restore point, taken while the OLD api container is still up ------
# Dumped INSIDE the running api container so the file lands in the volume
# the "Operaciones" restore screen reads — a dump on the runner would be
# unreachable by the only restore path this platform has.
- name: Pre-migrate backup
env:
PORTAINER_URL: ${{ secrets.PORTAINER_URL }}
PORTAINER_API_KEY: ${{ secrets.PORTAINER_API_KEY }}
PORTAINER_ENDPOINT_ID: ${{ secrets.PORTAINER_ENDPOINT_ID }}
DATABASE_URL: ${{ secrets.DATABASE_URL }}
BACKUP_TAG: ${{ github.event.inputs.tag }}
ALLOW_MISSING_CONTAINER: ${{ github.event.inputs.bootstrap }}
# Portainer serves a self-signed certificate. Scoped to this step
# only, which does nothing but talk to Portainer.
NODE_TLS_REJECT_UNAUTHORIZED: "0"
run: node deploy/scripts/pre-migrate-backup.mjs
# --- schema, forward-only ---------------------------------------------
# Prisma has no down-migrations: a code rollback does NOT roll the schema
# back. See docs/DEPLOY_AND_MIGRATIONS.md — every change must be
# expand/contract so the previous release still runs against the new
# schema. Run as a deploy STEP, never as the container CMD: N replicas
# would race each other applying the same migration.
- name: Apply database migrations
if: ${{ github.event.inputs.skip_migrate != 'true' }}
env:
DATABASE_URL: ${{ secrets.DATABASE_URL }}
run: |
set -e
SCHEMA=packages/database/prisma/schema.prisma
npx --yes prisma@5 migrate status --schema "$SCHEMA" || true
if ! npx --yes prisma@5 migrate deploy --schema "$SCHEMA"; then
echo "::error::migrate deploy failed. If this is P3005 (schema not empty),"
echo "::error::the database predates migration history — baseline it once with:"
echo "::error:: npx prisma@5 migrate resolve --applied 0000_init --schema $SCHEMA"
exit 1
fi
# --- make sure the host actually has the images ------------------------
# The deploy action's `pull: true` does not reliably refresh an already
# cached moving tag; without this a "successful" deploy can leave the host
# serving an older build of the same tag.
- name: Pull images
env:
PORTAINER_URL: ${{ secrets.PORTAINER_URL }}
PORTAINER_API_KEY: ${{ secrets.PORTAINER_API_KEY }}
PORTAINER_ENDPOINT_ID: ${{ secrets.PORTAINER_ENDPOINT_ID }}
REGISTRY: ${{ env.REGISTRY }}
REGISTRY_USERNAME: ${{ secrets.REGISTRY_USERNAME }}
REGISTRY_PASSWORD: ${{ secrets.REGISTRY_PASSWORD }}
IMAGES: ${{ github.repository_owner }}/jorgecuadros-api,${{ github.repository_owner }}/jorgecuadros-web
TAG: ${{ github.event.inputs.tag }}
NODE_TLS_REJECT_UNAUTHORIZED: "0"
run: node deploy/scripts/pull-images.mjs
# --- always: the app (web + api) -------------------------------------
- name: Deploy app stack
uses: cssnr/portainer-stack-deploy-action@v1
with:
url: ${{ secrets.PORTAINER_URL }}
token: ${{ secrets.PORTAINER_API_KEY }}
name: ${{ secrets.PORTAINER_APP_STACK_NAME }}
file: deploy/jorgecuadros-app.stack.yml
type: file
pull: true
endpoint: ${{ secrets.PORTAINER_ENDPOINT_ID }}
env_data: |
{
"APP_TAG": "${{ github.event.inputs.tag }}",
"API_PORT": "3001",
"WEB_PORT": "3000",
"S3_BUCKET": "jorgecuadros-documents",
"API_ORIGIN": "${{ secrets.APP_API_ORIGIN }}",
"WEB_ORIGIN": "${{ secrets.APP_WEB_ORIGIN }}",
"S3_ENDPOINT": "${{ secrets.APP_S3_ENDPOINT }}",
"DATABASE_URL": "${{ secrets.DATABASE_URL }}",
"SESSION_SECRET": "${{ secrets.SESSION_SECRET }}",
"MINIO_ROOT_USER": "${{ secrets.MINIO_ROOT_USER }}",
"MINIO_ROOT_PASSWORD": "${{ secrets.MINIO_ROOT_PASSWORD }}"
}
# --- prove it ----------------------------------------------------------
# A stack naming a tag is not proof the container is running it — a
# skipped pull leaves the old code up. Ask the API what it actually is.
- name: Verify running version
env:
API_ORIGIN: ${{ secrets.APP_API_ORIGIN }}
WEB_ORIGIN: ${{ secrets.APP_WEB_ORIGIN }}
WANT: ${{ github.event.inputs.tag }}
run: |
set -e
apk add --no-cache curl >/dev/null
fetch_version() {
for i in $(seq 1 30); do
if curl -fsS "$1/version" > "$2"; then return 0; fi
echo "waiting for $1 ($i/30)..."
sleep 5
done
echo "::error::$1/version never answered"
return 1
}
fetch_version "$API_ORIGIN" /tmp/api.json
fetch_version "$WEB_ORIGIN" /tmp/web.json
cat /tmp/api.json; echo; cat /tmp/web.json; echo
API_SHA=$(node -e 'console.log(require("/tmp/api.json").gitSha)')
WEB_SHA=$(node -e 'console.log(require("/tmp/web.json").gitSha)')
API_VER=$(node -e 'console.log(require("/tmp/api.json").version)')
# Compare the COMMIT, not the version string: on a branch build both
# tiers report "master", so version equality proves nothing.
if [ "$API_SHA" != "$WEB_SHA" ]; then
echo "::error::api and web are different builds — api $API_SHA, web $WEB_SHA"
echo "::error::one of the images was not replaced; check the Pull images step"
exit 1
fi
echo "api and web agree: $API_SHA"
case "$WANT" in
[0-9]*.[0-9]*.[0-9]*)
if [ "$API_VER" != "$WANT" ]; then
echo "::error::deployed $WANT but the API reports $API_VER"
exit 1
fi
echo "verified: running $API_VER"
;;
*)
echo "dispatched '$WANT'; tiers report '$API_VER' (not directly comparable)"
;;
esac