Files
padelnomics/deploy.sh
Deeman 5f7e8f1200 fix(deploy): move router config write to after health check passes
Router had no profile so it was always included in `up -d --wait`.
Writing the new target's config BEFORE the wait caused the router to become
unhealthy if the new slot failed — leaving it in a broken state for the next
deploy attempt.

Now: router keeps its old config (pointing to the still-running old slot)
during the health check wait, so it stays healthy throughout. Config is only
written and nginx -s reload triggered after the new slot passes its health
check. This is the correct blue-green pattern.

Also add `retries: 3` and `start_period: 10s` to the router health check
for resilience against transient startup failures.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-22 13:22:50 +01:00

99 lines
3.6 KiB
Bash
Executable File

#!/usr/bin/env bash
set -euo pipefail
COMPOSE="docker compose -f docker-compose.prod.yml"
LIVE_FILE=".live-slot"
ROUTER_CONF="router/default.conf"
# ── Determine slots ─────────────────────────────────────────
CURRENT=$(cat "$LIVE_FILE" 2>/dev/null || echo "none")
if [ "$CURRENT" = "blue" ]; then
TARGET="green"
else
TARGET="blue"
fi
echo "==> Current: $CURRENT → Deploying: $TARGET"
# ── Build ───────────────────────────────────────────────────
echo "==> Building $TARGET..."
$COMPOSE --profile "$TARGET" build
# ── Backup DB before migration ────────────────────────────────
BACKUP_TAG="pre-deploy-$(date +%Y%m%d-%H%M%S)"
echo "==> Backing up database (${BACKUP_TAG})..."
$COMPOSE run --rm --entrypoint "" "${TARGET}-app" \
sh -c "cp /app/data/app.db /app/data/app.db.${BACKUP_TAG} 2>/dev/null || true"
# ── Migrate ─────────────────────────────────────────────────
echo "==> Running migrations..."
$COMPOSE --profile "$TARGET" run --rm "${TARGET}-app" \
python -m padelnomics.migrations.migrate
# ── Start & health check ───────────────────────────────────
# Router config is NOT written yet — router keeps old config so it stays
# healthy while we wait for the new slot to pass its own health check.
echo "==> Starting $TARGET (waiting for health check)..."
if ! $COMPOSE --profile "$TARGET" up -d --wait; then
echo "!!! Health check failed — dumping logs"
$COMPOSE logs --tail=100 2>&1 || true
echo "!!! Rolling back"
$COMPOSE stop "${TARGET}-app" "${TARGET}-worker" "${TARGET}-scheduler"
LATEST=$($COMPOSE run --rm --entrypoint "" "${TARGET}-app" \
sh -c "ls -t /app/data/app.db.pre-deploy-* 2>/dev/null | head -1")
if [ -n "$LATEST" ]; then
echo "==> Restoring database from ${LATEST}..."
$COMPOSE run --rm --entrypoint "" "${TARGET}-app" \
sh -c "cp '${LATEST}' /app/data/app.db"
fi
exit 1
fi
# ── Write router config and reload (new slot is healthy) ────
echo "==> Writing router config for $TARGET..."
mkdir -p "$(dirname "$ROUTER_CONF")"
cat > "$ROUTER_CONF" <<NGINX
upstream app {
server ${TARGET}-app:5000;
}
server {
listen 80;
location / {
proxy_pass http://app;
proxy_set_header Host \$host;
proxy_set_header X-Real-IP \$remote_addr;
proxy_set_header X-Forwarded-For \$proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto \$scheme;
}
}
NGINX
echo "==> Reloading router..."
$COMPOSE exec router nginx -s reload
# ── Cleanup old pre-deploy backups (keep last 3) ─────────────
$COMPOSE run --rm --entrypoint "" "${TARGET}-app" \
sh -c "ls -t /app/data/app.db.pre-deploy-* 2>/dev/null | tail -n +4 | xargs rm -f" || true
# ── Stop old slot ───────────────────────────────────────────
if [ "$CURRENT" != "none" ]; then
echo "==> Stopping $CURRENT..."
$COMPOSE stop "${CURRENT}-app" "${CURRENT}-worker" "${CURRENT}-scheduler"
fi
# ── Record live slot ────────────────────────────────────────
echo "$TARGET" > "$LIVE_FILE"
echo "==> Deployed $TARGET successfully!"