Seu Nome 40e2c01abd fix: zero-downtime deploy com start-first e healthcheck ... - Remove docker service update --force que causava downtime - Agrupa env vars do Convex em um único update (evita múltiplos restarts) - Adiciona delay: 10s e monitor: 30s no update_config - Healthcheck do web usa /api/health com timeout - Ajusta start_period: 180s (web) e 60s (convex) - Convex backend não é mais forçado a reiniciar após stack deploy Fluxo correto de deploy: 1. docker stack deploy detecta mudança 2. Novo container é criado (start-first) 3. Swarm espera healthcheck passar 4. Swarm espera monitor period (30s) 5. Container antigo é removido 6. Zero downtime durante todo o processo 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>		2025-12-08 15:07:13 -03:00
..
ci-cd-web-desktop.yml	fix: zero-downtime deploy com start-first e healthcheck	2025-12-08 15:07:13 -03:00
desktop-release.yml	Switch workflows to Bun install/test and update pnpm	2025-11-04 23:21:41 -03:00
quality-checks.yml	Hardening CI for prisma and convex deploy	2025-11-18 09:43:28 -03:00