docs(pulse-docs): recovery commands + DOCKER-CHECKLIST + SESSION-CHECKLIST — docs de auto-administracao

runbook/RECOVERY-COMMANDS.md

@@ -1,20 +1,43 @@
-# Recovery Commands — Docker Swarm
+# Comandos de Recuperação — Docker Swarm
 
-## Emergency
-\`\`\`bash
-docker node ls                              # verificar saude do no
-docker stack rm <stack> && sleep 3           # remover stack problematica
+_Alfabeto de comandos para o Pulse usar quando algo quebrar._
+
+## Emergency — todos os serviços down
+
+```bash
+docker node ls                              # verificar saúde do nó
+docker stack rm <stack> && sleep 3           # remover stack problemática
+docker swarm init                           # só se necessário
 docker stack deploy -c <stack>.yml <stack>  # re-deploy
-\`\`\`
+```
 
-## Servico especifico
-\`\`\`bash
-docker service ps <stack>_<service>                # ver tasks
-docker service update --force <stack>_<service>    # forc@r nova task
-\`\`\`
+## Serviço específico — forçar restart
 
-## Health check manual
-\`\`\`bash
+```bash
+docker service ps <stack>_<service>                # ver tasks atuais
+docker service update --force <stack>_<service>    # forçar nova task
+```
+
+## Limpar containers órfãos
+
+```bash
+docker ps -a -f "status=exited" --format '{{.Names}}' | xargs -r docker rm -f
+docker ps -a -f "status=dead"   --format '{{.Names}}' | xargs -r docker rm -f
+```
+
+## Swarm reset (extreme)
+
+```bash
+docker swarm leave --force && docker swarm init --advertise-addr <ip>
+```
+
+## Health check manual de um container
+
+```bash
+# Status geral
+docker inspect --format '{{json .State.Health}}' <container_id> | python3 -m json.tool
+
+# Com health check definido
 docker inspect --format '{{.State.Health.Status}}' <container_id>
-# → healthy | unhealthy | starting | <sem-health>
-\`\`\`
+# → "healthy" | "unhealthy" | "starting"
+```