archestra.llm.complete() call is failing 100% of the time with error code llm_unavailable ("The LLM completion could not be produced") β it's not a quota issue (llm_quota), and it happens on every request.llm.complete calls? If it needs enabling, could you turn it on (or tell me who can)?Read-only live mirror of Archestra.AI Slack
πJoin the discussion withAI enthusiasts!assistant:writecommandsapp_mentions:readfiles:readgroups:historyim:writereactions:readusers:read.emailbash docker pull archestra/platform:latest docker run -p 127.0.0.1:9000:9000 -p 127.0.0.1:3000:3000 \ -e ARCHESTRA_QUICKSTART=true \ -v /var/run/docker.sock:/var/run/docker.sock \ -v archestra-postgres-data:/var/lib/postgresql/data \ -v archestra-app-data:/app/data \ archestra/platform `` Then open: - Admin UI β http://localhost:3000 - API β http://localhost:9000 The docker.sock mount enables the embedded Kubernetes (KinD) cluster for running MCP servers. This mode is great for tinkering but not for production β the bundled Postgres isn't persisted the same way a managed DB would be. :wheelofdharma: Helm (recommended for production) Requires a Kubernetes cluster, Helm 3+, and kubectl: ``bash helm upgrade archestra-platform \ <oci://europe-west1-docker.pkg.dev/friendly-path-465518-r6/archestra-public/helm-charts/archestra-platform> \ --install \ --namespace archestra \ --set archestra.env.HOSTNAME="0.0.0.0" \ --create-namespace \ --wait `` For production, a few key recommendations: - Use an external PostgreSQL (--set postgresql.enabled=false --set postgresql.external_database_url=...) β cloud-managed DBs give you HA, backups, and encryption. - Set proper load-balancer timeouts* (600s) so streaming responses don't get cut off β the docs have per-cloud snippets for GKE, EKS, AKS, nginx, and Traefik. - Set ARCHESTRA_AUTH_SECRET, ARCHESTRA_FRONTEND_URL, and change the default admin password. Both methods expose the Admin UI on port 3000 and the API on port 9000. :pagefacingup: Full guide (env vars, IaC via Terraform/Crossplane, scaling, etc.): https://github.com/archestra-ai/archestra/blob/main/docs/pages/platform-deployment.md Want me to dig into any particular part β e.g. a specific cloud, external Postgres setup, or the Helm config options? π€ Community Agent