FAQ

Do I need a connector for every tool?

No. You only need the Edge Connector for tools that must run close to your systems (for example on-prem databases or private APIs). Pure SaaS tools or cloud-only workflows can run entirely in nara’s cloud without an edge deployment.

Can one connector handle multiple tools?

Yes. A single connector instance can host multiple tools as long as they are part of the same bundle and belong to the same organization. Use edge-connector generate for each tool, then package and upload them together.

How do I separate staging and production?

Use separate:

organizations or environments in nara,
deployments (for example edge-staging, edge-prod),
bundles (for example edge-bundle-org-123-staging, edge-bundle-org-123-prod).

Point staging connectors at a staging Tool RPC URL and staging bundle; do the same for production. Avoid sharing tokens or deployments across environments.

Can I run the connector in containers?

Yes. For containerized deployments:

mount persistent volumes for config.json, auth-token.json, and deployment-credentials.json,
pass tokens and deployment IDs via environment variables for initial provisioning,
use --auto-resume or your own process manager (systemd, Kubernetes, etc.) to keep the connector running.