Support
FAQ
Short answers for common CloudWatcher setup and feature questions.
Frequently asked questions
No. AWS has the deepest coverage, but Azure and GCP are supported in provider-aware setup, dashboards, billing, alerts, simulations, and live infrastructure where the capability is ready.
Read permissions and write permissions are different. A provider can allow inventory discovery while blocking resource creation, update, or deletion.
Cloud Shell cannot reach your laptop through localhost. Use a hosted backend callback URL or paste service-account JSON manually.
No. Core telemetry can work with metadata such as provider, model, tokens, latency, cost, status, service, endpoint, and spans. Prompt capture can stay optional.
The model may not have pricing configured yet. Tokens can be recorded while cost remains unavailable or zero.
Editing a canvas and previewing Terraform are planning steps. Deploying a simulation can create real resources.
No. Delete removes the saved design. Use Destroy in Simulation History to clean up deployed resources.
It syncs existing cloud resources into visual service canvases. Supported actions affect real resources.
It is a guided job workflow for moving a server to a better size with preflight, launch, validation, cutover, and source-preservation steps.
Yes. It works as a focused AI telemetry tool, though it is most useful when cloud cost and reliability data are also connected.
Need more help?
Keep moving with the right next guide
If your team still runs into setup issues, empty dashboards, billing delays, callback failures, or alerting problems, continue with troubleshooting before re-running the entire onboarding flow.