Provisioning

Provisioning, shown on the pages as preparing the work environment, builds the isolated environment your tasks run in, and it starts automatically the moment you connect a repository. You do not trigger it and you cannot speed it up. It usually takes 10–20 seconds, and the page refreshes itself while it runs, so you can leave it open.

​

What happens during provisioning

Orbi creates an environment dedicated to your repository, clones the repository into it, installs the delivery engine, and runs the engine’s setup step against your repository. That setup is what creates the ai-* labels (ai-ready, ai-in-progress, ai-pr-opened, ai-fix-needed, ai-merged, ai-blocked and the rest) in your repository’s label list. Setup is declarative: a missing label is created, a drifted one is corrected, and nothing is ever deleted. Your own labels (bug, enhancement, whatever you use) are never touched. Each environment belongs to exactly one tenant and one repository. Environments are isolated from one another at the operating-system level.

​

Watching it

While provisioning runs, the status page headline says Orbi is preparing the work environment, and in the setup checklist the Prepare work environment step shows Orbi is working. The card below lists four steps: Set up your own work environment, Pull your code, Install Orbi and Start reading your repository. Each one gets a ✓ and the seconds it took, and the line under the list says Usually 10–20 seconds. After 30 seconds it changes to Taking longer than usual; still preparing…. When all four are done, the Prepare work environment step turns to ✓ Done. The same state is listed per repository on the status page under Usage and settings → Connected projects, in the Preparation status column:

​

When the AI model service is unavailable

If Orbi’s AI model service is temporarily unavailable and you have not set your own API key, the repository still connects, but provisioning does not start. The page after connecting is titled Repository connected, waiting for the AI model service. Click Continue preparation later; this does not add a second connection. Or click Use your own API key to start right away with your own AI provider, which bills you directly.

​

When provisioning fails

A failure is never a dead end. The main status card says Work environment preparation failed: and tells you which of three situations you are in; View failure reasons and repair steps jumps to that repository’s cell in the Connected projects table, with the reason and the next step. In the project overview on the status page, the repository’s row carries the same distinction as a chip: Needs you, Retrying automatically or On Orbi’s side. The causes behind those headlines fall into four classes.

​

The GitHub App installation or authorization is no longer valid

The reason reads Orbi’s GitHub App installation or authorization is no longer valid. Orbi can no longer mint a token for your installation, usually because the App was uninstalled or its access to the repository was removed. What to do: follow Reauthorize the GitHub App in the next step, reinstall or re-authorize the App, then come back to the status page.

​

The API key was not accepted

The reason reads The API key was rejected. Your own API key was rejected by the provider. What to do: follow Set up AI model in the next step, or open Settings → AI model from the account menu, and correct the key. See model configuration. Saving retries provisioning automatically; you do not need to reconnect the repository.

​

The AI model service could not be reached

The reason reads Unable to reach the AI model service (the API key or baseUrl may be wrong). The key or the baseUrl may be wrong, or the provider is unreachable. What to do: check the baseUrl and key in Settings → AI model, and use Test connection before saving. As above, saving retries provisioning automatically.

​

A platform-side failure

Anything not attributable to your installation or your AI model settings is ours, for example Work environment preparation failed (on Orbi’s side) or Work environment configuration sync failed. What to do: nothing. The next step says This is on Orbi’s side; no action needed. or it will retry automatically; no action is needed. If it has not recovered after a while, contact support. The failure details include an Error ID that identifies your exact provisioning attempt, so quote it.

​

Retries

Provisioning retries automatically a limited number of times. While it still will, the next step says it will retry automatically; no action is needed. When the cause is on your side, fixing the cause retries it: saving corrected AI model settings retries even after the automatic retries are used up. When the cause is on the platform side and the retries are used up, we take it from there without any action from you.

​

After provisioning succeeds

The environment stays in place for later tasks. Orbi then reads the repository and suggests work it can do, usually in about 2 minutes. This is separate from preparing the environment. You can leave and return to review a suggested draft, or write your own request. Connecting another repository provisions a fresh environment for it, and disconnecting a repository reclaims its environment.