No error.

A database/SQL-level error occurred while processing the request.

The request data failed validation because a value was invalid.

The requested object could not be found.

The requested object exists but has already been deleted.

An unexpected internal error occurred while processing the request.

One or more mandatory fields were missing from the request.

The organization/implementation id could not be resolved for the current request.

Model validation (data annotations) failed for the submitted entity.

The entity could not be created.

The entity could not be updated.

The entity could not be saved (create or update).

The entity could not be deleted.

The operation could not be performed because the job has been canceled.

The operation could not be performed because the job has already been completed.

The referenced job could not be found.

The serialized stock item has already been disposed and cannot be used.

The serialized stock item's current status does not allow this operation.

Updating the dispatch credit usage failed while processing a job dispatch log entry.

Accepting the job dispatch failed.

The job has already been accepted by another technician.

The job dispatch has already been rejected.

The user has already rejected this job dispatch.

The referenced serialized stock item is not available for use on this job part usage.

The referenced serialized stock item could not be found for this job part usage.

There is not enough stock available to cover the requested job part usage quantity.

The one-time password has expired.

The one-time password could not be found.

The one-time password does not match.

The requested domain could not be found.

The requested domain has been archived.

The requested domain is currently undergoing maintenance and cannot be reached.

The API could not get into the database because of how its Managed Identity is configured, not because the request was wrong. Distinct from SqlError so a missing grant is recognisable as something to fix in Azure.

The user's credentials did not resolve to a valid account. See FSA-6049.

The user has been archived/disabled. See FSA-6049.

The user has no user group assigned. See FSA-6049.

The user (and its user group) is not allowed to use the service app. See FSA-6049.

The user's licence has expired. See FSA-6049.

The user access check itself could not be completed (e.g. an OCC/database lookup failed), as distinct from a genuine "not allowed" outcome. See FSA-6049.

The AI service backing a text rewrite request could not be reached, or did not return a usable response, after retries. See FSA-6091.

The AI service was reached and answered normally, but declined to rewrite the text it was given. The request was valid and the service is healthy, so this is neither a DataInvalid nor an AiServiceUnavailable case: retrying the same text will not help, and the client should keep the technician's original text rather than display the refusal. See FSA-6154.

Too many AI actions in the last minute (HTTP 429). The Retry-After header tells when to try again. See FSA-6217.

The user reached the AI usage limit of the day or the month (HTTP 429). The Retry-After header and the error detail tell when it resets. See FSA-6217.

The domain reached its AI usage limit of the day or the month (HTTP 429). The Retry-After header and the error detail tell when it resets. See FSA-6217.

The AI features are not turned on for this domain (HTTP 403): it has no enabled route in OCC for this app and feature, or the key or the provider of its route is switched off. The AI is not called and retrying will not help. See FSA-6215.

A work order update asked to change the status to Cancelled (HTTP 400). Cancelling must go through the dedicated cancel action, which also stops the work order's used parts from counting against stock; the generic update does not. Nothing was changed. See FSA-6237.