2.9 Error Handling

There are several potential failure scenarios for the Active Directory system. "Failure", in this context, does not refer to an error returned by a member protocol due to an invalid or not permitted request (for example, a request to modify a directory object when the requestor does not have the necessary permissions to do so). Such errors are part of the normal processing behavior of the system. Instead, "failure" in this context refers to conditions that can prevent the system from successfully servicing requests made by a client that uses the member protocols.

These failure scenarios are the following:

  • Transient unavailability of durable storage, without loss or corruption of data

  • Permanent unavailability of durable storage

  • Corruption of data on the durable storage

  • Unavailability of networking

  • Unavailability of DNS

  • Failures while joining or unjoining a domain

Additionally, individual member protocols can have their own failure scenarios. Such scenarios are documented in the protocols' Technical Documents and are not repeated here.

The Active Directory system does not define any error handling requirements beyond those that are described in the Technical Documents of the protocols that the system supports, as listed in section 2.4, and in the failure scenarios described in the sections that follow.

Various kinds of errors might occur that affect the system. More precisely, an error condition might affect one or more protocols supported by the system. Such error conditions and the resulting protocol semantics are described in the corresponding protocol Technical Documents. The system does not constrain the types of errors that can be received through the member protocols.