Distributed database exception handling

A computer-implemented method, comprising:

• receiving a first database request from a client driver;
• generating a second database request based, at least in part, on translating the first database request;
• sending the second database request to a storage subsystem of a distributed database;
• and
• in response to receiving an error message from the storage subsystem in response to the second database request:
• determining whether the second database request can be retried based, at least in part, on the error message;
• resending the second database request to the storage subsystem on behalf of the client driver in response to determining that the second database request can be retried;
• receiving a first response in a first format from the storage subsystem in response to the resent second database request;
• translating the error message into a second format expected by an account associated with the client driver, the second format different than the first format, in response to determining that the second database request cannot be retried;
• and
• sending a second response, in the second format, to the client driver based, at least in part, on the translated error message or translating the first response from the first format to the second format.

A non-transitory computer-readable storage medium having stored thereon executable instructions that, in response to being executed by one or more processors of a computing device, cause the computing device to at least:

• receive a first database request from a client driver;
• generate a second database request based, at least in part, on translating the first database request;
• send the second database request to a storage subsystem of a distributed database;
• and
• in response to receiving an error message generated by the storage subsystem in response to the second database request:
• determine whether the second database request can be retried based, at least in part, on the error message;
• resend the second database request on behalf of the client driver in response to determining that the second database request can be retried;
• receive a first response in a first format from the storage subsystem in response to the resent second database request;
• translate the error message into a second format expected by a database client associated with the client driver, the second format different than the first format, in response to determining that the second database request cannot be retried;
• and
• send a second response, in the second format, to the client driver based, at least in part, on the translated error message or translating the first response from the first format to the second format.