Error recovery in a client/server application using two independent sockets for communication

A first host, comprising:

• a first socket and a second socket;
• wherein the first host's first socket is used for sending requests to a second host's first socket, and receiving responses from the second host's first socket, and the first host's second socket is used for receiving requests from the second host's second socket, and sending responses to the second host's second socket;
• a memory for storing software instructions;
• and
• a control associated with the memory for executing the software instructions for:
• detecting when a failure condition occurs at the first host's second socket, and, when the failure condition is detected:
• attempting to send a communication from the first host's first socket to the second host's first socket;
• if the attempt to send succeeds, closing the first host's second socket, then attempting to reconnect the first host's second socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A program storage device, tangibly embodying a program of instructions executable by a first host to perform a method for detecting a failure, the first host having a first socket and a second socket, wherein the first host's first socket is used for sending requests to a second host's first socket, and receiving responses from the second host's first socket, and the first host's second socket is used for receiving requests from the second host's second socket, and sending responses to the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's second socket, and, when the failure condition is detected:
• attempting to send a communication from the first host's first socket to the second host's first socket;
• if the attempt to send succeeds, closing the first host's second socket, then attempting to reconnect the first host's second socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A method for detecting a failure at a first host having a first socket and a second socket, wherein the first host's first socket is used for sending requests to a second host's first socket, and receiving responses from the second host's first socket, and the first host's second socket is used for receiving requests from the second host's second socket, and sending responses to the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's first socket, and, when the failure condition is detected:
• checking an internal state to determine whether the first host is attempting to recover from a failure condition detected at the first host's second socket;
• if the internal state indicates the first host is attempting to recover, allowing the first host to attempt to recover;
• if the internal state indicates the first host is not attempting to recover, attempting to reconnect the first host's first socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A first host, comprising:

• a first socket and a second socket;
• wherein the first host's first socket is used for sending requests to a second host's first socket, and receiving responses from the second host's first socket, and the first host's second socket is used for receiving requests from the second host's second socket, and sending responses to the second host's second socket;
• a memory for storing software instructions;
• and
• a control associated with the memory for executing the software instructions for:
• detecting when a failure condition occurs at the first host's first socket, and, when the failure condition is detected:
• checking an internal state to determine whether the first host is attempting to recover from a failure condition detected at the first host's second socket;
• if the internal state indicates the first host is attempting to recover, allowing the first host to attempt to recover;
• if the internal state indicates the first host is not attempting to recover, attempting to reconnect the first host's first socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A program storage device, tangibly embodying a program of instructions executable by a first host to perform a method for detecting a failure, the first host having a first socket and a second socket, wherein the first host's first socket is used for sending requests to a second host's first socket, and receiving responses from the second host's first socket, and the first host's second socket is used for receiving requests from the second host's second socket, and sending responses to the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's first socket, and, when the failure condition is detected:
• checking an internal state to determine whether the first host is attempting to recover from a failure condition detected at the first host's second socket;
• if the internal state indicates the first host is attempting to recover, allowing the first host to attempt to recover;
• if the internal state indicates the first host is not attempting to recover, attempting to reconnect the first host's first socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A method for detecting a failure at a first host, the first host having a first socket and a second socket, wherein the first host's first socket is used for receiving requests from a second host's first socket, and sending responses to the second host's first socket, and the first host's second socket is used for sending requests to the second host's second socket, and receiving responses from the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's first socket, and, when the failure condition is detected:
• checking an internal state to determine whether the first host is attempting to recover from a failure condition detected at the first host's second socket;
• if the internal state indicates the first host is attempting to recover, allowing the first host to attempt to recover;
• if the internal state indicates the first host is not attempting to recover, closing the first host's first socket and waiting for the second host to reconnect the first host's first socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A first host, comprising:

• a first socket and a second socket;
• wherein the first host's first socket is used for receiving requests from a second host's first socket, and sending responses to the second host's first socket, and the first host's second socket is used for sending requests to the second host's second socket, and receiving responses from the second host's second socket;
• a memory for storing software instructions;
• and
• a control associated with the memory for executing the software instructions for:
• detecting when a failure condition occurs at the first host's first socket, and, when the failure condition is detected:
• checking an internal state to determine whether the first host is attempting to recover from a failure condition detected at the first host's second socket;
• if the internal state indicates the first host is attempting to recover, allowing the first host to attempt to recover;
• if the internal state indicates the first host is not attempting to recover, closing the first host's first socket and waiting for the second host to reconnect the first host's first socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A program storage device, tangibly embodying a program of instructions executable by a first host to perform a method for detecting a failure, the first host having a first socket and a second socket, wherein the first host's first socket is used for receiving requests from a second host's first socket, and sending responses to the second host's first socket, and the first host's second socket is used for sending requests to the second host's second socket, and receiving responses from the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's first socket, and, when the failure condition is detected:
• checking an internal state to determine whether the first host is attempting to recover from a failure condition detected at the first host's second socket;
• if the internal state indicates the first host is attempting to recover, allowing the first host to attempt to recover;
• if the internal state indicates the first host is not attempting to recover, closing the first host's first socket and waiting for the second host to reconnect the first host's first socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A method for detecting a failure at a first host, the first host having a first socket and a second socket, wherein the first host's first socket is used for receiving requests from a second host's first socket, and sending responses to the second host's first socket, and the first host's second socket is used for sending requests to the second host's second socket, and receiving responses from the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's second socket, and, when the failure condition is detected:
• attempting to send a communication from the first host's first socket to the second host's first socket;
• if the attempt to send succeeds, closing the first host's second socket and waiting for the second host to reconnect the first host's second socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A first host, comprising:

• a first socket and a second socket;
• wherein the first host's first socket is used for receiving requests from a second host's first socket, and sending responses to the second host's first socket, and the first host's second socket is used for sending requests to the second host's second socket, and receiving responses from the second host's second socket;
• a memory for storing software instructions;
• and
• a control associated with the memory for executing the software instructions for:
• detecting when a failure condition occurs at the first host's second socket, and, when the failure condition is detected:
• attempting to send a communication from the first host's first socket to the second host's first socket;
• if the attempt to send succeeds, closing the first host's second socket and waiting for the second host to reconnect the first host's second socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.

A program storage device, tangibly embodying a program of instructions executable by a first host to perform a method for detecting a failure, the first host having a first socket and a second socket, wherein the first host's first socket is used for receiving requests from a second host's first socket, and sending responses to the second host's first socket, and the first host's second socket is used for sending requests to the second host's second socket, and receiving responses from the second host's second socket, the method comprising:

• detecting when a failure condition occurs at the first host's second socket, and, when the failure condition is detected:
• attempting to send a communication from the first host's first socket to the second host's first socket;
• if the attempt to send succeeds, closing the first host's second socket and waiting for the second host to reconnect the first host's second socket;
• and
• if the attempt to reconnect succeeds, setting an internal state to indicate that normal operation is resumed.