Bi-directional HTTP-based reliable messaging protocol and system utilizing same

• US 7,562,147 B1
• Filed: 10/02/2000
• Issued: 07/14/2009
• Est. Priority Date: 10/02/2000
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Cohen, Shy
• Primary Examiner(s): Caldwell; Andrew
• Assistant Examiner(s): FORD, GRANT M
• Application Number:
  US09/676,924
• Time in Patent Office: 3,207 Days
• Field of Search: 709/218, 709227-229, 709/234, 709/203, 709/206, 709/223, 709/246, 709/226, 709/224, 707/10, 713/201, 370/230, 370/310, 370/312, 370/395.21, 370/238, 370/221.01, 370/401, 370/462, 370/442
• US Class Current: 709/228
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]

Abstract

A system and protocol for enabling bi-directional HTTP-based reliable messaging through a Web proxy server via the Internet is presented. The client processor residing on the private computer network side of the Web proxy establishes two virtual channels with a server. One of the channels is used for outgoing messages from the client to the server, and message delivery acknowledgments. The other channel is used for incoming messages from the server to the client, and message delivery acknowledgments. This incoming channel is established and maintained by the client through the Web proxy server so that the client has full control over the channel. The establishment is accomplished by the transmission of an HTTP-based “request” to the server. The server does not respond until and unless it has a message to be transmitted to the client. Once such a message is generated, the server generates an HTTP-based “reply” to this request, and transmits it to the client via the Web proxy server. The client then sends another HTTP-based request to the server with the message acknowledgment included therein. The client may request that the server reply to the request after some period of time to allow the client to assess the health of the connection and avoid connection time out disconnections by the Web proxy server. This period may be fixed or may be dynamically adjusted.

Claims

1. A method of bi-directionally communicating between an application residing on a first processor on a private computer network and an application residing on a second processor not on the private computer network in order to allow for the delivery of unsolicited messages to be received from the second processor, the communication path including a public computer network and a proxy server coupled to the private computer network and separating the private computer network from the public computer network, the method comprising:

transmitting a first HTTP-based “

request”

from the first processor to the second processor for establishing a first communication channel between the first processor and the second processor through the proxy server to allow the transfer of first messages from the first processor to the second processor, and the delivery of first message delivery acknowledgments from the second processor to the first processor;

transmitting a first parked HTTP-based “

request”

from the first processor to be parked at the second processor for establishing a persistent communication channel between the first processor and the second processor through the proxy server to allow the transfer of second messages from the second processor to the first processor, and the delivery of second message delivery acknowledgments from the first processor to the second processor, and wherein the first parked HTTP-based “

request”

includes therein a request that the second processor transmit a reply after the expiration of a time period even if there are no messages to send to the first processor so that the first processor can assess a status of the connection thereto;

receiving a first HTTP-based “

reply”

from the second processor to the first processor in response to the first parked HTTP-based “

request”

; and

in response to receiving the first HTTP-based “

reply”

, transmitting a second parked HTTP-based “

request”

via the proxy to the second processor, the second parked HTTP “

request”

including an acknowledgment to the first HTTP-based “

reply”

in order to maintain the persistent HTTP-based connection between the first processor and the second processor through the proxy server, and wherein the second parked HTTP-based “

request”

includes therein a request that the second processor transmit a reply after the expiration of a time period even if there are no messages to send to the first processor in order to ensure persistent connectivity between the first and second processor.

2. The method of claim 1, wherein the first HTTP-based “

request”

includes at least one of the first messages therein.

3. The method of claim 1, wherein the first HTTP-based “

reply”

includes at least one of the second messages therein.

4. The method of claim 1, wherein the first processor only receives the first HTTP-based “

reply”

from the second processor on the persistent communication channel when the second processor has at least one of the second messages to send to the first processor.

5. The method of claim 1, further comprising setting the time period to be less than two days.

6. The method of claim 1, further comprising setting the time period to be approximately five minutes.

7. The method of claim 1, further comprising dynamically adjusting the time period based upon a connection time out closure controlled by the proxy server.

8. A computer-readable storage medium having computer-readable instructions that, when executed by a processor of a computing device, cause the computing device to perform the method of claim 1.

9. The method of claim 7, wherein the dynamically adjusting of the time period comprises:

receiving a connection time out closure message from the proxy server;

determining a first time between transmitting the second HTTP-based “

request” and

receiving a connection time out closure message from the proxy server; and

calculating a new time period to be less than the first time and less than the time period.

10. A method of enabling transmission of unsolicited messages from a server to a client by ensuring that a persistent connection between the server and the client does not time out, wherein the client resides on a private computer network having a proxy server between the private computer network and a public computer network, and wherein the server transmits the unsolicited messages over the public computer network, the method comprising:

selecting by a client a connection time out period used in order to determine a time duration in which the client is to receive a “

reply”

message from a server in order to ensure persistent connectivity between the client and the server;

allowing the client to include the connection time out period in a parked HTTP-based “

request”

sent from the client to be parked at the server for requesting an HTTP-based “

reply”

from the server after the expiration of the connection time out period even if there are no messages to send to the client in order to avoid connection termination by the proxy server due to communication inactivity; and

transmitting the parked HTTP-based “

request”

to the server via the proxy server to open a persistent connection therewith.

11. The method of claim 10, further comprising dynamically adjusting the connection time out period to avoid connection termination by the proxy server due to communication inactivity.

12. The method of claim 10, further comprising:

receiving a connection time out closure message from the proxy server indicating that the proxy server has closed the persistent connection;

calculating a connection time out period from the transmitting of the parked HTTP-based “

request”

to the receiving of the connection time out closure message; and

transmitting a second parked HTTP-based “

request”

to the server via the proxy server to open a persistent connection therewith, the second parked HTTP-based “

request”

requesting a reply from the server when the server has messages to send to the client and after the expiration of the connection time out period if there are no messages to send to the client.

13. A computer-readable storage medium having computer-readable instructions that when executed by a processor of a computing device, cause the computing device to perform the method of claim 10.

14. The method of claim 11, wherein the step of dynamically adjusting the connection time out period comprises:

receiving a connection time out closure message from the proxy server;

upon receiving the time out closure message from the proxy server, calculating a new time period from the transmitting of the HTTP-based “

request”

to the receiving of the connection time out closure message;

reducing the connection time out period to be less than the new time period and less than a current value of the connection time out period in order to create a new connection time out period;

including the new connection time out period in a second parked HTTP-based “

request”

for requesting an HTTP-based “

reply”

from the server after the expiration of the new connection time out period even if there are no messages to send to the client in order to avoid connection termination by the proxy server due to communication inactivity; and

transmitting the second parked HTTP-based “

request”

to the server via the proxy server to maintain the persistent connection therewith.

15. The method of claim 12, further comprising:

receiving a second connection time out closure message from the proxy server;

reducing the connection time out period to form a new connection time out period shorter in duration than the connection time out period; and

transmitting a third parked HTTP-based “

request”

to the server via the proxy server to open a persistent connection therewith, the third parked HTTP-based “

request”

requesting a reply from the server when the server has messages to send to the client and after the expiration of the new connection time out period if there are no messages to send to the client.

16. A method of transmitting unsolicited HTTP-based messages via a public computer network to a client residing on a private computer network, the private computer network including a proxy server, the method comprising:

receiving an HTTP-based “

request”

originating from the client through the proxy server, wherein the HTTP-based “

request”

includes a first connection time out period used in order to determine a time duration in which the client is to receive a “

reply”

message in order to ensure persistent connectivity between the client and a server; and

parking the HTTP-based “

request”

without responding thereto unless a message is generated that needs to be transmitted to the client or unless the first connection time out period expires, the parking of the HTTP-based “

request”

establishing a persistent connection from the client through the proxy server; and

when the message is generated or the first connection time out period expires, generating an HTTP-based “

reply”

to the HTTP-based “

request”

parked for the client, the HTTP-based “

reply”

containing the message therein;

transmitting the HTTP-based “

reply”

;

receiving a second HTTP-based “

request”

including a message acknowledgment from the client through the proxy server acknowledging the receipt of the HTTP-based “

reply” and

also including a second connection time out period; and

parking the second HTTP-based request without responding thereto unless a message is generated that needs to be transmitted to the client or unless the second connection time out period expires, the parking the second HTTP-based request maintaining the persistent connection from the client through the proxy server.

17. The method of claim 16, wherein the second connection time out period is different than the first connection time out period.

18. A computer-readable storage medium having computer-executable instructions that when executed by a processor of a computing device, cause the computing device to perform the method of claim 16.