System and method for connection failover using redirection

• US 20060179147A1
• Filed: 02/07/2005
• Published: 08/10/2006
• Est. Priority Date: 02/07/2005
• Status: Active Grant

• Current Assignee: Veritas Technologies, LLC (Whitehouse Group Ltd.)
• Original Assignee: Veritas Operating Corporation (Gen Digital Inc.)
• Inventors: Tran, Cuong, Seetala, Partha, Thaker, Bhavin
• Granted Patent: US 7,668,962 B2
• Time in Patent Office: Days
• US Class Current: 709/227
• CPC Class Codes: G06F 11/2012 and using different communication protocols
• CPC Class Codes: G06F 11/2023 Failover techniques
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A system for connection failover using redirection includes a primary server comprising a primary connection failover driver (CFD), a shadow server comprising a shadow CFD, and a client. The primary and shadow servers and the client are coupled via a network. The primary server and the client are configured to cooperate to establish a network connection. The primary CFD is configured to redirect a first message packet, targeted for transmission to the client over the network connection, to the shadow server. The shadow CFD is configured to copy contents of the first message packet into a log, and forward the first message packet to the client after the contents have been copied.

Claims

1. A system comprising:

a primary server including a primary connection failover driver;

a shadow server including a shadow connection failover driver; and

a client;

wherein the primary server, the client and the shadow server are coupled via a network;

,

wherein the primary server and the client are configured to cooperate to establish a network connection;

wherein the primary connection failover driver is configured to redirect a first message packet, targeted for transmission over the network connection to the client, to the shadow server, and

wherein the shadow connection failover driver is configured to;

copy contents of the first message packet in a log; and

forward the first message packet to the client.

2. The system as recited in claim 1, wherein the network connection is established using the Transmission Control Protocol (TCP).

3. The system as recited in claim 1, wherein a networking software stack at the primary server includes the primary connection failover driver in a layer below a transport layer.

4. The system as recited in claim 1, wherein the primary connection failover driver is configured to redirect the first message packet to the shadow server according to a protocol that does not require an acknowledgment from the shadow server of a reception of the first message packet.

5. The system as recited in claim 3, wherein the protocol is the User Datagram Protocol (UDP).

6. The system as recited in claim 1, wherein, in response to a reception of a second message packet from the client, the primary connection failover driver is further configured to:

transmit a copy of the second message packet to the shadow server; and

redirect an acknowledgment packet, targeted for transmission over the network connection to the client, to the shadow server;

and wherein the shadow connection failover driver is further configured to;

copy contents of the second message packet in the log; and

forward the acknowledgment packet to the client.

7. The system as recited in claim 6, wherein the shadow connection failover driver is further configured to verify that the copy of the second message packet has been received at the shadow server, prior to forwarding the acknowledgment packet.

8. The system as recited in claim 1, wherein the network connection is established to allow communication between an application and the client, and wherein, in response to a detection of a failure, the shadow connection failover driver is configured to re-establish the network connection, using connection state information derived from the log, to allow communication between a failover version of the application and the client.

9. The system as recited in claim 1, wherein at least a portion of the log is maintained within a volatile memory.

10. The system as recited in claim 9, wherein at least a subset of contents of the log are asynchronously copied to a non-volatile storage device.

11. A method comprising:

establishing a network connection between a primary server and a client;

generating a first message packet targeted for transmission over the network connection from the primary server to the client;

redirecting the first message packet from the primary server to a shadow server;

copying contents of the first message packet in a log accessible from the shadow server; and

forwarding the first message packet to the client from the shadow server.

12. The method as recited in claim 11, wherein the primary server includes a networking software stack comprising a primary connection failover driver in a layer below a transport layer, and wherein the primary connection failover driver is configured to redirect the first message packet.

13. The method as recited in claim 11, wherein the first message packet is redirected to the shadow server according to a protocol that does not require an acknowledgment from the shadow server of a reception of the first message packet.

14. The method as recited in claim 11, further comprising:

receiving a second message packet from the client at the primary server;

transmitting a copy of the second message packet to the shadow server from the primary server;

generating an acknowledgment packet targeted at the client, responsive to the second message packet, at the primary server;

redirecting the acknowledgment packet to the shadow server from the primary server;

copying contents of the second message packet in the log accessible from the shadow server; and

forwarding the acknowledgment packet to the client from the shadow server.

15. The method as recited in claim 14, further comprising:

verifying that the copy of the second network packet has been received at the shadow server, prior to forwarding the acknowledgment packet.

16. The method as recited in claim 11, wherein the network connection is established to allow communication between an application and the client, further comprising:

in response to a detection of a failure, re-establishing the network connection at the shadow server using connection state information derived from the log, to allow communication between a failover version of the application and the client.

17. A computer readable medium comprising program instructions, wherein the instructions are executable to:

establish a network connection between a primary server and a client;

generate a first message packet targeted for transmission over the network connection from the primary server to the client;

redirect the first message packet from the primary server to a shadow server;

copy contents of the first message packet in a log accessible from the shadow server; and

forward the first message packet to the client from the shadow server.

18. The computer readable medium as recited in claim 17, wherein the primary server includes a networking software stack comprising a primary connection failover driver in a layer below a transport layer, and wherein the primary connection failover driver is configured to redirect the first message packet.

19. The computer readable medium as recited in claim 17, wherein the first message packet is redirected to the shadow server according to a protocol that does not require an acknowledgment from the shadow server of a reception of the first message packet.

20. The computer readable medium as recited in claim 17, wherein the instructions are further executable to:

receive a second message packet from the client at the primary server;

transmit a copy of the second message packet to the shadow server from the primary server;

generate an acknowledgment packet targeted at the client, responsive to the second message packet, at the primary server;

redirect the acknowledgment packet to the shadow server from the primary server;

copy contents of the second message packet in the log accessible from the shadow server; and

forward the acknowledgment packet to the client from the shadow server.

21. The computer readable medium as recited in claim 20, wherein the instructions are further executable to:

verify that the copy of the second network packet has been received at the shadow server, prior to forwarding the acknowledgment packet.

22. The computer readable medium as recited in claim 17, wherein the network connection is established to allow communication between an application and the client, and wherein the instructions are further executable to:

in response to a detection of a failure, re-establish the network connection at the shadow server using connection state information derived from the log, to allow communication between a failover version of the application and the client.