Seamless fail-over support for virtual interface architecture (VIA) or the like

• US 20030033412A1
• Filed: 08/08/2001
• Published: 02/13/2003
• Est. Priority Date: 08/08/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Sundaresan, Sharad, Sriram, Balasubramanian
• Granted Patent: US 7,870,258 B2
• Time in Patent Office: Days
• US Class Current: 709/227
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer
• 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
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

To connect a client application to a server ‘server’ on a cluster ‘cluster’ having a plurality of servers instantiated thereon, ‘cluster’ and ‘server’ are received from the client application, a first request message is sent to ‘cluster’ requesting first connection information for connecting to ‘server’, a first reply message containing the requested first connection information is received from ‘cluster’, and the client application is connected to ‘server’ on ‘cluster’ based on the received first connection information. Thereafter, the connection to ‘server’ fails, and a second request message is sent to ‘cluster’ requesting second connection information for connecting to ‘server’, a second reply message containing the requested second connection information is received from ‘cluster’, and the client application is again connected to ‘server’ on ‘cluster’ based on the received second connection information.

Claims

1. A method of connecting a client application to a server ‘

server’

on a cluster ‘

cluster’

having a plurality of servers instantiated thereon, the method comprising;

receiving ‘

cluster’ and

‘

server’

from the client application;

sending a first request message to ‘

cluster’

requesting first connection information for connecting to ‘

server’

;

receiving from ‘

cluster’

a first reply message containing the requested first connection information;

connecting the client application to ‘

server’

on ‘

cluster’

based on the received first connection information, wherein once connected, the client application and ‘

server’

may transact business;

determining that the connection to ‘

server’

has failed, ‘

server’

presumably having been moved from a first server of the cluster to a second server of the cluster, the received first connection information corresponding to the first server;

sending a second request message to ‘

cluster’

requesting second connection information for connecting to ‘

server’

, the requested second connection information corresponding to the second server;

receiving from ‘

cluster’

a second reply message containing the requested second connection information; and

connecting the client application to ‘

server’

on ‘

cluster’

based on the received second connection information, wherein once again connected, the client application and ‘

server’

may again transact business.

2. The method of claim 1 comprising sending each request message as a UDP (user datagram protocol) packet addressed to a UDP address of ‘

cluster’

.

3. The method of claim 1 comprising receiving each requested connection information including an address of ‘

cluster’

on which ‘

server’

is listening.

4. The method of claim 1 comprising receiving each requested connection information including an address of ‘

cluster’

on which ‘

server’

is listening for packets formatted according to a VIA (Virtual Interface Architecture) protocol.

5. The method of claim 1 comprising receiving each requested connection information including each address of ‘

cluster’

on which ‘

server’

is listening and a corresponding protocol associated with the address by ‘

server’

.

6. The method of claim 1 comprising connecting the client application to ‘

server’

on ‘

cluster’

at an address of ‘

cluster’

on which ‘

server’

is listening for packets formatted according to a VIA (Virtual Interface Architecture) protocol.

7. The method of claim 1 further comprising:

caching the received second connection information in a cache;

subsequently again receiving ‘

cluster’ and

‘

server’

from the client application;

retrieving the cached connection information from the cache;

connecting the client application to ‘

server’

on ‘

cluster’

based on the retrieved cached connection information.

8. The method of claim 1 further comprising:

determining whether the connected-to server is in fact ‘

server’ and

if not;

sending a new request message to ‘

cluster’

requesting new connection information for connecting to ‘

server’

;

receiving from ‘

cluster’

a new reply message containing the requested new connection information; and

connecting the client application to ‘

server’

on ‘

cluster’

based on the received new connection information.

9. A computer-readable medium having stored thereon computer-executable instructions implementing a method of connecting a client application to a server ‘

server’

on a cluster ‘

cluster’

having a plurality of servers instantiated thereon, the method comprising;

receiving ‘

cluster’ and

‘

server’

from the client application;

sending a first request message to ‘

cluster’

requesting first connection information for connecting to ‘

server’

;

receiving from ‘

cluster’

a first reply message containing the requested first connection information;

connecting the client application to ‘

server’

on ‘

cluster’

based on the received first connection information, wherein once connected, the client application and ‘

server’

may transact business;

determining that the connection to ‘

server’

has failed, ‘

server’

presumably having been moved from a first server of the cluster to a second server of the cluster, the received first connection information corresponding to the first server;

sending a second request message to ‘

cluster’

requesting second connection information for connecting to ‘

server’

, the requested second connection information corresponding to the second server;

receiving from ‘

cluster’

a second reply message containing the requested second connection information; and

connecting the client application to ‘

server’

on ‘

cluster’

based on the received second connection information, wherein once again connected, the client application and ‘

server’

may again transact business.

10. The medium of claim 9 wherein the method comprises sending each request message as a UDP (user datagram protocol) packet addressed to a UDP address of ‘

cluster’

.

11. The medium of claim 9 wherein the method comprises receiving each requested connection information including an address of ‘

cluster’

on which ‘

server’

is listening.

12. The medium of claim 9 wherein the method comprises receiving each requested connection information including an address of ‘

cluster’

on which ‘

server’

is listening for packets formatted according to a VIA (Virtual Interface Architecture) protocol.

13. The medium of claim 9 wherein the method comprises receiving each requested connection information including each address of ‘

cluster’

on which ‘

server’

is listening and a corresponding protocol associated with the address by ‘

server’

.

14. The medium of claim 9 wherein the method comprises connecting the client application to ‘

server’

on ‘

cluster’

at an address of ‘

cluster’

on which ‘

server’

is listening for packets formatted according to a VIA (Virtual Interface Architecture) protocol.

15. The medium of claim 9 wherein the method further comprises:

caching the received second connection information in a cache;

subsequently again receiving ‘

cluster’ and

‘

server’

from the client application;

retrieving the cached connection information from the cache;

connecting the client application to ‘

server’

on ‘

cluster’

based on the retrieved cached connection information.

16. The medium of claim 9 wherein the method further comprises:

determining whether the connected-to server is in fact ‘

server’ and

if not;

sending a new request message to ‘

cluster’

requesting new connection information for connecting to ‘

server’

;

receiving from ‘

cluster’

a new reply message containing the requested new connection information; and

connecting the client application to ‘

server’

on ‘

cluster’

based on the received new connection information.