Apparatus and method for server network management to provide zero planned retrofit downtime

• US 8,521,853 B2
• Filed: 06/29/2007
• Issued: 08/27/2013
• Est. Priority Date: 06/29/2007
• Status: Active Grant

• Current Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Original Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Inventors: Rathunde, Dale Frank, Susa, Paul M., Oommen, Saju George
• Primary Examiner(s): Blair, Douglas B.
• Assistant Examiner(s): Nguyen, Angela
• Application Number:
  US11/771,429
• Publication Number:
  US 20090007135A1
• Time in Patent Office: 2,251 Days
• Field of Search: 709/221
• US Class Current: 709/221
• CPC Class Codes: G06F 8/65 Updates security arrangements therefor G06F21/57

Abstract

Methods and systems are presented for updating software applications in a processor cluster, in which the cluster is divided into first and second processor groups and the first group is isolated from clients and from the second group with respect to network and cluster communications by application of IP filters. The first group of processors is updated or retrofitted with the new software and brought to a ready-to-run state while the second group is active to serve clients. The first group is then transitioned to an in-service state after isolating the then-active service providing application on second group. Thereafter, the second group of processors is offlined, updated or retrofitted, and transitioned to an in-service state to complete the installation of the new application version across the cluster with reduced or zero downtime and without requiring backward software compatibility.

Claims

1. A method of updating or retrofitting software applications in a cluster of operatively coupled processors running an application to service clients, the method comprising:

segmenting the cluster of operatively coupled processors running the software application into first and second groups of processors, the first and second groups of processors individually including at least one processor of the cluster;

applying IP filters to the processors and ports of each of the first group of processors to isolate the first group of processors from service clients and from the second group of processors with respect to a network and a cluster communication related to the software application;

updating or retrofitting the software application on the first group of processors;

bringing application states on the first group of processors from an active state to a ready-to-run state in which the software application on the first group of processors is active and partially initialized but not in-service;

transitioning application states on the second group of processors from an active state to a ready-to-run state in which the application on the second group of processors is active and partially initialized but not in-service;

isolating the second group of processors from service clients and from processors of the first group;

bringing application states on the first group of processors to an in-service state;

updating or retrofitting the software application on the second group of processors; and

bringing application states on the second group of processors to an in-service state.

2. The method of claim 1, wherein the first and second groups of processors include roughly the same number of processors.

3. The method of claim 1, wherein the first and second groups of processors include different numbers of processors, and wherein the smaller of the first and second groups of processors includes enough processors to service the clients when the other group is offline.

4. The method of claim 1, further comprising removing the IP filters from the processors and ports of the first group of processors after bringing application states on the first group of processors to the in-service state.

5. The method of claim 1, further comprising waiting a predetermined period of time after bringing application states on the first group of processors to an in-service state and before updating or retrofitting the software application on the second group of processors.

6. The method of claim 1, further comprising disabling an address resolution protocol in the first group of processors and setting up static routes for the first group of processors after updating or retrofitting the software application on the first group of processors and before bringing application states on the first group of processors to a ready-to-run state.

7. The method of claim 1, wherein bringing application states on the first group of processors to the ready-to-run state temporarily establishes a controlled split cluster environment with two isolated instances of the software application on different software application versions, with the second group of processors being in-service or active while the first group of processors is managed in a partial operating condition on a newer version of the application.

8. The method of claim 1, wherein the IP filters are applied to the processors and ports of each of the first group of processors using firewall software running on the first group of processors.

9. The method of claim 5, further comprising:

during the predetermined period of time and before updating or retrofitting the software application on the second group of processors, determining whether a problem exists with the updated or retrofit version of the software application on the first group of processors; and

if the problem exists with the updated or retrofit version of the software application on the first group of processors;

taking the first group of processors offline,

removing the IP filters for the first group of processors,

publishing IP addresses for the second group of processors,

changing application state to in-service for the second group of processors,

backing the first group of processors out to an earlier software application version,

transitioning a processor state of the first group of processors to online, and

transitioning application state of the first group of processors to in-service.

10. A system for updating or retrofitting software applications in a cluster of operatively coupled processors running a software application to service clients, comprising:

a network server operatively coupled with the cluster of operatively coupled processors running the software application, the network server running an update/retrofit application;

the network server operative to segment the cluster of operatively coupled processors into first and second groups of processors, the first and second groups individually including at least one processor of the cluster;

the network server operative to apply IP filters to processors and ports of each of the first group of processors to isolate the first group of processors from service clients and from the second group of processors with respect to network and cluster communication related to the application;

the network server operative to update or retrofit the software application on the first group of processors;

the network server operative to bring application states on the first group of processors from an active state to a ready-to-run state in which the software application on the first group of processors is active and partially initialized but not in-service;

the network server is operative to transition application states on the second group of processors from an active state to the ready-to-run state in which the software application on the second group of processors is active and partially initialized but not in-service;

the network server is operative to isolate the second group of processors from clients and from the first group of processors after bringing application states on the first group of processors to the ready-to-run state and before bringing application states on the first group of processors to the in-service state;

the network server operative to bring application states on the first group of processors to the in-service state;

the network server operative to update or retrofit the software application on the second group of processors; and

the network server operative to bring application states on the second group of processors to the in-service state.

11. The system of claim 10, wherein the network server is operative to disable an address resolution protocol in the first group of processors and wherein the network server is operative to set up static routes for the first group of processors after updating or retrofitting the software application on the first group of processors and before bringing the application states on the first group of processors to the ready-to-run state.

12. The system of claim 10, wherein the network server is operative to bring application states on the first group of processors to the ready-to-run state to temporarily establish a controlled split cluster environment with two isolated instances of the software application on different software application versions, with the second group of processors being in-service or active while the first group of processors is managed in a partial operating condition on a newer version of the software application.

13. The system of claim 10, wherein the network server is operative:

to wait a predetermined period of time after bringing application states on the first group of processors to the in-service state and before updating or retrofitting the software application on the second group of processors;

during the predetermined period of time and before updating or retrofitting the application on the second group of processors to determine whether a problem exists with the updated or retrofit version of the application on the first group of processors; and

if the problem exists with the updated or retrofit version of the software application on the first group of processors, to;

take the first group of processors offline,

remove the IP filters for the first group of processors,

publish IP addresses for the second group of processors,

change application state to in-service for the second group of processors,

back the first group of processors out to an earlier software application version,

transition a processor state of the first group of processors to online, and

transition application state of the first group of processors to in-service.

14. The system of claim 12, wherein the network server is operative to disable an address resolution protocol in the first group of processors and setting up static routes for the first group of processors after updating or retrofitting the software application on the first group of processors and before bringing application states on the first group of processors to the ready-to-run state.

15. A non-transitory computer-readable medium having computer-executable instructions for performing steps comprising:

segmenting a cluster of operatively coupled processors running a software application into first and second groups of processors, the first and second groups of processors individually including at least one processor of the cluster;

applying IP filters to processors and ports of each of the first group of processors to isolate the first group of processors from service clients and from the second group of processors with respect to a network and cluster a communication related to the software application;

updating or retrofitting the software application on the first group of processors;

bringing application states on the first group of processors from an active state to a ready-to-run state in which the software application on the first group of processors is active and partially initialized but not in-service;

transitioning application states on the second group of processors from an active state to the ready-to-run state in which the software application on the second group of processors is active and partially initialized but not in-service;

isolating the second group of processors from service clients and from the first group of processors;

bringing application states on the first group of processors to an in-service state;

updating or retrofitting the software application on the second group of processors; and

bringing application states on the second group of processors to the in-service state.

16. The non-transitory computer-readable medium of claim 15, comprising computer-executable instructions for bringing application states on the first group of processors to the ready-to-run state to temporarily establish a controlled split cluster environment with two isolated instances of the software application on different software application versions, with the second group of processors being in-service or active while the first group of processors is managed in a partial operating condition on a newer version of the software application.

17. The non-transitory computer-readable medium of claim 15, comprising computer-executable instructions for:

waiting a predetermined period of time after bringing the application states on the first group of processors to an in-service state and before updating or retrofitting the application on the second group of processors;

during the predetermined period of time and before updating or retrofitting the application on the second group of processors, determining whether a problem exists with the updated or retrofit version of the application on the first group of processors; and

if a problem exists with the updated or retrofit version of the application on the first group of processors;

taking the first group of processors offline,

removing the IP filters for the first group of processors,

publishing IP addresses for the second group of processors,

changing the application state to in-service for the second group of processors,

backing the first group of processors out to an earlier application version,

transitioning a processor state of the first group of processors to online, and

transitioning the application state of the first group of processors to in-service.

18. The non-transitory computer-readable medium of claim 16, comprising computer-executable instructions for disabling an address resolution protocol in the first group of processors and setting up static routes for the first group of processors after updating or retrofitting the software application on the first group of processors and before bringing the software application states on the first group of processors to the ready-to-run state.

19. A method of updating or retrofitting software applications in a cluster of operatively coupled processors running a software application to service clients, the method comprising:

segmenting a cluster of operatively coupled processors running an application into first and second groups of processors, the first and second groups of processors individually including at least one processor of the cluster;

applying IP filters to processors and ports of each of the first group of processors to isolate the first group of processors from service clients and from the second group of processors with respect to network and cluster communication related to the application;

updating or retrofitting the software application on the first group of processors;

bringing application states on the first group of processors from an active state to a ready-to run state in which the application on the first group of processors is active and partially initialized but not in-service;

isolating the second group of processors from service clients and from processors of the first group;

bringing application states on the first group of processors to an in-service state;

updating or retrofitting the software application on the second group of processors; and

bringing application states on the second group of processors to the in-service state.

20. The method of claim 19, comprising:

waiting a predetermined period of time after bringing application states on the first group of processors to in-service state and before updating or retrofitting the software application on the second group of processors;

during the predetermined period of time and before updating or retrofitting the software application on the second group of processors, determining whether a problem exists with the updated or retrofit version of the software application on the first group of processors; and

if the problem exists with the updated or retrofit version of the software application on the first group of processors;

taking the first group of processors offline,

removing the IP filters for the first group of processors,

publishing IP addresses for the second group of processors,

changing application state to in-service for the second group of processors,

backing the first group of processors out to an earlier software application version,

transitioning a processor state of the first group of processors to online, and

transitioning application state of the first group of processors to in-service.

21. The method of claim 19, wherein updating or retrofitting the application on the first group of processors includes retrofitting the first group of processors with a new version of the software application that is not backward compatible with a prior version of the software application on the second group of processors, and wherein bringing the application states on the first group of processors to the ready-to run state includes allowing initializing processes in the first group of processors to access new database schema in the first group of processors and initializing all internal data structures with the first group of processors.