REDUCING DOWNTIME WHEN PATCHING MULTIPLE INTER-DEPENDENT SOFTWARE COMPONENTS

• US 20100257517A1
• Filed: 04/01/2009
• Published: 10/07/2010
• Est. Priority Date: 04/01/2009
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Mukherjee, Sourav, Sah, Rohit, Ganguly, Shamik, Sriram, Vijay Nichinbatl
• Granted Patent: US 9,195,455 B2
• Time in Patent Office: Days
• US Class Current: 717/168
• CPC Class Codes: G06F 8/658 Incremental updates; Differential updates

Abstract

According to an aspect of the present invention, the dependency information of software components implementing an enterprise application, is used to minimize the down time of the components when applying patches. In an embodiment, all the software components are shut down before applying patches. The patches are then applied and the components are started in a dependency order starting from an independent component. The down time is reduced as a result. According to another aspect, the shutdown also is performed in the reverse of the dependency order. The shutdown and starting are performed using recursive logic.

Claims

1. A computing system comprising:

a set of systems to store a plurality of software components implementing an enterprise application, wherein said plurality of software components are inter-dependent on each other according to a dependency order;

a developer system to send a plurality of patches sought to be applied to said plurality of software components; and

a patch tool to apply said plurality of patches on respective software components after all of said plurality of software components are in a shutdown state,

said patch tool to start up said plurality of software components in said dependency order starting first with an independent component, wherein said independent component is contained in said plurality of software components.

2. The computing system of claim 1, wherein said dependency order is represented by a dependency data, wherein said patch tool examines said dependency data to determine said dependency order.

3. The computing system of claim 2, wherein said dependency data comprises a plurality of metadata, each metadata being included associated with a respective component, the metadata associated with a component indicating a set of components which are dependent on the component,

each of said plurality of software components being associated with a corresponding one of a plurality of deployment logics,

said patch tool being operable to first send a shutdown command to the deployment logic associated with said independent component,

wherein the deployment logic of each of said plurality of software components, including said independent component, is designed to send a shutdown request to the deployment logics of dependent components indicated by corresponding one of said plurality of metadata and is designed to shutdown the associated software component after receiving an indication that corresponding dependent components are shutdown, whereby shutdown happens in a reverse dependency order ending with said independent component,

said patch tool being operable to then send a patch and startup command to the deployment logic associated with said independent component,

wherein the deployment logic of each of said plurality of software components, including said independent component, is designed to first apply to applicable ones of said plurality of patches to the associated software component, then startup the associated software component and then send another indication to the deployment logics of the dependent components according to the associated metadata, whereby said plurality of software components are started in said dependency order starting with said independent component.

4. The computing system of claim 3, wherein said indications for shutdown of the dependent components of a component are sent in parallel.

5. The computing system of claim 4, wherein said indications for patch and startup of the dependent components of a component are also sent in parallel.

6. The computing system of claim 2, wherein said dependency data is formed in said patch tool.

7. The computing system of claim 2, wherein said dependency data is formed in said developer system and sent to said patch tool along with said plurality of patches in the form a deployment package.

8. The computing system of claim 1, wherein said plurality of software components comprise a database system, a cluster ready service, a automated storage management and a application server software.

9. A machine readable medium carrying one or more sequences of instructions for causing a system to facilitate applying of patches to a plurality of software components implementing an enterprise application, said plurality of software components being inter-dependent on each other, wherein execution of said one or more sequences of instructions by one or more processors contained in said system causes said system to perform the actions of:

receiving a plurality of patches that are sought to be applied to said plurality of software components;

forming a metadata corresponding to each of said plurality of software components, wherein the metadata corresponding to a software component indicates the specific ones of said plurality of software components that are dependent on the software component;

including the metadata corresponding to the software component and a set of patches sought to be applied to the software component in a component package, the set of patches being contained in said plurality of patches; and

generating a deployment package containing the component packages corresponding to the plurality of software components for deployment.

10. The machine readable medium of claim 9, wherein said system is a developer system, further comprising one or more instructions for:

sending said deployment package to a patch tool which applies said plurality of patches to said plurality of software components.

11. The machine readable medium of claim 9, wherein said system also implements a patch tool which applies said plurality of patches to said plurality of software components after said generating.

12. The machine readable medium of claim 9, further comprising one or more instructions for:

associating each of said plurality of software components with a corresponding one of a plurality of deployment logics;

sending a shutdown command to the deployment logic associated with an independent component, wherein said independent component is contained in said plurality of software components;

sending further shutdown commands to the deployment logics of dependent components indicated by corresponding one of said plurality of metadata and is designed to shutdown the associated software component after receiving an indication that corresponding dependent components are shutdown; and

sending patch and startup commands to the deployment logic of each of said plurality of software components, including said independent component,

where each of the deployment logic is designed to first apply to applicable ones of said plurality of patches to the associated software component, then startup the associated software component and then send another indication to the deployment logics of the dependent components according to the associated metadata.

13. The machine readable medium of claim 12, wherein said plurality of software components comprise a database system, a cluster ready service, a automated storage management and a application server software.

14. A method of applying patches to a plurality of software components implementing an enterprise application, said plurality of software components being inter-dependent on each other according to a dependency order, said method comprising:

receiving a plurality of patches sought to be applied on said plurality of software components;

shutting down said plurality of software components executing on a set of systems;

applying said plurality of patches to said plurality of software components; and

starting said plurality of software components in said dependency order starting with an independent component comprised in said plurality of software components.

15. The method of claim 14, wherein said dependency order is represented by a dependency data, said method further comprising examining said dependency data to determine said dependency order, wherein said starting is performed based on said examining.

16. The method of claim 15, wherein said plurality of software components are also shutdown in said dependency order determined based on said dependency data.

17. The method of claim 16, wherein said shutting down and starting are performed using a recursive logic implemented associated with each of said plurality of software components.

18. The method of claim 17, wherein said dependency data comprises a plurality of metadata, each metadata being included associated with a respective component, the metadata associated with a component indicating a set of components which are dependent on the component,

each of said plurality of software components being associated with a corresponding one of a plurality of deployment logics,

said method further comprising:

sending a shutdown command to the deployment logic associated with said independent component,

wherein the deployment logic of each of said plurality of software components, including said independent component, is designed to send a shutdown request to the deployment logics of dependent components indicated by corresponding one of said plurality of metadata and is designed to shutdown the associated software component after receiving an indication that corresponding dependent components are shutdown, whereby shutdown happens in a reverse dependency order ending with said independent component, and

sending a patch and startup command to the deployment logic associated with said independent component after completion of shutdown of all of said plurality of software components,

wherein the deployment logic of each of said plurality of software components, including said independent component, is designed to first apply to applicable ones of said plurality of patches to the associated software component, then startup the associated software component and then send another indication to the deployment logics of the dependent components according to the associated metadata, whereby said plurality of software components are started in said dependency order starting with said independent component.

19. The method of claim 18, wherein said indications for shutdown of the dependent components of a component are sent in parallel.

20. The method of claim 18, wherein said indications for patch and startup of the dependent components of a component are also sent in parallel.