Capturing and restoring database session state

• US 9,104,739 B2
• Filed: 03/28/2013
• Issued: 08/11/2015
• Est. Priority Date: 07/01/2008
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Liang, Chao, Cao, Feng, Pingte, Rajendra
• Primary Examiner(s): Bullock, Joshua
• Application Number:
  US13/852,986
• Publication Number:
  US 20130232113A1
• Time in Patent Office: 866 Days
• Field of Search: 707/655, 707/615
• US Class Current: 1/1
• CPC Class Codes: G06F 11/3414 Workload generation, e.g. scripts, playback
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 2201/87 Monitoring of transactions

Abstract

Techniques are described herein for capturing and restoring database session state. Production database server components save the session state of each of a plurality of database sessions. The components store workload units that are processed in these sessions. The components store updated session states in response to certain events. Thus, the components may capture multiple session states, pertaining to various different points in time, for each session. The captured session states and the captured workload are moved to a test database server. A user selects, from among the time points represented by the session states, a point in time at which the user would like workload replay to begin. Sessions are re-created on the test database server. Session states of these sessions are set to reflect the session states as they existed at the user-selected time point. Workload units are replayed in the sessions relative to the test database server.

Claims

1. A computer-implemented method comprising steps of:

determining that a database session in a first database server is at a transaction boundary;

in response to determining that said database session is at a transaction boundary, capturing a session state of said database session;

after capturing said session state, capturing a workload that is processed in said database session by said first database server;

recreating, based on said session state, said database session in a second database server that is separate from said first database server; and

replaying at least a portion of said workload in said recreated database session in said second database server.

2. The method of claim 1, wherein:

capturing a session state of said database session comprises capturing a session state of said database session that is current as of a time that said transaction boundary was reached.

3. The method of claim 1, wherein said step of capturing said session state comprises storing said session state on a persistent non-volatile computer-readable storage medium.

4. The method of claim 1, wherein said step of recreating said database session comprises loading said session state from a persistent non-volatile computer-readable storage medium.

5. The method of claim 1, the steps further comprising receiving, from a user, an indication of a particular time point from which workload replay should begin;

and

wherein replaying at least a portion of said workload includes replaying at least a first portion of said workload starting at said particular time point, in said recreated database session in said second database server.

6. The method of claim 5, wherein replaying at least a first portion of said workload starting at said particular time point, in said recreated database session in said second database server comprises replaying said first portion of said workload without replaying any portion of said workload that is associated with a time that is prior to said particular time point.

7. The method of claim 5, the steps further comprising:

after replaying said first portion of said workload, replaying, in said second database server, another portion of said workload starting at another user-specified point in time that is different than said particular time point.

8. The method of claim 1, wherein:

capturing a session state of said database session comprises storing, on a non-volatile computer-readable storage medium, one or more variables that are contained in volatile memory of a first computer on which said first database server executes, along with one or more values of those one or more variables; and

recreating, based on said session state, said database session in said second database server comprises loading, from said non-volatile computer-readable storage medium, and into a volatile memory of a second computer on which said second database server executes, said one or more variables along with said one or more values of those one or more variables;

wherein said first computer is separate from said second computer; and

wherein said first database server does not execute concurrently with said second database server.

9. A non-transitory computer-readable medium storing one or more sequences of instructions which, when executed by one or more processors, cause:

determining that a database session in a first database server is at a transaction boundary;

in response to determining that said database session is at a transaction boundary, capturing a session state of said database session;

after capturing said session state, capturing a workload that is processed in said database session by said first database server;

recreating, based on said session state, said database session in a second database server that is separate from said first database server; and

replaying at least a portion of said workload in said recreated database session in said second database server.

10. The non-transitory computer-readable medium of claim 9, wherein:

capturing a session state of said database session comprises capturing a session state of said database session that is current as of a time that said transaction boundary was reached.

11. The non-transitory computer-readable medium of claim 9, wherein capturing said session state comprises storing said session state on a persistent non-volatile computer-readable storage medium.

12. The non-transitory computer-readable medium of claim 9, wherein recreating said database session comprises loading said session state from a persistent non-volatile computer-readable storage medium.

13. The non-transitory computer-readable medium of claim 9, the sequences of instructions comprising instructions, that when executed by one or more precessors, cause receiving, from a user, an indication of a particular time point from which workload replay should begin;

and

wherein replaying at least a portion of said workload includes replaying at least a first portion of said workload starting at said particular time point, in said recreated database session in said second database server.

14. The non-transitory computer-readable medium of claim 13, wherein replaying at least a first portion of said workload starting at said particular time point, in said recreated database session in said second database server comprises replaying said first portion of said workload without replaying any portion of said workload that is associated with a time that is prior to said particular time point.

15. The non-transitory computer-readable medium of claim 13, the sequences of instructions comprising instructions, that when executed by one or more processors, cause:

after replaying said first portion of said workload, replaying, in said second database server, another portion of said workload starting at another user-specified point in time that is different than said particular time point.

16. The non-transitory computer-readable medium of claim 9, wherein:

capturing a session state of said database session comprises storing, on a non-volatile computer-readable storage medium, one or more variables that are contained in volatile memory of a first computer on which said first database server executes, along with one or more values of those one or more variables; and

recreating, based on said session state, said database session in said second database server comprises loading, from said non-volatile computer-readable storage medium, and into a volatile memory of a second computer on which said second database server executes, said one or more variables along with said one or more values of those one or more variables;

said first computer is separate from said second computer; and

the sequences of instructions comprise instructions, that when executed by one or more processors, cause said first database server to not execute concurrently with said second database server.