System Upgrade Under High Availability Constraints

• US 20140258999A1
• Filed: 03/07/2013
• Published: 09/11/2014
• Est. Priority Date: 03/07/2013
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: TOEROE, Maria, KATIHAR, Ekansh Singh
• Granted Patent: US 9,430,214 B2
• Time in Patent Office: Days
• US Class Current: 717/168
• CPC Class Codes: G06F 8/60 Software deployment
• CPC Class Codes: G06F 8/61 Installation
• CPC Class Codes: G06F 8/65 Updates security arrangements therefor G06F21/57
• CPC Class Codes: G06F 8/656 while running
• CPC Class Codes: G06F 8/70 Software maintenance or management

Abstract

An embedded step is executed to upgrade multiple layers of a system. The embedded step includes multiple nested steps, and each nested step includes a sequence of actions for upgrading one of the layers. A finite state machine defines the state transitions for each nested step. During the execution of the nested steps, an outer step of the nested steps is transitioned from an executing state into a waiting state after the outer step completes a tear-down. During the transitioning of the outer step, the outer step sends a trigger to an inner step of the nested steps to cause the inner step to transition into the executing state. When the inner step is completed, the outer step is transitioned out of the waiting state to execute a build-up phase. Execution of the embedded step is completed when an outermost step of the nested steps is completed.

Claims

1. A method performed by a computer system for executing an embedded step to upgrade a plurality of layers in a software subsystem, the embedded step including a plurality of nested steps, each of the nested steps including a sequence of actions for upgrading one of the layers, the method comprising the steps of:

executing, by the computer system for each of the nested steps, state transitions according to a finite state machine (FSM), wherein the step of executing further comprises;

transitioning an outer step of the nested steps from an executing state into a waiting state of the FSM after the outer step completes a tear-down phase of the sequence of actions, wherein, during the transitioning of the outer step, the outer step sends a trigger to an inner step of the nested steps to cause the inner step to transition into the executing state; and

transitioning the outer step out of the waiting state to execute a build-up phase of the sequence of actions when the inner step is completed; and

completing the execution of the embedded step when an outermost step of the nested steps is completed.

2. The method of claim 1, further comprising the steps of:

detecting that an error occurs in a given one of the nested steps during the upgrade;

retrying the given nested step in a same phase in which the error occurs; and

if the number of retries reaches a predetermined number without success, undoing the upgrade for each of the nested steps.

3. The method of claim 2, wherein subsequent to the predetermined number of retries, the method further comprises the step of:

transitioning the given nested step into a partial undone state of the FSM to wait for all of its inner steps to undo the upgrade.

4. The method of claim 1, wherein, in response to a rollback trigger, the method further comprises the steps of:

transitioning the outer step into a rollback waiting state of the FSM after the outer step rolled back the build-up phase, wherein, during the transitioning of the outer step, the outer step sends another trigger to the inner step to cause the inner step to roll back; and

transitioning the outer step out of the rollback waiting state to roll back the tear-down phase when the inner step has rolled back.

5. The method of claim 4, further comprising the steps of:

detecting that an error occurs during the rollback of a given one of the nested steps; and

retrying the given nested step in a same phase in which the error occurs; and

if the number of retries reaches a predetermined number without success, ending the rollback without all of the nest steps being in a same state of the FSM.

6. The method of claim 1, wherein the nested steps include a hardware layer.

7. The method of claim 1, wherein the nested steps include an operating system layer.

8. The method of claim 1, wherein at least one of the nested steps is a locked reduced upgrade step, which includes actions that lock a symmetric activation unit and unlocks the symmetric activation unit, wherein the symmetric activation unit includes entities that are de-activated and re-activated by the upgrade.

9. The method of claim 1, wherein the nested steps include one or more selected from a group consisting of:

an in-phase normal step, an out-of-phase normal step, a reduced upgrade step, and a locked reduced upgrade step.

10. The method of claim 1, further comprising the step of:

extending a schema for embedded steps that upgrade the software subsystem to specify each embedded step to include one or more nested steps.

11. A computer system adapted to execute an embedded step to upgrade a plurality of layers in a software subsystem, the embedded step including a plurality of nested steps, each of the nested steps including a sequence of actions for upgrading one of the layers, the computer system comprising:

one or more processors;

one or more memory devices coupled to the one or more processors; and

an upgrade engine coupled to the one or more processors and the one or more memory devices, the upgrade engine adapted to;

execute, for each of the nested steps, state transitions according to a finite state machine (FSM), the upgrade engine is further adapted to;

transition an outer step of the nested steps from an executing state into a waiting state of the FSM after the outer step completes a tear-down phase of the sequence of actions, wherein, during the transitioning of the outer step, the outer step sends a trigger to an inner step of the nested steps to cause the inner step to transition into the executing state; and

transition the outer step out of the waiting state to execute a build-up phase of the sequence of actions when the inner step is completed; and

complete the execution of the embedded step when an outermost step of the nested steps is completed.

12. The computer system of claim 11, wherein the upgrade engine is further adapted to:

detect that an error occurs in a given one of the nested steps during the upgrade;

retry the given nested step in a same phase in which the error occurs; and

if the number of retries reaches a predetermined number without success, undo the upgrade for each of the nested steps.

13. The computer system of claim 12, wherein subsequent to the predetermined number of retries, the upgrade engine is further adapted to transition the given nested step into a partial undone state of the FSM to wait for all of its inner steps to undo the upgrade.

14. The computer system of claim 11, wherein, in response to a rollback trigger, the upgrade engine is further adapted to:

transition the outer step into a rollback waiting state of the FSM after the outer step rolled back the build-up phase, wherein, during the transitioning of the outer step, the outer step sends another trigger to the inner step to cause the inner step to roll back; and

transition the outer step out of the rollback waiting state to roll back the tear-down phase when the inner step has rolled back.

15. The computer system of claim 14, wherein the upgrade engine is further adapted to:

detect that an error occurs during the rollback of a given one of the nested steps; and

retry the given nested step in a same phase in which the error occurs; and

if the number of retries reaches a predetermined number without success, ending the rollback without all of the nested steps being in a same state of the FSM.

16. The computer system of claim 11, wherein the nested steps include a hardware layer.

17. The computer system of claim 11, wherein the nested steps include an operating system layer.

18. The computer system of claim 11, wherein at least one of the nested steps is a locked reduced upgrade step, which includes actions that lock a symmetric activation unit and unlocks the symmetric activation unit, wherein the symmetric activation unit includes entities that are de-activated and re-activated by the upgrade.

19. The computer system of claim 11, wherein the nested steps include one or more selected from a group consisting of:

an in-phase normal step, an out-of-phase normal step, a reduced upgrade step, and a locked reduced upgrade step.

20. The computer system of claim 11, wherein the embedded step is defined according to a schema that specifies the embedded step to include one or more nested steps.