AUTONOMOUS PRIMARY NODE ELECTION WITHIN A VIRTUAL INPUT/OUTPUT SERVER CLUSTER

• US 20120179798A1
• Filed: 01/11/2011
• Published: 07/12/2012
• Est. Priority Date: 01/11/2011
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Pafumi, James A., Rosales, Jacob Jason, Rosas, Morgan Jeffrey, Vallabhaneni, Vasu
• Granted Patent: US 8,583,773 B2
• Time in Patent Office: Days
• US Class Current: 709/223
• CPC Class Codes: G06F 9/5072 Grid computing

Abstract

In a data processing system having a plurality of virtual input/output servers (VIOSes) configured within a VIOS cluster, a method, data processing system and computer program product provide for autonomous election of a primary node within a virtual input/output server (VIOS) cluster. A first VIOS performs the functions of: detecting that a primary node is required for the VIOS cluster; and autonomously initiating an election process to elect a next primary node from among the VIOSes within the VIOS cluster. When the first VIOS meets the pre-established requirements for becoming a primary node, the first VIOS obtains a lock on a primary node ID field within a VIOS database (DB) and then initiates a primary node commit process to assign the first VIOS as the primary node. The first VIOS issues a notification to the VIOS cluster to notify the other VIOSes that a primary node has been elected.

Claims

1. In a data processing system having one or more computing electronic complexes (CECs) that comprise a plurality of virtual input/output (I/O) servers (VIOSes) configured within a VIOS cluster, which has an associated VIOS database (DB) and in which each VIOS provides I/O resources to one or more logical partitions (LPARs) executing within the one or more CECs, a first VIOS of the VIOS cluster executing code on a processing resource to perform a method comprising:

the first VIOS of the VIOS cluster detecting that a primary node is required for the VIOS cluster; and

autonomously initiating an election process to elect a next primary node from among one or more VIOSes within the VIOS cluster.

2. The method of claim 1, further comprising:

determining whether a first node meets one or more pre-established requirements to become a primary node by checking one or more configuration parameters of the first node against the one or more pre-established requirements for a node to become the primary node; and

in response to the first node not meeting the pre-established requirements, activating a primary node discovery process which discovers and notifies a next node from within an elector list of a need for electing a primary node;

wherein the first node becomes an elector node if there is a next node that meets the pre-established requirements to be a candidate for election.

3. The method of claim 2, further comprising:

in response to the first node meeting the pre-established requirements, determining whether a primary node identifier (ID) field within the VIOS DB is locked by another node;

in response to the primary node ID field being locked, setting a primary node election timer to track an elapsed time from when the first node made the attempt to become the primary node, wherein the timer expires after a pre-set amount of time and the expiration of the timer prior to another node being elected as the primary node triggers the first node to re-initiates an attempt to become the primary node; and

in response to the primary node ID field being unlocked;

locking the Primary Node ID field within the VIOS DB; and

initiating a primary node commit process to assign the first node as the primary node.

4. The method of claim 3, wherein the initiating the primary node commit process comprises:

sending a commit signal to the VIOS DB to update the primary node ID field with the unique ID of the first node;

unlocking the Primary Node ID field following the update of primary node ID field with the unique ID of the first node; and

issuing a notification message to all other nodes in the cluster to inform the other nodes that a new primary node has been elected.

5. The method of claim 3, further comprising:

determining that performance of a primary node function/action is required;

issuing a query to the VIOS DB to determine whether the first node is still the primary node by checking a current value of the primary node ID field; and

performing the primary node function only if the first node is still the primary node.

6. The method of claim 5, further comprising:

in response to the first node no longer being the primary node, determining whether another node has committed as the primary node; and

in response to another node having committed as the primary node;

retrieving the primary node ID from the primary node ID field; and

updating a node local copy of the primary node ID within the first node.

7. The method of claim 2, wherein the activating a primary node discovery process further comprises:

retrieving a list of active nodes in the VIOS cluster;

communicating with a next active node in the list of the need to elect a primary node; and

in response to detecting that a next node is elected as the primary node;

stopping the election timer; and

updating a local primary node ID that is maintained in local storage.

8. The method of claim 7, further comprising, in response to none of the active nodes being able to be elected as the primary node, issuing a notification of a failure to elect a primary node from among the active nodes within the cluster.

9. The method of claim 3, further comprising:

determining that the first node is no longer capable of acting as the primary node;

relinquishing the primary node responsibility by updating the Primary Node ID in the VIOS DB to a null value; and

discovering and notifying a next candidate in the list of active nodes, wherein the notifying includes one or more or a directly messaging the next candidate and broadcasting a primary node election message so that the other nodes that meet one or more primary node criteria can initiate the election process.

10. The method of claim 3, wherein:

different primary nodes perform different functions and the broadcasting comprises indentifying, within the broadcasted message, information about what primary role is being relinquished;

the primary node is a role specific VIOS operating as one of a primary notification node and a primary clean-up node and is configured to provide one or more services to other VIOSes within the VIOS cluster, the one or more services comprising;

posting life cycle events;

posting alerts;

journaling alerts; and

providing database and general clean up; and

each role is provided by at least one primary node and multiple roles can be provided by a single primary node;

11. The method of claim 2, wherein the VIOS DB comprises an elector node identifier (ID), said method further comprising:

checking whether an elector node ID is set in the VIOS DB; and

in response to the elector node ID not being set, setting the elector node ID to the unique ID of the first node and initiating the election process.

18. The computer program product of claim 2, wherein the function of activating a primary node discovery process further comprises:

retrieving a list of active nodes in the VIOS cluster;

communicating with a next active node in the list of the need to elect a primary node; and

in response to detecting that a next node is elected as the primary node;

stopping the election timer; and

updating a local primary node ID that is maintained in local storage.

19. The computer program product of claim 7, further comprising program instructions that execute to perform the functions of:

in response to none of the active nodes being able to be elected as the primary node, issuing a notification of a failure to elect a primary node from among the active nodes within the cluster.

12. A computer program product comprising:

a computer readable medium; and

program code on the computer readable medium that when executed by a processor of a first virtual input/output server within a VIOS cluster comprised of a plurality of VIOSes and a VIOS database (DB) and where the plurality of VIOSes provide I/O resources to one or more logical partitions (LPARs), causes the first VIOS to perform the functions of;

detecting that a primary node is required for the VIOS cluster; and

autonomously initiating an election process to elect a next primary node from among one or more VIOSes within the VIOS cluster;

13. The computer program product of claim 12, further comprising program instructions that execute to perform the functions of:

determining whether a first node meets one or more pre-established requirements to become a primary node by checking one or more configuration parameters of the first node against the one or more pre-established requirements for a node to become the primary node; and

in response to the first node not meeting the pre-established requirements, activating a primary node discovery process which discovers and notifies a next node from within an elector list of a need for electing a primary node;

wherein the first node becomes an elector node if there is a next node that meets the pre-established requirements to be a candidate for election.

14. The computer program product of claim 12, further comprising program instructions that execute to perform the functions of:

in response to the first node meeting the pre-established requirements, determining whether a primary node identifier (ID) field within the VIOS DB is locked by another node;

in response to the primary node ID field being locked, setting a primary node election timer to track an elapsed time from when the first node made the attempt to become the primary node, wherein the timer expires after a pre-set amount of time and the expiration of the timer prior to another node being elected as the primary node triggers the first node to re-initiates an attempt to become the primary node; and

in response to the primary node ID field being unlocked;

locking the Primary Node ID field within the VIOS DB; and

initiating a primary node commit process to assign the first node as the primary node.

15. The computer program product of claim 14, wherein the function of initiating the primary node commit process comprises:

sending a commit signal to the VIOS DB to update the primary node ID field with the unique ID of the first node;

unlocking the Primary Node ID field following the update of primary node ID field with the unique ID of the first node; and

issuing a notification message to all other nodes in the cluster to inform the other nodes that a new primary node has been elected.

16. The computer program product of claim 14, further comprising program instructions that execute to perform the functions of:

determining that performance of a primary node function/action is required;

issuing a query to the VIOS DB to determine whether the first node is still the primary node by checking a current value of the primary node ID field; and

performing the primary node function only if the first node is still the primary node.

17. The computer program product of claim 16, further comprising program instructions that execute to perform the functions of:

in response to the first node no longer being the primary node, determining whether another node has committed as the primary node; and

in response to another node having committed as the primary node;

retrieving the primary node ID from the primary node ID field; and

updating a node local copy of the primary node ID within the first node.

20. The computer program product of claim 14, further comprising program instructions that execute to perform the functions of:

determining that the first node is no longer capable of acting as the primary node;

relinquishing the primary node responsibility by updating the Primary Node ID in the VIOS DB to a null value; and

discovering and notifying a next candidate in the list of active nodes, wherein the notifying includes one or more or a directly messaging the next candidate and broadcasting a primary node election message so that the other nodes that meet one or more primary node criteria can initiate the election process.

21. The computer program product of claim 14, wherein:

different primary nodes perform different functions and the broadcasting comprises indentifying, within the broadcasted message, information about what primary role is being relinquished;

the primary node is a role specific VIOS operating as one of a primary notification node and a primary clean-up node and is configured to provide one or more services to other VIOSes within the VIOS cluster, the one or more services comprising;

posting life cycle events;

posting alerts;

journaling alerts; and

providing database and general clean up; and

each role is provided by at least one primary node and multiple roles can be provided by a single primary node;

22. The computer program product of claim 13, wherein the VIOS DB comprises an elector node identifier (ID), said program instructions further comprising program instructions that execute to perform the functions of:

checking whether an elector node ID is set in the VIOS DB; and

in response to the elector node ID not being set, setting the elector node ID to the unique ID of the first node and initiating the election process.

23. A data processing system comprising:

one or more processors providing processing resources;

one or more memories coupled to the one or more processors;

a cluster-aware (CA) operating system (OS) which generates a first virtual input/output server (VIOS) that is communicatively couple to other VIOSes and to a VIOS database (DB) to create a VISO cluster, wherein the CA_OS comprises a utility executing on a processing resource of the first VIOS, which utility causes the first VIOS to perform the functions of;

detecting that a primary node is required for the VIOS cluster;

autonomously initiating an election process to elect a next primary node from among one or more VIOSes within the VIOS cluster;

determining whether a first node meets one or more pre-established requirements to become a primary node by checking one or more configuration parameters of the first node against the one or more pre-established requirements for a node to become the primary node;

in response to the first node not meeting the pre-established requirements, activating a primary node discovery process which discovers and notifies a next node from within an elector list of a need for electing a primary node;

wherein the first node becomes an elector node if there is a next node that meets the pre-established requirements to be a candidate for election.

in response to the first node meeting the pre-established requirements, determining whether a primary node identifier (ID) field within the VIOS DB is locked by another node;

in response to the primary node ID field being locked, setting a primary node election timer to track an elapsed time from when the first node made the attempt to become the primary node, wherein the timer expires after a pre-set amount of time and the expiration of the timer prior to another node being elected as the primary node triggers the first node to re-initiates an attempt to become the primary node; and

in response to the primary node ID field being unlocked;

locking the Primary Node ID field within the VIOS DB; and

initiating a primary node commit process to assign the first node as the primary node.

24. The data processing system of claim 23, wherein the function of initiating the primary node commit process comprises:

checking whether an elector node ID is set in the VIOS DB;

in response to the elector node ID not being set, setting the elector node ID to the unique ID of the first node;

sending a commit signal to the VIOS DB to update the primary node ID field with the unique ID of the first node;

unlocking the Primary Node ID field following the update of primary node ID field with the unique ID of the first node; and

issuing a notification message to all other nodes in the cluster to inform the other nodes that a new primary node has been elected.

25. The data processing system of claim 23, said utility further executes to cause the first VIOS to perform the functions of:

determining that performance of a primary node function/action is required;

issuing a query to the VIOS DB to determine whether the first node is still the primary node by checking a current value of the primary node ID field;

performing the primary node function only if the first node is still the primary node;

in response to the first node no longer being the primary node, determining whether another node has committed as the primary node; and

in response to another node having committed as the primary node;

retrieving the primary node ID from the primary node ID field; and

updating a node local copy of the primary node ID within the first node.

26. The data processing system of claim 23, said utility further executes to cause the first VIOS to perform the functions of:

determining that the first node is no longer capable of acting as the primary node;

relinquishing the primary node responsibility by updating the Primary Node ID in the VIOS DB to a null value; and

retrieving a list of active nodes in the VIOS cluster;

discovering and notifying a next candidate in the list of active nodes, wherein the notifying includes one or more or a directly messaging the next candidate and broadcasting a primary node election message so that the other nodes that meet one or more primary node criteria can initiate the election process;

in response to detecting that a next node is elected as the primary node;

stopping the election timer; and

updating a local primary node ID that is maintained in local storage; and

in response to none of the active nodes being able to be elected as the primary node, issuing a notification of a failure to elect a primary node from among the active nodes within the cluster.

27. The data processing system of claim 23, wherein:

different primary nodes perform different functions and the broadcasting comprises indentifying, within the broadcasted message, information about what primary role is being relinquished;

the primary node is a role specific VIOS operating as one of a primary notification node and a primary clean-up node and is configured to provide one or more services to other VIOSes within the VIOS cluster, the one or more services comprising;

posting life cycle events;

posting alerts;

journaling alerts; and

providing database and general clean up; and

each role is provided by at least one primary node and multiple roles can be provided by a single primary node.