Failover scheme with service-based segregation

• US 8,209,405 B1
• Filed: 06/12/2006
• Issued: 06/26/2012
• Est. Priority Date: 06/12/2006
• Status: Expired due to Fees

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: Juniper Networks Incorporated
• Inventors: Trivedi, Atul Narendra
• Primary Examiner(s): Serrao, Ranodhi
• Application Number:
  US11/423,566
• Time in Patent Office: 2,206 Days
• Field of Search: 709/223, 714/26
• US Class Current: 709/223
• CPC Class Codes: G06F 11/2023 Failover techniques
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 2221/2149 Restricted operating environment
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A system provides a set of services. The system includes nodes that are in communication with each other. The system segregates the services into at least first and second groups of services, assigns the first group of services to a first set of the nodes, and assigns the second group of services to a second set of nodes. The first set of nodes provides the first group of services, and the second set of nodes provides the second group of services.

Claims

1. A failover system, comprising:

a first cluster of nodes to provide a first group of services, nodes of the first cluster of nodes being implemented at least partially in hardware, a node, of the first cluster of nodes, to function as a master for the first cluster of nodes and at least one other node, of the first cluster of nodes, to function as a standby for the first cluster of nodes; and

a second cluster of nodes, connected to the first cluster of nodes, to provide a second group of services, nodes of the second cluster of nodes being implemented at least partially in hardware, a node, of the second cluster of nodes, to function as a master for the second cluster of nodes and at least one other node, of the second cluster of nodes, to function as a standby for the second cluster of nodes,

the nodes of the first cluster of nodes and the nodes of the second cluster of nodes to exchange information,

at least one node of the first cluster of nodes to;

process the information received from other nodes of the first cluster of nodes,

determine, based on the information received from the nodes of the second cluster of nodes, that the nodes of the second cluster of nodes do not provide the first group of services, and

disregard the information received from the nodes of the second cluster of nodes, based on determining that the nodes of the second cluster of nodes do not provide the first group of services,

the first group of services and the second group of services being selected from an entire set of services associated with the failover system,

the first group of services including a first list of services and the second group of services including a second list of services,

the first list of services being exclusive relative to the second list of services,

the master for the first cluster of nodes providing the first group of services without providing the second group of services, and

the master for the second cluster of nodes providing the second group of services without providing the first group of services.

2. The system of claim 1, the entire set of services associated with the failover system being segregated into the first group of services, the second group of services, and a third group of services, services in the third of group of services being different than the first list of services and the second list of services,

the first group of services being assigned to the first cluster of nodes, and

the second group of services being assigned to the second cluster of nodes,

the third group of services being assigned to a third cluster of nodes.

3. The system of claim 1, one of the nodes of the first cluster of nodes to discover other ones of the nodes of the first cluster of nodes and the nodes of the second cluster of nodes.

4. The system of claim 3, the one of the nodes of the first cluster of nodes to broadcast data to dynamically discover the other nodes of the first cluster of nodes and the nodes of the second cluster of nodes.

5. The system of claim 4, the one of the nodes of the first cluster of nodes to broadcast the data to a particular port of each of the other nodes of the first cluster of nodes and the nodes of the second cluster of nodes.

6. The system of claim 3, the one of the nodes of the first cluster of nodes to:

receive data from the other nodes of the first cluster of nodes and the nodes of the second cluster of nodes, and

generate a list that identifies the other nodes of the first cluster of nodes and the nodes of the second cluster of nodes.

7. The system of claim 1, the exchanged information respectively including information regarding the first group of services provided by the nodes of the first cluster of nodes and the second group of services provided by the nodes of the second cluster of nodes.

8. The system of claim 7, one of the nodes of the first cluster of nodes to identify other ones of the nodes in the first cluster of nodes as part of the first cluster of nodes based on the exchanged information.

9. The system of claim 1, the nodes of the first cluster of nodes to identify the node to function as the master and the at least one other node to function as the standby.

10. The system of claim 1, the master for the first cluster of nodes to provide the first group of services for the first cluster of nodes when the master for the first cluster is active, and

the standby of the first cluster of nodes to provide the first group of services when the master of the first cluster of nodes is inactive.

11. The system of claim 1, the master for the first cluster of nodes to provide the first group of services for the first cluster of nodes when the master for the first cluster of nodes is active,

the standby of the first cluster of nodes to provide the first group of services for the first cluster of nodes when the master of the first cluster of nodes becomes inactive.

12. The system of claim 11, the master of the second cluster of nodes to provide the second group of services for the second cluster of nodes independent of the master of the first cluster of nodes becoming inactive.

13. A method, comprising:

configuring a plurality of nodes in a failover system, the plurality of nodes providing a plurality of services;

segregating the plurality of services into at least a first group of services and a second groups of services;

forming clusters of nodes based on the plurality of nodes,

each of the clusters of nodes providing services associated with the first group of services or the second group of services;

providing, by one of the clusters of nodes, the first group of services,

nodes of the one of the clusters of nodes being implemented at least partially in hardware, a node, of the one of the cluster of nodes, functioning as a master for the one of the cluster of nodes;

providing, by a different one of the clusters of nodes, the second group of services,

nodes of the different one of the clusters of nodes being implemented at least partially in hardware, a node, of the different one of the cluster of nodes, functioning as a master for the different one of the clusters of nodes, and

the nodes of the one of the clusters of nodes and the nodes of the different one of the clusters of nodes to exchange information;

processing, by at least one node of the one of the cluster of nodes, the information received from other nodes of the first cluster of nodes; and

ignoring, by the at least one node, the information received from one or more nodes of the different one of the cluster of nodes, based on the information, received from the one or more nodes, indicating that the one or more nodes do not provide the first group of services,

the first group of services including a first list of services and the second group of services including a second list of services,

the first list of services being exclusive relative to the second list of services,

the master for the one of the clusters of nodes providing the first group of services without providing the second group of services, and

the master for the different one of the clusters of nodes providing the second group of services without providing the first group of services.

14. The method of claim 13, further comprising:

identifying, for each of the clusters of nodes, one of the nodes of the cluster of nodes to function as a master for the cluster of nodes and at least one other one of the nodes of the cluster of nodes to function as a standby for the cluster of nodes.

15. The method of claim 14, further comprising:

detecting that the master of the one of the clusters of nodes, providing the first group of services, becomes inactive; and

providing, by a standby of the one of the clusters, the first group of services when the master becomes inactive.

16. The method of claim 15, further comprising:

providing, by the master of the different one of the clusters of nodes, the second group of services prior to the master of the one of the clusters of nodes becoming inactive and after the master of the one of the clusters of nodes becomes inactive.

17. The method of claim 13, further comprising:

discovering, by at least one node of the clusters of nodes, other nodes of each of the clusters of nodes.

18. The method of claim 17, where discovering the other nodes includes broadcasting, by the at least one of the nodes, data to dynamically discover the other nodes.

19. The method of claim 18, where broadcasting the data includes broadcasting the data to a particular port of the other nodes.

20. The method of claim 17, where discovering the other nodes includes:

receiving data from the other nodes; and

generating a list that identifies the other nodes independent of a respective one of the clusters of nodes with which the other nodes are associated.

21. The method of claim 13, where forming the clusters of nodes includes exchanging information regarding services provided by each node of the clusters of nodes.

22. The method of claim 21, where forming the clusters of nodes further includes identifying, by at least one node of each cluster of nodes of the clusters of nodes, other nodes in the cluster of node as being part of the cluster of nodes based on the exchanged information.

23. A system, comprising:

a first cluster of nodes to provide a first group of services, a node of the first cluster of nodes to function as a master for the first cluster of nodes and at least one other node of the first cluster of nodes to function as a standby for the first cluster of nodes;

a second cluster of nodes, connected to the first cluster of nodes, to provide a second group of services, a node of the second cluster of nodes to function as a master for the second cluster of nodes,

nodes of the first cluster of nodes and nodes of the second cluster of nodes being implemented at least partially in hardware,

nodes of the first cluster of nodes and nodes of the second cluster of nodes to exchange information; and

one or more nodes of the first cluster of nodes to;

process the information received from other nodes of the first cluster of nodes, and

disregard the information received from the nodes of the second cluster of nodes, based on the information, received from the nodes of the second cluster of nodes, indicating that the nodes of the second cluster of nodes do not provide the first group of services,

the first group of services and the second group of services being selected from an entire set of services associated with the system,

the first group of services, provided by the first cluster of nodes, being different than the second group of services, provided by the second cluster of nodes,

the master for the first cluster of nodes providing the first group of services without providing the second group of services, and

the master for the second cluster of nodes providing the second group of services without providing the first group of services.

24. The system of claim 23, where the node, of the first cluster of nodes, is to receive information, from the at least one other node of the first cluster of nodes, indicating that the at least one other node provides the first group of services, and

where the node and the at least one other node are included in the first cluster of nodes based on the information received from the at least one other node.

25. The system of claim 23, where the node, of the second cluster of nodes, is to:

broadcast data to discover one or more additional nodes included in the second cluster of nodes,

receive, in response to the broadcast data, information from another node, the received information identifying the other node as providing the second group of services, and

determine that the other node is included in the second cluster based on the received information from the other node.

26. The system of claim 23, where the master for the second cluster of nodes provides the second group of services prior to the master for the first cluster of nodes becoming inactive and after the master for the first cluster of nodes becoming inactive.