CN105306545A - Failover method and system for external service node of cluster - Google Patents

Abstract

The invention discloses a failover method and system for an external service node of a cluster. The method comprises the following steps: allocating an intranet address, a node serial number and a priority level for each node in the cluster; sending broadcast to the intranet addresses of other nodes excluding the current node through an intranet by each node; wherein, the broadcast by each non-external service node comprises a node serial number per se, the broadcast by the external service node comprises the node serial number per se and external service identification information; determining node serial number information not receiving the broadcast by each node according to the received node serial number information; selecting nodes corresponding to the same node serial number information not receiving the broadcast in all nodes as failure nodes; and if the external service node is present in the failure nodes, selecting a node having the highest priority level in effective nodes to serve as the external service node. The method and the system can be used for achieving reasonable and effective failover in the case of failure of the external service node.

Description

A method of cluster node failure to take over the foreign service and system

FIELD

[0001] The present invention relates to computers, and more particularly relates to a method and system for external services cluster node failure takeover.

Background technique

[0002] - like clusters usually only be switched between the external services of a few nodes and configuration changes very troublesome time, when the master node cluster services outside face of failure, can only enable has been set several a spare node to take over external services. And the need for master node and a few spare nodes are arranged ahead of time, more time to configure the backup node will be very troublesome. If you want to add a new spare node or delete the old standby node, modify the configuration is very complex, so the child would seriously reduce the scalability of the cluster. Furthermore this method to configure multiple backup node to take over when the confusion prone procedure.

[0003] Therefore, how fast, simple to implement cluster node fails to take over the Foreign Service, it is skilled in the art to be solved technical problems.

SUMMARY

[0004] The present invention is a method and system for providing a cluster node failure to take over the foreign service, when the method and system capable of external services node failure of rational and effective takeover, was greatly reduced when adding or removing cluster nodes modify the configuration of the time-consuming, can be prevented due to the modification of human errors caused by operation of the complex.

[0005] To solve the above problems, the present invention provides a method of cluster node failure takeover external services, to assign network addresses, the node number for each node in the cluster and priority, further comprising:

[0006] Each node sends a broadcast to the other nodes of the network address other than the present node through the network; wherein each non-broadcast transmitted external service node comprises its own node number, node sends a broadcast foreign service itself comprises the node number and external service identification information;

[0007] Each node does not receive the node number information is broadcast to the node number information received, according to the determined;

[0008] Select all nodes in the node number does not receive the broadcast information corresponding to the node representing the same section as invalid.

[0009] If the presence of the foreign service node failure node, the node with the highest priority is selected as a valid node from the external service node.

[0010] wherein, further comprising:

When [0011] If the external service node receives the broadcast, there is a high ratio of the external service node priority node, and selecting from the highest priority node is higher than the priority of the external node serving node;

[0012] The external service node sends to the node with the highest priority takes over the request, and stop the foreign service;

The [0013] highest priority node, after receiving the takeover request to take over the external services.

[0014] wherein each node sends the network addresses of other nodes other than this node within the network by broadcasting comprising:

[0015] The table corresponding to each node of the network addresses, sending a broadcast to the remaining network addresses of each node through the network.

[0016] wherein, each node receives the node number information, determines the node number information is not received broadcast includes:

[0017] Each node according to the node number information is received within the preset time, determining the node number is not received broadcast

ί project information.

[0018] wherein, further comprising:

[0019] The basic configuration of the timing information of each node in the cluster, the node number, and updates the priority.

[0020] A service node outside the cluster of the present invention provides a system failure takeover, comprising:

[0021] setting module for each node in the cluster assigned network address, the node number and the priority;

[0022] broadcasting module, for each node sends a broadcast to the other nodes of the network address other than the present node through the network; wherein each non-broadcast transmitted external service node comprises its own node number, the external service node send broadcast includes its own node number and external service identification information;

[0023] determination means for each node fails to receive broadcast information destination node ID ί node number information is received, determined in accordance with;

[0024] selection module for selecting all of the nodes in the same node number does not receive the broadcast information corresponding to a node as a failed node;

When [0025] The takeover means for the presence of foreign service node if the node fails, the node with the highest priority is selected as the active node from the external service node.

[0026] wherein, further comprising:

[0027] The comparison module for broadcasting if the received external service node, a node when there is a high priority than the external service node, and selecting the highest priority than the external service node from a higher priority node node;

[0028] Alternatively means for taking over the foreign service node sends a request to the node with the highest priority, and stop foreign service; the highest priority node, after receiving the takeover request to take over the external services.

[0029] wherein the determining module comprises:

[0030] The table corresponding to each node of the network addresses, sending a broadcast to the remaining network addresses of each node through the network.

[0031] wherein said selection module comprising:

[0032] Each node according to the node number information is received within the preset time, determining the node number is not received broadcast

ί project information.

[0033] wherein, further comprising:

[0034] The updating module, for timing the basic configuration information of each node in the cluster, the node number, and updates the priority.

[0035] The method and system of the present invention, the cluster nodes external services provided by the failure to take over, within each node within the cluster distribution network address, the node number and the priority, further comprising: by each node in the network nodes in addition to those the network address of other nodes outside the broadcast transmission; wherein each non-broadcast transmitted external service node comprises its own node number, node sends a broadcast external services include its own node number and identification information of external services; each node according to the received to the node number information, determines the node number information is not received broadcast; select all the nodes in the same node ID is not received broadcast information corresponding to the relative node as the failed node; if the failure nodes present in the external service node, from the highest priority node selected as the active node in the foreign service node;

[0036] Since the process by advancing the nodes configured for each node is assigned a network address, the node number and the priority, each node has its own number and priority, only the corresponding parameters delete and add modified, so this way when, after extending a need to add or operation will be very simple when you delete a node; and each node to confirm the identity of the form of broadcast, and whether the failure can be quickly and accurately in the Foreign service when node failure of rational and effective takeover. And no human interference, can prevent human errors; when the method and system capable of external services node failure of rational and effective takeover, have greatly reduce the cost of modifying the configuration when adding or removing cluster nodes time, can be prevented modifying operation caused due to the complexity of human errors.

BRIEF DESCRIPTION

[0037] In order to more clearly illustrate the technical solutions in the embodiments or the prior art embodiment of the present invention, briefly introduced hereinafter, embodiments are described below in the accompanying drawings or described in the prior art needed to be used in describing the embodiments the drawings are only examples of the present invention, those of ordinary skill in the art is concerned, without creative efforts, can derive other drawings from the accompanying drawings provided.

[0038] FIG. 1 is a flowchart of a method of external service node cluster failure takeover embodiment provided embodiment of the invention;

[0039] Figure 2 a block diagram of the external service node cluster failure takeover embodiment provided a system embodiment of the invention.

detailed description

[0040] The core of the invention is a method and system for providing a cluster node failure to take over the foreign service, when the method and system capable of external services node failure of rational and effective takeover, was greatly reduced when adding or removing cluster nodes modify the configuration of the time-consuming, can be prevented due to the modification of human errors caused by operation of the complex.

[0041] In order that the invention object, technical solutions, and advantages of the embodiments more clearly, the following the present invention in the accompanying drawings, technical solutions of embodiments of the present invention are clearly and completely described, obviously, the described the embodiment is an embodiment of the present invention is a part, but not all embodiments. Based on the embodiments of the present invention, all other embodiments of ordinary skill in the art without any creative effort shall fall within the scope of the present invention.

[0042] Please refer to FIG 1, FIG. 1 is a flowchart of a method embodiment of the cluster nodes external services provided by the failure to take over the embodiment; This method first requires each node within the cluster distribution network address, the node number and the priority provided, wherein the basic configuration of each node are consistent; after setting the method may comprise:

[0043] slOO, each node sends a broadcast to the other nodes of the network address other than the present node through the network; wherein each non-broadcast transmitted external service node comprises its own node number, the node sends a broadcast external services including its node number and external service identification information;

[0044] The destination node sends a broadcast is to have each node of the circumstances precluding other nodes outside of itself. Therefore, the contents of each node sends its own information may include, for example, contains the number of the node itself, if the need for the foreign service node also includes Foreign node identification information indicating own identities; information on other nodes dispensable. For example, the priority information may be transmitted through broadcast, may be broadcast to other nodes according to the obtained node number, and then stored in the lookup priority list in each node according to the node number. Here are just mentioned two ways.

[0045] sllO, each node does not receive the node number information is broadcast to the node number information received, according to the determined;

[0046] s 120, selected the same for all nodes in the node number information is not received corresponding to the broadcast node as the failed node;

[0047] wherein the number information by comparing the respective nodes, the failed node to obtain number information, i.e. determine the failure nodes; under normal circumstances all nodes, each node should receive the ID information of other nodes, i.e., plus on their number of all node information; when there is node failure, the node does not transmit the broadcast, therefore, a normal node may be received by the node number of all number information in a numbered list for comparison, identify missing number information, i.e., determines the node number is not received information broadcast. Node number information broadcast is not received all of the normal node may aggregate obtained, making comparisons, it is determined that all nodes have not received the same broadcast information corresponding to the node number of the node as the failed node.

[0048] s 130, if the presence of the foreign service node failure node, the node with the highest priority is selected as a valid node from the external service node.

[0049] When there are foreign service node node failure, you need to select the highest priority of all active nodes in the node to take over as the Foreign Service Node. Wherein the priority of each node may be set and modified by the actual situation. By setting priorities may make foreign service node to maintain an optimal state of active nodes.

[0050] Based on the above technical solutions, the external service node cluster method provided in embodiments of the present invention, the failure to take over, the process by advancing the nodes configured for each node is assigned a network address, the node number and the priority, each node It has its own number and priority, and only needs to be deleted when adding the corresponding parameter modification, so this way when after extended operations in to add or delete nodes when needed will be very simple; and each node confirm the identity of the form of broadcast, and can be quickly and accurately whether the failure to make reasonable and effective to take over when the foreign service node failure. And no human interference, can prevent human errors; when the method and system capable of external services node failure of rational and effective takeover, have greatly reduce the cost of modifying the configuration when adding or removing cluster nodes time, can be prevented modifying operation caused due to the complexity of human errors.

[0051] Based on the above technical solution, the method may further comprise:

When [0052] If the external service node receives the broadcast, there is a high ratio of the external service node priority node, and selecting from the highest priority node is higher than the priority of the external node serving node;

[0053] The external service node sends to the node with the highest priority takes over the request, and stop the foreign service;

The [0054] highest priority node, after receiving the takeover request to take over the external services.

[0055] wherein, the node number of the law can be set according to usage, for example, all the nodes are arranged in the intake butoxy left foot from still to low priority order.

[0056] wherein the system in real-time, real-time to take over the case according to the priority of active nodes and external service node, a node always maintained the highest priority of all active nodes for the foreign service node. Include a specific process may be as follows:

[0057] By using the network segments in the distribution network, preparing a fixed network address, node number and the priority correspondence table on all nodes, all nodes just give assigned node number, node itself automatically so that the configuration according to the correspondence table network address. Then by using the internal network broadcast, all nodes including the network time to send to the corresponding table containing all addresses within the network broadcasts its own node number, the node sends a broadcast in which external services will add some extra special message that It is a node external service provider.

[0058] If all the nodes in the process are monitored by a node can not receive the broadcast, then depending on the failed node is a node; otherwise depending on the node is a valid node. If a node failure to provide services, from the node with the highest priority valid node started to provide selected external services. If provide external services node receives a broadcast from the node of higher priority than its own, it stops the external service, and transmits the request to take over the node, the node when the higher priority node receives the takeover request, will be outside to take over the service, and then began to provide external services.

[0059] Based on each of the above technical solution, optionally, each node sends the network addresses of other nodes other than this node may broadcast by the network comprising:

[0060] The table corresponding to each node of the network addresses, sending a broadcast to the remaining network addresses of each node through the network.

[0061] wherein, the network may be a list of URLs of all nodes, and may be stored in each node; each node correspondence table according to the network address, through the network to the rest of each of the remaining nodes other than this node network broadcast address to send.

[0062] where time may be sent at any transmission, may be transmitted in accordance with a predetermined time interval. Several models exist or can be set by a user. That is set according to the actual situation.

[0063] Based on each of the above technical solution, optionally, each node of the node number does not receive the broadcast information may comprise information received by the node number, determined according to:

[0064] Each node according to the node number information is received within the preset time, determining the node number is not received broadcast

ί project information.

[0065] Among them, the node sends a broadcast is not sent all the time, therefore, in determining the node is not valid, there must be a time limit, otherwise it will cause confusion. Thus, each node according to the node number information is received within the preset time, determining the node number information is not received broadcast. Here the preset time may be a node of a preset radio transmission time interval, may be smaller than the time interval.

[0066] Based on the above any particular aspect, the method may further comprise:

[0067] The basic configuration of the timing information of each node in the cluster, the node number, and updates the priority.

[0068] wherein, based on the use of improved technology, or a device, the user may at any time or periodically basic configuration information of each node in the cluster, the node number, and updates the priority. So that the node information is always the best information, so as to ensure the ability of foreign service node.

[0069] The specific process is:

Under the total number of [0070] node is not fixed, and the need to build a cluster to provide high availability of external services, we need to be consistent on the basis of the configuration of each node, and then number each node, each assigned to take over the Foreign Service priority, the highest priority node obtains permission to external services. All nodes broadcast to other nodes, if a node does not receive all the broadcast of a node within a certain time, depending on the node failure. If the foreign service node failure, and on the remaining active node with the highest priority to take control of external services; if a higher priority node appeared, the current provision of external services node will stop the foreign service, and to a higher priority the node sends a request to take over the higher priority node after receiving the request to take over the Foreign service.

[0071] Based on the above technical solution, a method of external cluster node provided service failure takeover of the present invention, changing the cluster switch for external services usually only a few nodes between the prior art, and configuration changes when very troublesome. This method can quickly configure the number of nodes in the cluster nodes are not fixed, so that the foreign service node failure when reasonably and effectively over, and add or delete a simple operation required when a node fast, was reduced significantly add modify the configuration or time consuming to remove a cluster node, human errors can be prevented due to changes in the operation due to complex. And node configuration is simple, just need a common configuration with node number can be set independently, all nodes have the ability to take over the external services, all nodes have the ability to take over the external services.

[0072] Example embodiments provide a method of cluster node failure takes over the foreign service of the present invention, when the external services may be capable of node failure by the method described above takes over reasonable and effective.

[0073] The following system cluster foreign service node provided failure takeover is described embodiment of the present invention, a method cluster external service node cluster external services nodes described below failure to take over the system described above failure takeover may correspond to cross-reference.

[0074] Please refer to FIG 2, a block diagram of FIG. 2 cluster external services provided by the node embodiment of the failure to take over the system embodiment of the invention; The system may comprise:

[0075] The setting module 100, is used to assign each node in the cluster network address, the node number and the priority;

[0076] The broadcasting module 200, for each node sends a broadcast to the other nodes of the network address other than the present node through the network; wherein each non-broadcast transmitted external service node comprises its own node number, external services including broadcasting node sends its node number and external service identification information;

[0077] module 300 determines, for each node according to the node number information received, determining the node number is not received broadcast information;

[0078] The selection module 400, configured to select the node number information is not received corresponding to the broadcast with respect to all the nodes in the same node as the failed node;

[0079] module 500 to take over, if the presence of a foreign service node failure node, the node with the highest priority is selected as a valid node from the external service node.

[0080] Optionally, the system may further comprise:

[0081] The comparison module for broadcasting if the received external service node, a node when there is a high priority than the external service node, and selecting the highest priority than the external service node from a higher priority node node;

[0082] Alternatively means for taking over the foreign service node sends a request to the node with the highest priority, and stop foreign service; the highest priority node, after receiving the takeover request to take over the external services.

[0083] Optionally, the determination module 300 may include:

[0084] The table corresponding to each node of the network addresses, sending a broadcast to the remaining network addresses of each node through the network.

[0085] Optionally, the selecting module 400 may include:

[0086] Each node according to the node number information is received within the preset time, determining the node number is not received broadcast

ί project information.

[0087] Based on the above technical solution any, the system may further comprise:

[0088] updating module, for timing the basic configuration information of each node in the cluster, the node number, and updates the priority.

[0089] In various embodiments, the specification progressive manner described, is different from the embodiment and the other embodiments each of which emphasizes embodiment, the same or similar portions between the various embodiments refer to each other. For the disclosed embodiment of the apparatus embodiment, since it corresponds to the method disclosed embodiments, the description is relatively simple, see Methods of the correlation can be described.

[0090] professionals may further appreciate that the various means disclosed herein and algorithm steps described exemplary embodiments, by electronic hardware, computer software, or a combination thereof. In order to clearly illustrate the hardware and software interchangeability, in the above description, according to functions generally described compositions and steps of the examples. Whether these functions are performed by hardware or software depends upon the particular application and design constraints of the technical solutions. Professional technical staff may use different methods for each specific application to implement the described functionality, but such implementation should not be considered outside the scope of the present invention.

[0091] The steps of a method or algorithm described in the embodiments disclosed herein may be implemented in hardware, or a combination thereof, in a software module executed by a processor implemented directly. A software module may be placed in a random access memory (RAM), a memory, a read only memory (ROM), electrically programmable R0M, electrically erasable programmable R0M, registers, a hard disk, a removable disk, CD-ROM, or within the technical field known any other form of storage medium.

[0092] The foregoing method and system of the present invention, the cluster nodes external services provided by the failed takeover described in detail. Herein through specific examples and embodiments of the principles of the present invention are set forth in the above described embodiments are only used to help understand the method and core idea of ​​the present invention. It should be noted that those of ordinary skill in the art, in the present invention without departing from the principles of the premise, may also be a number of modifications and improvements of the present invention, and these improvements and modifications also fall within the scope of the claims of the invention.

Claims (10)

1. 1. A method of cluster node failure takeover external services, wherein for each node in the cluster assigned network address, the node number and the priority, further comprising: each node in addition to the network node by the network address of the other node broadcasts transmitted; wherein each non-transmitted on the broadcast Bu Xi service node includes its own node number, node sends a broadcast external services include its own node number and identification information of external services; each node according to the received to the node number information, determines the node number information is not received broadcast; select all the nodes in the same node ID is not received broadcast information corresponding to the relative node as the failed node; if the failure nodes present in the external service node, from the highest priority node selected as the active node in the foreign service node.
2. If the external service node receives the broadcast, there is a high ratio of the external service node priority node from the priority than the external service node: 2. The method according to claim 1, characterized in that, further comprising high-level nodes selected node with the highest priority; the external service node sends a takeover request to the node with the highest priority, and stop foreign service; the highest priority node, after receiving the takeover request to take over the external service.
3. 3. The method according to claim 1, wherein each said node includes transmitting a broadcast to the other nodes of the network address other than the present node through the network: the node corresponding to each network address in accordance with table, sending a broadcast to the remaining network addresses of each node through the network.
4. 4. The method according to claim 3, wherein said each node does not receive the node number information of the broadcast information including the received node number, determined according to: for each node according to the received node within a preset time number information, determines the node number information is not received broadcast.
5. 5. A method according to any one of claim 1 to claim 4, characterized in that, further comprising: a timing of the basic configuration information of each node in the cluster, the node number, and updates the priority.
6. A cluster node failure takeover external service system, characterized by comprising: a setting module, configured to assign each node in the cluster network address, the node number and the priority; broadcasting module, for each node the network sends a broadcast to the other nodes of the network address other than the present node; wherein each non-broadcast transmitted external service node comprises its own node number, node sends a broadcast external services comprise own node identification numbers and external services information; determining module, for each node according to the node number information received, determining the node number information is not received broadcast; selection module for selecting all the nodes have not received the broadcast information of the same node numbers of nodes corresponding as failure nodes; taking over means for the presence of foreign service node if the node failure, a node is selected as the highest priority node from the active foreign service node.
7. 7. The system according to claim 6, characterized in that, further comprising: a comparison module configured if foreign service node receives the broadcast, there is a high ratio of the external service node priority node, from the ratio said high priority node in the foreign service node selected node with the highest priority; replacing module for taking over the foreign service node sends a request to the node with the highest priority, and stop foreign service; the highest priority node after receiving the takeover request to take over the external services.
8. 8. The system according to claim 6, wherein the determining module comprises: each node of the network address according to the correspondence table, a broadcast transmission to the remaining network addresses of each node through the network.
9. 9. The system according to claim 8, wherein said selection module comprising: each node according to the node number information is received within the preset time, determining the node number information is not received broadcast.
10. 10. The system according to any one of claims 6-9, characterized in that, further comprising: an updating module, for timing the basic configuration information of each node in the cluster, the node number and the priority of the update.