CN105407087B - Message method and device - Google Patents
Message method and device Download PDFInfo
- Publication number
- CN105407087B CN105407087B CN201510714154.1A CN201510714154A CN105407087B CN 105407087 B CN105407087 B CN 105407087B CN 201510714154 A CN201510714154 A CN 201510714154A CN 105407087 B CN105407087 B CN 105407087B
- Authority
- CN
- China
- Prior art keywords
- node
- host
- group message
- master
- leave group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/10—Architectures or entities
- H04L65/1013—Network architectures, gateways, control or user entities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/1066—Session management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- General Business, Economics & Management (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Mobile Radio Communication Systems (AREA)
- Computer And Data Communications (AREA)
Abstract
The present invention proposes a kind of message method and device, and in the cluster applied to deployment centralized management model, which includes: that the host node in resource-area has monitored that node leaves；If the network between host node and control node is unreachable, or host node does not receive the confirmation message of control node, and host node reaches preset threshold to the number of control node sending node leave group message, then host node is to adjacent non-master sending node leave group message, so that network of the adjacent non-master between control node up to when, node leave group message is sent to control node.The present invention can be in the host node in resource-area and the unreachable network between control node, control node is sent by node leave group message most possibly, to which control node can notify other to have the node of slack resources to carry out fault recovery to the resource left on node, and then the high availability of cluster can be improved.
Description
Technical field
The present invention relates to field of communication technology more particularly to a kind of message method and devices.
Background technique
Cluster high availability (High Availability；Hereinafter referred to as: HA) refer to by shortening as far as possible because of daily maintenance
Downtime caused by the system crash (unplanned) of (plan) and burst is operated, to improve the availability of system and application.
The cluster HA software of mainstream uses monitoring (corosync) software and resource management (pacemaker) software at present
Combination.Monitoring of software realizes the function of node relationships management and communication service, and resource management software realizes the function of resource management.
The cluster for disposing centralized management model can be divided into resource-area, network and control three parts of (controller) node,
In resource-area, each host (host) is required to deployment monitoring and resource management software.Resource management selection in resource-area
Host node (master node) of one node as resource-area.If master node monitors a certain node in resource-area
It leaves, is then responsible for from master node to controller node sending node leave group message.Controller node receives section
After point leave group message, after finding the host that other there are slack resources, idle money has been sent by above-mentioned node leave group message
The host in source carries out fault recovery.
But when the host node of resource-area monitors that some node leaves, and host node and controller node it
Between network it is unreachable, and can not carry out in resource-area there are the resource when host of slack resources, left on node therefore
Barrier restores.
Summary of the invention
The purpose of the present invention is intended to solve at least some of the technical problems in related technologies.
For this purpose, the first purpose of this invention is to propose a kind of message method.This method can be in resource-area
Host node and control node between network it is unreachable when, send control node for node leave group message most possibly,
To which control node can notify other to have the node of slack resources to carry out fault recovery, Jin Erke to the resource left on node
To improve the high availability of cluster.
Second object of the present invention is to propose a kind of message sending device.
To achieve the goals above, the message method of first aspect present invention embodiment is applied to deployment concentrate tube
In the cluster for controlling model, it includes at least two nodes in the resource-area that the cluster, which includes resource-area and control node, described
At least two nodes and the control node pass through network connection, which comprises the host node monitoring in the resource-area
There is node to leave into the resource-area；If the network between the host node and the control node is unreachable, Huo Zhesuo
It states host node and disappears to not receiving the control node after the control node sending node leave group message and left for the node
The confirmation message of breath, and the host node reaches default threshold to the number that the control node sends the node leave group message
It is worth, then adjacent non-master sending node leave group message of the host node to the host node, the node leave group message packet
Include the essential information for leaving node and the resource left on node so that the adjacent non-master of the host node with institute
State network between control node up to when, the node leave group message is sent to the control node.
In the message method of the embodiment of the present invention, the host node in resource-area monitors there is node in above-mentioned resource-area
After leaving, if the network between host node and control node is unreachable or host node sends to above-mentioned control node and saves
The confirmation message that above-mentioned control node is directed to above-mentioned node leave group message is not received after point leave group message, and host node is to control
The number of node sending node leave group message reaches preset threshold, then host node is sent to the adjacent non-master of above-mentioned host node
Node leave group message so that network of the adjacent non-master of above-mentioned host node between control node up to when, by node
Leave group message is sent to above-mentioned control node, so as to can not in the host node in resource-area and the network between control node
Up to when, control node is sent by node leave group message most possibly, so that control node notifies other to have slack resources
Node carries out fault recovery to the resource left on node, and then the high availability of cluster can be improved.
To achieve the goals above, the message method of second aspect of the present invention embodiment is applied to deployment concentrate tube
In the cluster for controlling model, it includes at least two nodes in the resource-area that the cluster, which includes resource-area and control node, described
At least two nodes and the control node pass through network connection, which comprises host node is adjacent in the resource-area
Non-master receives the node leave group message that the host node is sent, and the node leave group message includes leaving the basic letter of node
Breath and the resource left on node, the node leave group message is that the host node monitors there is node in the resource-area
It leaves, the network between the host node and the control node is unreachable or the host node is sent out to the control node
The confirmation message that the control node is directed to the node leave group message, and the main section are not received after sending node leave group message
What point was sent after reaching preset threshold to the number that the control node sends the node leave group message；If the main section
Network between the adjacent non-master and the control node of point is reachable, then the adjacent non-master of the host node will be described
Node leave group message is sent to the control node.
In the message method of the embodiment of the present invention, the adjacent non-master of host node receives host node hair in resource-area
After the node leave group message sent, if the network between the adjacent non-master and control node of above-mentioned host node is reachable,
Above-mentioned node leave group message is sent to above-mentioned control node by the adjacent non-master of host node, so as in resource-area
When network between host node and control node is unreachable, control node is sent by node leave group message most possibly, with
Just control node notifies other to have the node of slack resources to carry out fault recovery to the resource left on node, and then can be improved
The high availability of cluster.
To achieve the goals above, the message sending device of third aspect present invention embodiment is applied to deployment concentrate tube
In the cluster for controlling model, it includes at least two nodes in the resource-area that the cluster, which includes resource-area and control node, described
By network connection, the message sending device is set at least two node at least two nodes and the control node
Host node in, whether the message sending device includes: monitoring modular, have node to leave for monitoring in the resource-area；
Sending module, for when the monitoring module monitors have node to leave into the resource-area, if the host node and institute
It states that the network between control node is unreachable or the host node is not to receiving after the control node sending node leave group message
The confirmation message of the node leave group message is directed to the control node, and the host node is sent to the control node
The number of the node leave group message reaches preset threshold, then leaves and disappear to the adjacent non-master sending node of the host node
Breath, the node leave group message includes leaving the essential information and the resource left on node of node, so that the main section
Network of the adjacent non-master between the control node of point up to when, the node leave group message is sent to described
Control node.
In above-mentioned message sending device, after monitoring module monitors have node to leave into above-mentioned resource-area, if main section
Between point and control node network is unreachable or host node is not to receiving control after control node sending node leave group message
Node is directed to the confirmation message of above-mentioned node leave group message, and host node sends above-mentioned node and leave and disappears to above-mentioned control node
The number of breath reaches preset threshold, then adjacent non-master sending node leave group message from sending module to above-mentioned host node, with
Make network of the adjacent non-master of above-mentioned host node between control node up to when, node leave group message is sent to
Control node is stated, so as in the host node in resource-area and the unreachable network between control node, most possibly
Control node is sent by node leave group message, so that control node notifies other to have the node of slack resources to leaving on node
Resource carry out fault recovery, and then the high availability of cluster can be improved.
To achieve the goals above, the message sending device of fourth aspect present invention embodiment is applied to deployment concentrate tube
In the cluster for controlling model, it includes at least two nodes in the resource-area that the cluster, which includes resource-area and control node, described
By network connection, the message sending device is set at least two node at least two nodes and the control node
In in the non-master adjacent with host node, the message sending device includes: receiving module, for receiving the host node hair
The node leave group message sent, the node leave group message include leaving the essential information and the money left on node of node
Source, the node leave group message are that the host node monitors have node to leave in the resource-area, the host node with it is described
Network between control node is unreachable or the host node is not to receiving after the control node sending node leave group message
The control node is directed to the confirmation message of the node leave group message, and the host node sends institute to the control node
The number for stating node leave group message reaches what preset threshold was sent later；Sending module, for working as the adjacent non-of the host node
Network between host node and the control node up to when, the node leave group message is sent to the control node.
In above-mentioned message sending device, after receiving module receives the node leave group message that host node is sent, if above-mentioned
Network between the adjacent non-master and control node of host node is reachable, then sending module sends above-mentioned node leave group message
To above-mentioned control node, so as to which in the host node in resource-area and the unreachable network between control node, maximum can
Can ground by node leave group message send control node, so that control node notifies other to have the node of slack resources to leaving section
Resource on point carries out fault recovery, and then the high availability of cluster can be improved.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is the flow chart of message method one embodiment of the present invention；
Fig. 2 is the flow chart of another embodiment of message method of the present invention；
Fig. 3 is the flow chart of message method further embodiment of the present invention；
Fig. 4 is the architecture diagram of cluster one embodiment of present invention deployment centralized management model；
Fig. 5 is the structural schematic diagram of message sending device one embodiment of the present invention；
Fig. 6 is the structural schematic diagram of another embodiment of message sending device of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not considered as limiting the invention.On the contrary, this
The embodiment of invention includes all changes fallen within the scope of the spiritual and intension of attached claims, modification and is equal
Object.
Fig. 1 is the flow chart of message method one embodiment of the present invention, and above-mentioned message method is applied to deployment
In the cluster of centralized management model, it includes at least two sections in above-mentioned resource-area that above-mentioned cluster, which includes resource-area and control node,
Point, above-mentioned at least two node and above-mentioned control node pass through network connection.Each of resource-area node is required to dispose
Monitoring and resource management software, the resource management software in resource-area select host node of the node as resource-area
(master node).
As shown in Figure 1, the message method may include:
Step 101, the host node in resource-area monitors have node to leave in above-mentioned resource-area.
Step 102, if the network between above-mentioned host node and above-mentioned control node is unreachable or above-mentioned host node to
The confirmation that above-mentioned control node is not received after above-mentioned control node sending node leave group message for above-mentioned node leave group message disappears
It ceases, and above-mentioned host node reaches preset threshold to the number that above-mentioned control node sends above-mentioned node leave group message, then main section
For point to the adjacent non-master sending node leave group message of above-mentioned host node, above-mentioned node leave group message includes leaving the base of node
This information and the above-mentioned resource left on node, so that net of the adjacent non-master of host node between above-mentioned control node
Network up to when, above-mentioned node leave group message is sent to control node.
That is, when the host node in resource-area monitors has node to leave in above-mentioned resource-area, if above-mentioned master
Network between node and above-mentioned control node is unreachable, and at this moment host node is without normal direction control node sending node leave group message；
Or host node is left for above-mentioned node to not receiving above-mentioned control node after above-mentioned control node sending node leave group message
The confirmation message of message, and above-mentioned host node reaches default to the number that above-mentioned control node sends above-mentioned node leave group message
Threshold value, under both situations, node leave group message can be sent to the adjacent non-master of host node by host node, if main section
Network between the adjacent non-master and above-mentioned control node of point is reachable, then the adjacent non-master of host node can will be above-mentioned
Node leave group message is sent to control node, and if network between the adjacent non-master and above-mentioned control node of host node
Also unreachable, then the adjacent non-master of host node can continue above-mentioned node leave group message to be sent to adjacent next non-master
Node, and so on, until the network between the non-master and control node for receiving node leave group message is reachable, then with control
The node leave group message that the reachable non-master of network between node processed can will receive is sent to control node.
Wherein, above-mentioned control node is not received after host node is to above-mentioned control node sending node leave group message for upper
When stating the confirmation message of node leave group message, illustrate that above-mentioned node leave group message sends failure.At this time host node need again to
Control node sending node leave group message, at this moment, if host node reaches to the number of control node sending node leave group message
Above-mentioned node leave group message can be sent to the adjacent non-master of host node by preset threshold, host node.
Above-mentioned preset threshold can in specific implementation, according to the realization sets itselfs such as demand and/or system performance, this reality
It applies example to be not construed as limiting the size of above-mentioned preset threshold, for example, above-mentioned preset threshold can be 5.
Network in the worst case, if the network in above-mentioned resource-area between all nodes and control node can not
It reaches, the above-mentioned node leave group message that above-mentioned host node issues finally is sent back to host node, that is to say, that when in above-mentioned resource-area
When network between all nodes and control node is unreachable, the node leave group message that host node issues is being had sent a circle
It is transferred back to host node again later.At this moment, after by scheduled duration, host node is again to the adjacent non-master of above-mentioned host node
Node sending node leave group message.Specifically, a timer can be set in host node, timer then after, host node
Retransmit node leave group message, it is ensured that have in resource-area the network between node and control node up to when, can be timely
Node leave group message is sent to control node.Wherein, the length of above-mentioned scheduled duration can sets itself in specific implementation,
The present embodiment is not construed as limiting this, for example, above-mentioned scheduled duration can be 10 seconds.
Reliable communication service is supported due to monitoring process in resource-area, and host node is to the adjacent non-of above-mentioned host node
Host node sending node leave group message can be with are as follows: above-mentioned node leave group message is sent to above-mentioned by the Resource Management process of host node
Above-mentioned node leave group message is sent to the adjacent of above-mentioned host node by the monitoring process of the monitoring process of host node, above-mentioned host node
The monitoring process of non-master.
In above-mentioned message method, the host node in resource-area is monitored after having node to leave in above-mentioned resource-area,
If the network between host node and control node is unreachable or host node is to above-mentioned control node sending node leave group message
The confirmation message that above-mentioned control node is directed to above-mentioned node leave group message is not received afterwards, and host node is sent out to above-mentioned control node
The number of the node leave group message is sent to reach preset threshold, then host node sends to the adjacent non-master of above-mentioned host node and saves
Point leave group message so that network of the adjacent non-master of above-mentioned host node between control node up to when, by node from
It opens message and is sent to above-mentioned control node, so as to unreachable in the host node in resource-area and the network between control node
When, control node is sent by node leave group message most possibly, so that control node notifies other to have the section of slack resources
Point carries out fault recovery to the resource left on node, and then the high availability of cluster can be improved.
Fig. 2 is the flow chart of another embodiment of message method of the present invention, and above-mentioned message method is applied to portion
In the cluster for affixing one's name to centralized management model, it includes at least two in above-mentioned resource-area that above-mentioned cluster, which includes resource-area and control node,
Node, above-mentioned at least two node and above-mentioned control node pass through network connection.Each of resource-area node is both needed to mian part
Administration monitors and resource management software, and the resource management software in resource-area selects host node of the node as resource-area
(master node).
As shown in Fig. 2, above-mentioned message method may include:
Step 201, the adjacent non-master of host node receives the node leave group message that host node is sent in above-mentioned resource-area,
Above-mentioned node leave group message includes leaving the essential information and the above-mentioned resource left on node of node, above-mentioned node leave group message
It is that host node monitors have node to leave in above-mentioned resource-area, the network between above-mentioned host node and control node is unreachable, or
Person's host node is not to receiving control node after above-mentioned control node sending node leave group message for above-mentioned node leave group message
Confirmation message, and above-mentioned host node to the number that above-mentioned control node sends above-mentioned node leave group message reach preset threshold it
It sends afterwards.
Support reliable communication service due to monitoring process in resource-area, step 201 can be with are as follows: host node it is adjacent
Monitoring process in non-master receives the node leave group message that the monitoring process of above-mentioned host node is sent.
Step 202, if the network between the adjacent non-master and above-mentioned control node of above-mentioned host node is reachable, on
Above-mentioned node leave group message is sent to above-mentioned control node by the adjacent non-master for stating host node.
Specifically, the adjacent non-master of above-mentioned host node above-mentioned node leave group message is sent to above-mentioned control node can
With are as follows: above-mentioned node leave group message is sent to the phase of above-mentioned host node by the monitoring process of the adjacent non-master of above-mentioned host node
The Resource Management process of the Resource Management process of adjacent non-master, the adjacent non-master of above-mentioned host node leaves above-mentioned node
Message is sent to above-mentioned control node.
Fig. 3 is the flow chart of message method further embodiment of the present invention, as shown in figure 3, after step 201, also
May include:
Step 301, above-mentioned if the network between the adjacent non-master and control node of above-mentioned host node is unreachable
Above-mentioned node leave group message is sent to adjacent next non-master by the adjacent non-master of host node, and so on, until
The network received between the non-master and above-mentioned control node of above-mentioned node leave group message is reachable, then with above-mentioned control node
Between the above-mentioned node leave group message that will receive of the reachable non-master of network be sent to above-mentioned control node.
Specifically, what the monitoring process that the monitoring process in the adjacent non-master of host node receives above-mentioned host node was sent
After node leave group message, node leave group message is sent to the Resource Management process in the adjacent non-master of host node, this
When, since the network between the adjacent non-master and control node of above-mentioned host node is unreachable, host node it is adjacent non-
Node leave group message can not be sent to control node, therefore the adjacent non-master section of host node by the Resource Management process in host node
Node leave group message is sent back the monitoring process in the adjacent non-master of host node by the Resource Management process in point, by main section
Node leave group message is sent to the monitoring process of adjacent next non-master by the monitoring process in the adjacent non-master of point.
In above-mentioned message method, in resource-area host node adjacent non-master receive host node send node from
After opening message, if the network between the adjacent non-master and control node of above-mentioned host node is reachable, the phase of host node
Above-mentioned node leave group message is sent to above-mentioned control node by adjacent non-master, so as in resource-area host node and control
When network between node processed is unreachable, control node is sent by node leave group message most possibly, so as to control node
Other are notified to have the node of slack resources to carry out fault recovery to the resource left on node, and then the height that cluster can be improved can
The property used.
The message method that FIG. 1 to FIG. 3 illustrated embodiment of the present invention provides is applied to the collection of deployment centralized management model
In group, the architecture diagram of above-mentioned cluster can be as shown in figure 4, Fig. 4 be cluster one implementation that the present invention disposes centralized management model
The architecture diagram of example, referring to fig. 4, cluster shown in Fig. 4 include resource-area and control node, include N number of node in above-mentioned resource-area,
Host in N >=2, Fig. 41~hostNN number of node as in resource-area, host1~hostNPass through net with above-mentioned control node
Network connection.Host in resource-area1~hostNIt is required to deployment monitoring and resource management software, the resource management in resource-area is soft
Part selects host node (master node) of the node as resource-area, it is assumed here that host1Master as resource-area
Node, and assume host1~hostI-1Network between control node is unreachable, hostI~hostNBetween control node
Network it is reachable, hostNIt is the host for having slack resources.If master node monitors that a certain node leaves in resource-area,
And the network between master node and control node is unreachable, then host1Node is to host2Node sending node leave group message.
host2After node receives node leave group message, host2Network between node and control node is also unreachable, then host2Node
Continue node leave group message being sent to host3Node；Dotted line in referring to fig. 4 1., the transmission process of node leave group message are as follows:
host1Node leave group message is sent to host by the Resource Management process of node1The monitoring process of node, host1The monitoring of node
Node leave group message is sent to host by process2The monitoring process of node, host2The monitoring process of node is by node leave group message
It is sent to host2The Resource Management process of node, due to host2Network between node and control node is also unreachable, then
host2Node leave group message is returned to host by the Resource Management process of node2The monitoring process of node, by host2The prison of node
Survey process continues node leave group message being sent to host3The monitoring process of node.
And so on, work as hostINode receives node leave group message, due to hostINetwork between node and control node
It is reachable, then hostIThe message that the Resource Management process of node leaves to controller node sending node, message
Transmission process is as 1. shown such as dotted line in Fig. 4.After control node receives node leave group message, slack resources and network have been found
Reachable node sends node leave group message to hostNNode, as in Fig. 4 solid line 2. shown in.hostNNode receives node
After leave group message, fault recovery is carried out to the resource left on node.In Fig. 4, the transmission order of node leave group message are as follows:
host1->host2->...hostI-> control node -> hostN。
Network between nodes all in the situation that network is the worst, resource-area and control node is unreachable, then main section
Point (host1) the node leave group message that issues is passed back to host node again after being had sent a circle.Host node can be at this time
One timer is set, timer then after, host node retransmit node leave group message, it is ensured that have section in resource-area
Point control node between network up to when, node leave group message can be sent to control node in time.Wherein, above-mentioned fixed
When device the length of timing length sets itself, the present embodiment this can be not construed as limiting in specific implementation, for example, on
The timing length for stating timer can be 10 seconds.
The message method that FIG. 1 to FIG. 3 illustrated embodiment of the present invention provides successively transmits section in the way of token ring
Node leave group message can be sent to control most possibly in the case where no load for increasing network by point leave group message
Node processed, it is ensured that leave the resource on node and be able to carry out fault recovery, and then the high availability of cluster can be improved.
Fig. 5 is the structural schematic diagram of message sending device one embodiment of the present invention, and above-mentioned message sending device is applied to
It disposes in the cluster of centralized management model, it includes at least two in above-mentioned resource-area that above-mentioned cluster, which includes resource-area and control node,
A node, above-mentioned at least two node and above-mentioned control node pass through network connection.Each of resource-area node is required to
Deployment monitors and resource management software, and the resource management software in resource-area selects host node of the node as resource-area
(master node).In the present embodiment, above-mentioned message sending device is set in the host node of above-mentioned at least two node.
As shown in figure 5, above-mentioned message sending device may include: monitoring modular 51 and sending module 52；
Whether monitoring modular 51 has node to leave for monitoring in above-mentioned resource-area；
Sending module 52, for when monitoring modular 51 monitors has node to leave in above-mentioned resource-area, if host node
Network between control node is unreachable or above-mentioned host node is not to receiving after above-mentioned control node sending node leave group message
The confirmation message of above-mentioned node leave group message is directed to above-mentioned control node, and above-mentioned host node is sent to above-mentioned control node
The number of above-mentioned node leave group message reaches preset threshold, then leaves and disappear to the adjacent non-master sending node of above-mentioned host node
Breath, above-mentioned node leave group message includes leaving the essential information and the above-mentioned resource left on node of node, so that above-mentioned main section
Network of the adjacent non-master between above-mentioned control node of point up to when, above-mentioned node leave group message is sent to above-mentioned
Control node.
That is, when monitoring modular 51 monitors has node to leave in above-mentioned resource-area, if host node with it is above-mentioned
Network between control node is unreachable, then sending module 52 is without normal direction control node sending node leave group message；Or main section
Above-mentioned control node is not received after putting to above-mentioned control node sending node leave group message and is directed to above-mentioned node leave group message really
Recognize message, and above-mentioned host node reaches preset threshold to the number that above-mentioned control node sends above-mentioned node leave group message, this
Under two kinds of situations, node leave group message can be sent to the adjacent non-master of host node by sending module 52, if host node
Adjacent non-master and above-mentioned control node between network it is reachable, then the adjacent non-master of host node can be by above-mentioned section
Point leave group message is sent to control node, and if network between the adjacent non-master and above-mentioned control node of host node
Unreachable, then the adjacent non-master of host node can continue above-mentioned node leave group message to be sent to adjacent next non-master section
Point, and so on, until the network between the non-master and control node for receiving node leave group message is reachable, then with control
The node leave group message that the reachable non-master of network between node can will receive is sent to control node.
Wherein, above-mentioned control node is not received after host node is to above-mentioned control node sending node leave group message for upper
When stating the confirmation message of node leave group message, illustrate that above-mentioned node leave group message sends failure.At this time host node need again to
Control node sending node leave group message, at this moment, if host node reaches to the number of control node sending node leave group message
Above-mentioned node leave group message can be sent to the adjacent non-master of host node by preset threshold, host node.
Above-mentioned preset threshold can in specific implementation, according to the realization sets itselfs such as demand and/or system performance, this reality
It applies example to be not construed as limiting the size of above-mentioned preset threshold, for example, above-mentioned preset threshold can be 5.
Further, sending module 52 is also used to when the network in above-mentioned resource-area between all nodes and control node is equal
When unreachable, the above-mentioned node leave group message that host node issues finally is sent back to above-mentioned host node, then is passing through scheduled duration
Later, again to the adjacent non-master sending node leave group message of above-mentioned host node.That is, the situation the worst in network
Under, if the network in above-mentioned resource-area between all nodes and control node is unreachable, the section of the sending of sending module 52
Point leave group message is transferred back to host node again after being had sent a circle, at this moment, after by scheduled duration, sending module
52 can be again to the adjacent non-master sending node leave group message of above-mentioned host node.Specifically, a timing can be set
Device, timer then after, sending module 52 retransmit node leave group message, it is ensured that have in resource-area node and control
Network between node up to when, node leave group message can be sent to control node in time.Wherein, above-mentioned scheduled duration
Length sets itself, the present embodiment can be not construed as limiting this in specific implementation, for example, above-mentioned scheduled duration can be
10 seconds.
In the present embodiment, since reliable communication service is supported in monitoring in resource-area, sending module 52 is specifically used for
Above-mentioned node leave group message is sent to the monitoring process of above-mentioned host node by the Resource Management process of above-mentioned host node, then by
The monitoring process of above-mentioned host node by above-mentioned node leave group message be sent to above-mentioned host node adjacent non-master monitoring into
Journey.
In above-mentioned message sending device, monitoring modular 51 is monitored after having node to leave in above-mentioned resource-area, if main
Network between node and control node is unreachable or above-mentioned host node is not to receiving after control node sending node leave group message
The confirmation message of above-mentioned node leave group message is directed to above-mentioned control node, and above-mentioned host node is sent to above-mentioned control node
The number of above-mentioned node leave group message reaches preset threshold, then sending module 52 is sent to the adjacent non-master of above-mentioned host node
Node leave group message so that network of the adjacent non-master of above-mentioned host node between control node up to when, by node
Leave group message is sent to above-mentioned control node, so as to can not in the host node in resource-area and the network between control node
Up to when, control node is sent by node leave group message most possibly, so that control node notifies other to have slack resources
Node carries out fault recovery to the resource left on node, and then the high availability of cluster can be improved.
Fig. 6 is the structural schematic diagram of another embodiment of message sending device of the present invention, is applied to deployment centralized management mould
In the cluster of type, above-mentioned cluster includes resource-area and control node, includes at least two nodes in above-mentioned resource-area, it is above-mentioned at least
Two nodes and above-mentioned control node pass through network connection.Each of resource-area node is required to deployment monitoring and resource pipe
Software is managed, the resource management software in resource-area selects host node (master node) of the node as resource-area.It is above-mentioned
Message sending device is set in non-master adjacent with host node in above-mentioned at least two node.
As shown in fig. 6, above-mentioned message sending device may include: receiving module 61 and sending module 62；
Wherein, receiving module 61, the node leave group message sent for receiving above-mentioned host node, above-mentioned node leave group message
Essential information and the above-mentioned resource left on node including leaving node, above-mentioned node leave group message is that host node monitors
Stating in resource-area has node to leave, and the network between above-mentioned host node and above-mentioned control node is unreachable or above-mentioned host node
The confirmation that above-mentioned control node is directed to above-mentioned node leave group message is not received after to above-mentioned control node sending node leave group message
Message, and above-mentioned host node reaches preset threshold to the number that above-mentioned control node sends above-mentioned node leave group message and sends out later
It send；Reliable communication service is supported due to monitoring process in resource-area, and in the present embodiment, receiving module 61 is specific to use
Monitoring process in the adjacent non-master by above-mentioned host node receives the node that the monitoring process of above-mentioned host node is sent
Leave group message.
Sending module 62, the network between adjacent non-master and above-mentioned control node for working as above-mentioned host node are reachable
When, above-mentioned node leave group message is sent to above-mentioned control node.In the present embodiment, sending module 62 is specifically used for by upper
Above-mentioned node leave group message is sent to the adjacent non-master of above-mentioned host node by the monitoring process for stating the adjacent non-master of host node
The Resource Management process of node, then above-mentioned node is left by the Resource Management process of the adjacent non-master of above-mentioned host node and is disappeared
Breath is sent to above-mentioned control node.
Further, sending module 62, be also used to when above-mentioned host node adjacent non-master and above-mentioned control node it
Between network it is unreachable when, above-mentioned node leave group message is sent to adjacent next non-master, and so on, until receive
It is reachable to the network between the non-master and above-mentioned control node of above-mentioned node leave group message, then between above-mentioned control node
The node leave group message that will receive of the reachable non-master of network be sent to above-mentioned control node.
Specifically, the monitoring process in adjacent non-master that receiving module 61 passes through host node receives above-mentioned host node
After the node leave group message that monitoring process is sent, node leave group message is sent to main section by monitoring process by sending module 62
Resource Management process in the adjacent non-master of point, at this moment, due to the adjacent non-master and control node of above-mentioned host node
Between network it is unreachable, therefore sending module 62 can not pass through the Resource Management process in the adjacent non-master of host node will
Node leave group message is sent to control node, therefore sending module 62 passes through the resource management in the adjacent non-master of host node
Node leave group message is sent back the monitoring process in the adjacent non-master of host node by process, by the adjacent non-master section of host node
Node leave group message is sent to the monitoring process of adjacent next non-master by the monitoring process in point.
In above-mentioned message sending device, after receiving module 61 receives the node leave group message that host node is sent, if on
The network stated between the adjacent non-master of host node and control node is reachable, then sending module 62 is by above-mentioned node leave group message
It is sent to above-mentioned control node, so as in the host node in resource-area and the unreachable network between control node, most
Possibly send control node for node leave group message greatly, so as to control node notify other have the node of slack resources to from
The resource opened on node carries out fault recovery, and then the high availability of cluster can be improved.
It should be noted that in the description of the present invention, term " first ", " second " etc. are used for description purposes only, without
It can be interpreted as indication or suggestion relative importance.In addition, in the description of the present invention, unless otherwise indicated, the meaning of " multiple "
It is two or more.
Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes
It is one or more for realizing specific logical function or process the step of executable instruction code module, segment or portion
Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussed suitable
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, to execute function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
It should be appreciated that each section of the invention can be realized with hardware, software, firmware or their combination.Above-mentioned
In embodiment, software that multiple steps or method can be executed in memory and by suitable instruction execution system with storage
Or firmware is realized.It, and in another embodiment, can be under well known in the art for example, if realized with hardware
Any one of column technology or their combination are realized: having a logic gates for realizing logic function to data-signal
Discrete logic, with suitable combinational logic gate circuit specific integrated circuit, programmable gate array
(Programmable Gate Array；Hereinafter referred to as: PGA), field programmable gate array (Field Programmable
Gate Array；Hereinafter referred to as: FPGA) etc..
Those skilled in the art are understood that realize all or part of step that above-described embodiment method carries
It suddenly is that relevant hardware can be instructed to complete by program, the program can store in a kind of computer-readable storage medium
In matter, which when being executed, includes the steps that one or a combination set of embodiment of the method.
In addition, each functional module in each embodiment of the present invention can integrate in a processing module, it is also possible to
Modules physically exist alone, and can also be integrated in a module with two or more modules.Above-mentioned integrated module
Both it can take the form of hardware realization, can also have been realized in the form of software function module.If the integrated module
In the form of software function module realize and when sold or used as an independent product, also can store can in a computer
It reads in storage medium.
Storage medium mentioned above can be read-only memory, disk or CD etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
Centainly refer to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be any
One or more embodiment or examples in can be combined in any suitable manner.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of message method, applied to disposing in the cluster of centralized management model, the cluster includes resource-area and control
Node processed includes that at least two nodes, at least two node and the control node pass through network company in the resource-area
It connects, which is characterized in that the described method includes:
Host node in the resource-area monitors have node to leave in the resource-area；
If the network between the host node and the control node is unreachable or the host node is to the control node
The confirmation message that the control node is directed to the node leave group message, and the master are not received after sending node leave group message
Node reaches preset threshold to the number that the control node sends the node leave group message, then the host node is to the master
The adjacent non-master sending node leave group message of node, the node leave group message include leaving essential information and the institute of node
The resource left on node is stated, so that network of the adjacent non-master of the host node between the control node is reachable
When, the node leave group message is sent to the control node；
Wherein, the host node includes: to the adjacent non-master sending node leave group message of the host node
The node leave group message is sent to the monitoring process of the host node by the Resource Management process of the host node, described
The node leave group message is sent to the monitoring process of the adjacent non-master of the host node by the monitoring process of host node.
2. the method according to claim 1, wherein further include:
If the network in the resource-area between all nodes and the control node is unreachable, what the host node issued
The node leave group message is finally sent back to the host node, then by after scheduled duration, the host node again to
The adjacent non-master sending node leave group message of the host node.
3. a kind of message method, applied to disposing in the cluster of centralized management model, the cluster includes resource-area and control
Node processed includes that at least two nodes, at least two node and the control node pass through network company in the resource-area
It connects, which is characterized in that the described method includes:
The adjacent non-master of host node receives the node leave group message that the host node is sent, the node in the resource-area
Leave group message includes leaving the essential information and the resource left on node of node, and the node leave group message is the master
Node monitors have node to leave in the resource-area, and the network between the host node and the control node is unreachable, or
Host node described in person to do not receive after the control node sending node leave group message control node for the node from
The confirmation message of message is opened, and the host node reaches pre- to the number that the control node sends the node leave group message
If being sent after threshold value；
If the network between the adjacent non-master and the control node of the host node is reachable, the phase of the host node
The node leave group message is sent to the control node by adjacent non-master；
Wherein, the adjacent non-master of host node receives the node leave group message packet that the host node is sent in the resource-area
It includes:
Monitoring process in the adjacent non-master of the host node receive the host node monitoring process send node from
Open message.
4. according to the method described in claim 3, it is characterized in that, the adjacent non-master of host node receives in the resource-area
After the node leave group message that the host node is sent, further includes:
If the network between the adjacent non-master and the control node of the host node is unreachable, the host node
The node leave group message is sent to adjacent next non-master by adjacent non-master, and so on, until receiving institute
The network stated between the non-master of node leave group message and the control node is reachable, then the net between the control node
The node leave group message received is sent to the control node by the reachable non-master of network.
5. the method according to claim 3 or 4, which is characterized in that the adjacent non-master of the host node is by the section
Point leave group message is sent to the control node and includes:
The node leave group message is sent to the phase of the host node by the monitoring process of the adjacent non-master of the host node
The Resource Management process of the Resource Management process of adjacent non-master, the adjacent non-master of the host node leaves the node
Message is sent to the control node.
6. a kind of message sending device, applied to disposing in the cluster of centralized management model, the cluster includes resource-area and control
Node processed includes that at least two nodes, at least two node and the control node pass through network company in the resource-area
It connects, the message sending device is set in the host node of at least two node, which is characterized in that the message sends dress
It sets and includes:
Whether monitoring modular has node to leave for monitoring in the resource-area；
Sending module, for when the monitoring module monitors have node to leave into the resource-area, if the host node
Network between the control node is unreachable or the host node is to after the control node sending node leave group message
The confirmation message that the control node is directed to the node leave group message is not received, and the host node is to the control node
The number for sending the node leave group message reaches preset threshold, then to the adjacent non-master sending node of the host node from
Message is opened, the node leave group message includes leaving the essential information and the resource left on node of node, so that described
Network of the adjacent non-master of host node between the control node up to when, the node leave group message is sent to
The control node；
Wherein, the sending module, specifically for the Resource Management process by the host node by the node leave group message
It is sent to the monitoring process of the host node, then the node leave group message is sent to institute by the monitoring process of the host node
State the monitoring process of the adjacent non-master of host node.
7. device according to claim 6, which is characterized in that
The sending module is also used to when the network in the resource-area between all nodes and the control node is unreachable
When, the node leave group message that the host node issues finally is sent back to the host node, then by scheduled duration it
Afterwards, again to the adjacent non-master sending node leave group message of the host node.
8. a kind of message sending device, applied to disposing in the cluster of centralized management model, the cluster includes resource-area and control
Node processed includes that at least two nodes, at least two node and the control node pass through network company in the resource-area
It connects, the message sending device is set in non-master adjacent with host node at least two node, and feature exists
In the message sending device includes:
Receiving module, the node leave group message sent for receiving the host node, the node leave group message include leaving section
The essential information and the resource left on node of point, the node leave group message is that the host node monitors the resource
There is node to leave in area, the network between the host node and the control node is unreachable or the host node is to described
The confirmation message that the control node is directed to the node leave group message is not received after control node sending node leave group message, and
And the host node reaches what preset threshold was sent later to the number that the control node sends the node leave group message；
Sending module, for when the network between the adjacent non-master of the host node and the control node up to when, will
The node leave group message is sent to the control node；
Wherein, the receiving module receives institute specifically for the monitoring process in the adjacent non-master by the host node
State the node leave group message that the monitoring process of host node is sent.
9. device according to claim 8, which is characterized in that
The sending module is also used to when the network between the adjacent non-master and the control node of the host node can not
Up to when, the node leave group message is sent to adjacent next non-master, and so on, until receive the node from
The network opened between the non-master of message and the control node is reachable, then the network between the control node is reachable
The node leave group message received is sent to the control node by non-master.
10. device according to claim 8 or claim 9, which is characterized in that
The sending module leaves the node specifically for the monitoring process of the adjacent non-master by the host node
Message is sent to the Resource Management process of the adjacent non-master of the host node, then the adjacent non-master by the host node
Resource Management process the node leave group message is sent to the control node.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811331220.7A CN109639640B (en) | 2015-10-28 | 2015-10-28 | Message sending method and device |
| CN201510714154.1A CN105407087B (en) | 2015-10-28 | 2015-10-28 | Message method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510714154.1A CN105407087B (en) | 2015-10-28 | 2015-10-28 | Message method and device |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811331220.7A Division CN109639640B (en) | 2015-10-28 | 2015-10-28 | Message sending method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105407087A CN105407087A (en) | 2016-03-16 |
| CN105407087B true CN105407087B (en) | 2019-01-01 |
Family
ID=55472342
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811331220.7A Active CN109639640B (en) | 2015-10-28 | 2015-10-28 | Message sending method and device |
| CN201510714154.1A Active CN105407087B (en) | 2015-10-28 | 2015-10-28 | Message method and device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811331220.7A Active CN109639640B (en) | 2015-10-28 | 2015-10-28 | Message sending method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN109639640B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107391236B (en) * | 2017-09-15 | 2020-03-06 | 郑州云海信息技术有限公司 | Cluster block storage implementation method and device |
| CN107656847A (en) * | 2017-09-19 | 2018-02-02 | 郑州云海信息技术有限公司 | Node administration method, system, device and storage medium based on distributed type assemblies |
| CN112954133B (en) * | 2021-01-20 | 2023-03-14 | 浙江大华技术股份有限公司 | Method, device, electronic device and storage medium for synchronizing node time |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101291337A (en) * | 2008-05-30 | 2008-10-22 | 同济大学 | Grid resource management system and method |
| CN101505272A (en) * | 2009-03-02 | 2009-08-12 | 浙江理工大学 | Method for searching related node subset on the basis of P2P |
| CN101534205A (en) * | 2008-03-11 | 2009-09-16 | 中国网通集团宽带业务应用国家工程实验室有限公司 | Application layer multicast service realizing method, terminal and system thereof |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1198407C (en) * | 2003-07-08 | 2005-04-20 | 大唐移动通信设备有限公司 | Device and method for realizing high availability in communication system |
| CN101640688B (en) * | 2009-08-20 | 2014-03-12 | 中兴通讯股份有限公司 | Content delivery network (CDN)-based switching method for main node controller and spare controller and CDN |
| CN101859480A (en) * | 2010-04-29 | 2010-10-13 | 顾翠红 | Electric appliance wireless control network |
| CN102769673B (en) * | 2012-07-25 | 2015-03-25 | 深圳市中博科创信息技术有限公司 | Failure detection method suitable to large-scale storage cluster |
| CN103118084B (en) * | 2013-01-21 | 2016-08-17 | 浪潮(北京)电子信息产业有限公司 | The electoral machinery of a kind of host node and node |
| CN104581954B (en) * | 2013-10-28 | 2017-12-19 | 株式会社理光 | Channel scheduling method and Wireless Telecom Equipment |
-
2015
- 2015-10-28 CN CN201811331220.7A patent/CN109639640B/en active Active
- 2015-10-28 CN CN201510714154.1A patent/CN105407087B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101534205A (en) * | 2008-03-11 | 2009-09-16 | 中国网通集团宽带业务应用国家工程实验室有限公司 | Application layer multicast service realizing method, terminal and system thereof |
| CN101291337A (en) * | 2008-05-30 | 2008-10-22 | 同济大学 | Grid resource management system and method |
| CN101505272A (en) * | 2009-03-02 | 2009-08-12 | 浙江理工大学 | Method for searching related node subset on the basis of P2P |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105407087A (en) | 2016-03-16 |
| CN109639640B (en) | 2020-07-14 |
| CN109639640A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105407087B (en) | Message method and device | |
| US9407582B2 (en) | Redundantly operable industrial communication system and method for operation thereof | |
| CN104054265A (en) | Fault - tolerant power semiconductor switching device control system | |
| CN103843286B (en) | For the method and apparatus based on the triggering router state change of exchanger connectivity | |
| CN103345435B (en) | Destination sever system for data backup | |
| EP2911341A1 (en) | Method and device for dynamically switching gateway of distributed resilient network interconnect | |
| CN100499507C (en) | Disaster recovery system, method and network device | |
| US9460035B2 (en) | Method for operating a slave node of a digital bus system | |
| CN104067516A (en) | Switching control system signalling techniques | |
| CN101217424A (en) | A detecting method and device of aggregated link failures | |
| US20190334760A1 (en) | Architecture for wireless avionics communication networks | |
| CN104054243A (en) | Switching control systems | |
| EP2023541A1 (en) | Method, device and communication system to avoid loops in an Ethernet Ring System with an underlying 802.3ad network | |
| CN103036756A (en) | Hybrid looped network protection method and system based on sharing channel | |
| CN104243259A (en) | Method and system for cooperation of distributed elastic network interconnection systems | |
| CN203340113U (en) | Source server system for through intranet-extranet long-distance data transmission and target server system | |
| CN107656847A (en) | Node administration method, system, device and storage medium based on distributed type assemblies | |
| CN100466582C (en) | Cross interconnection method for RPR and cross site, cross RPR | |
| CN103179013A (en) | Looped network protection achieving method and system | |
| KR101358995B1 (en) | Method and system for managing high availability | |
| CN101141195B (en) | Considerable alarm reporting method and system on optical communication transferring plane | |
| CN101009625A (en) | An instant loop monitoring network system | |
| CN102111299A (en) | Redundant network system and method for realizing redundancy | |
| JP5243138B2 (en) | Information processing apparatus and information processing method | |
| CN113992512A (en) | TTE system high-reliability configuration switching control protocol and reconfiguration method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |