CN1567889A - A method of automatic primary standby equipment switching - Google Patents

Abstract

The present invention provides a method for automatic switching of active and standby equipment. By setting different states of the transmission equipment in the data transmission ring network, it is ensured that when the main equipment fails, the standby node automatically switches the state of its own parameters, and the standby equipment is replaced and invalidated. The master device continues to work, thereby realizing the protection of the data of the master device and ensuring the stability and security of the data transmission ring network. The switching actions of all active and standby devices in the data transmission ring network are automatically completed, and because the solution of the present invention does not need to reserve a specific bandwidth, the cost is reduced; at the same time, the present invention has no impact on other devices in the data transmission ring network. Other additional requirements, thus satisfying common situations, and the protection action of the present invention is transparent to the user. The application of the present invention is simple and reliable. No manual intervention is required during the switching process of the active and standby servers, and the switching time can be completed within 50ms.

Description

A method for automatic switching of active and standby equipment

technical field

The invention relates to the technical field of master-standby switchover, in particular to a method for automatic switchover of master-standby devices on a ring network.

Background technique

With the development of the communication industry, people's study, life and work are increasingly inseparable from the network. Figure 1 shows a schematic diagram of data access to the Internet. The data of the user accessing the Internet is first connected to the access server, such as a digital subscriber line access server (DSLAM) and other equipment. After the user's data is processed by the access server, it enters the data transmission ring network, and the user's data The data is transmitted to the server, such as broadband access server (BAS), core switch and other equipment. After the server completes relevant authentication and other work for user data, it provides access to the Internet. The network management center completes the configuration, performance and alarm of the entire network. monitoring and management functions. In the figure, there are N DSLAM devices, M primary BAS devices and a backup BAS device, among which DSLAM1, DSLAM2 and DSALM3 access the Internet through BAS1, and DSLAM4, DSLAM5...DSLAMN use BAS2~M to access the Internet respectively . If a main BAS is damaged during use, under the deployment of the network management center, the damaged main BAS will be replaced by the backup BAS to work.

As the data transmission network has more and more influence on people's life and work, its security and stability are becoming more and more important. In some special industries such as: financial, electric power and government and other important departments, the interruption of data transmission network often brings immeasurable losses. It is very important to improve the stability of the data network operation and ensure the normal communication of data in the event of a data network failure. The existing application solutions mainly include the following two types:

1) Optical fiber direct connection. As shown in Figure 2, each DSLAM device provides two sets of optical cables at the same time, which are directly connected to the main and standby BAS respectively. In the figure, 101 and 102 represent the data flow connected to the main BAS, and 201 and 202 represent the data flow connected to the standby BAS. If the main BAS or the optical fiber at point A fails, it can be detected between the main BAS and the DSLAM. At this time, the backup channel of the DSLAM transmits the business flow to the backup BAS, and the backup BAS provides the service of accessing the Internet.

The defect of the above scheme is that although the switching process of the active and standby BAS can be completed automatically, since each DSLAM needs to provide two sets of optical cables to be connected to the active and standby BAS respectively, and each BAS is directly connected to all DSLAMs, therefore, not only The waste of optical cable resources is large, and the port resources of the BAS are seriously wasted. If the distance between the BAS and the DSLAM is long, and there is no aggregation process between them, the waste of optical cable resources is particularly prominent, resulting in some optical cable resources. Stressed cities cannot adopt this option.

2) Data transmission ring network mode. Between BAS and DSLAM, the data transmission network is used to transmit the data flow. As shown in Figure 3, the data transmission ring network can provide service transmission, convergence and protection functions. The transmission equipment in the figure is a synchronous digital hierarchy (SDH) or asynchronous transfer mode (ATM) equipment used to complete data stream transmission. The solid line 103 is a networking line. The data flow 101 accessed through DSLAM1 is transmitted to the two main and standby BASs at the same time, and the data flow 102 accessed through DSLAM2 is also transmitted to the two main and standby BASs at the same time. Whether it is the main BAS, or the optical fiber at point A, or the transmission equipment connected to the main BAS, the network management center will find the problem. At this time, the BAS network management personnel will notify the network management personnel of the data transmission ring network that the main BAS cannot provide Internet access services. , the management personnel of the data transmission ring network re-configure the business, manually complete the switchover of the active and standby BAS, and the standby BAS provides the service of accessing the Internet. The above switchover time generally takes about 30 minutes.

The defect of the above scheme is that: because the switching process requires manual intervention, not only the service recovery time is very long, but also there may be other online accidents caused by improper operation. At the same time, because each DSLAM needs to occupy two paths to transmit the received Protected services waste a lot of transmission bandwidth.

Contents of the invention

In view of this, the purpose of the present invention is to provide a method for automatic switching of master and backup devices based on the data transmission ring network, and at the same time achieve the protection of data in one or more master devices by using one backup device, or use multiple backup devices The purpose of the device protecting data in multiple active devices.

For achieving the above object, technical scheme of the present invention is achieved in that way:

A method for automatic switching of active and standby equipment, the method comprising the following steps:

a. Set parameters identifying the state of the master and backup devices for all master and backup devices;

b. According to the obtained working status of the master device, judge whether the master device is working normally. If yes, the master device continues to process the received data flow normally, and the backup device transparently transmits the user's data stream. Otherwise, the master device and the backup device automatically Change the state of its own parameters, the backup device switches to the master device to process the received data stream, and the master device switches to the backup device to transparently transmit the user's data stream.

Preferably, the master device is a combination of a master mount device connected to the ring network and a transmission device serving as a ring network node, and the backup device is a standby mount device connected to the ring network and a ring network node The combination of the transmission equipment, the ring network nodes contained in the main and standby equipment are respectively the main and standby nodes,

In step a, the network management center sets parameter states for all active and standby nodes and performs initial configuration;

The step b is: the standby node regularly sends management messages to all the master nodes, and judges whether the master device works normally according to whether it receives a response message from the master node within a specified time.

Preferably, the primary and backup devices are node transmission devices forming a ring, the primary and secondary devices are respectively primary and secondary nodes, and the primary and secondary nodes are located in the same device containing the control module, and the control module specifies the primary node and the secondary node,

In step a, the control module in the device sets parameter states for all active and standby nodes and performs initial configuration;

The step b is: the control module directly determines whether the master device is working normally.

Preferably, the parameters at least include node business processing mode parameters, node business active/standby mode parameters, node working address parameters and node protection address parameters, and the parameter settings further include:

Set the node business processing mode parameter of the master node to normal mode, and the node business master/standby mode parameter to work mode;

Set the node business processing mode parameter of the standby node to pass-through mode, and the node business master/standby mode parameter to backup mode.

Preferably, the active and standby devices in step b automatically changing their own parameter states further include:

The master node sets its own node business processing mode parameter as the pass-through mode, and the node business master/standby mode parameter as the backup mode;

The standby node sets its own node business processing mode parameter to normal mode, the node business active standby mode parameter to work mode, and sets its own node working address to the address of the node to be replaced.

Preferably, the step b further includes: the backup node determines the address of the failed master node according to the unreceived response message, and replaces the failed master node with the backup device to process the received data flow.

Preferably, the method further includes: judging whether more than one master device fails at the same time, and if so, performing master-standby switchover according to the preset priority level of the master device, otherwise, directly performing master-standby switchover.

Preferably, there is more than one standby device in the system, and the method further includes: determining the current usage status of the other party through the interaction of service messages between the standby hook-up devices.

Preferably, the method further includes: judging whether there is more than one master device, or the mount device directly connected to the master device, or the optical cable between the master device and the mount device directly connected to the master device fails, if , the master-standby switchover will be performed according to the preset priority level of the master device, otherwise, the master-standby switchover will be performed directly.

Preferably, the ring network is an elastic packet ring network, or a data packet ring network, or an ordinary ring network with a data packet function.

By applying the present invention, by setting different states of the transmission equipment in the data transmission ring network, it is ensured that when the main equipment fails, the standby equipment switches its own state, and the standby equipment replaces the failed main equipment to continue working, thereby realizing the control of the main equipment. Data protection ensures the stability and security of the data transmission ring network. The switching actions of all active and standby devices in the data transmission ring network are automatically completed, and because the solution of the present invention does not need to reserve a specific bandwidth, the cost is reduced; at the same time, the present invention has no impact on other devices in the data transmission ring network. Other additional requirements, thus satisfying common situations, and the protection action of the present invention is transparent to the user. The application of the present invention is simple and reliable. No manual intervention is required during the switchover process of the active and standby equipment, and the switchover time can be completed within 50ms.

Description of drawings

Figure 1 shows a schematic diagram of data access to the Internet;

FIG. 2 is a schematic diagram of a scheme 1 of the prior art;

FIG. 3 is a schematic diagram of a second scheme of the prior art;

FIG. 4 is a schematic diagram of the flow direction of service data applying the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

The present invention is based on the data ring network, and utilizes the characteristics of transmission and aggregation of the data ring network to transmit the accessed data to the main device connected to the ring network through the transmission equipment of the ring network, and the main device is responsible for receiving the received data. The data are processed accordingly.

Figure 4 shows an embodiment of the application of the present invention. In this embodiment, the data ring network with elastic grouping characteristics and the data servers attached to the ring network are taken as an example for specific description, and the active and standby data servers and the transmission devices on the ring network directly connected to the active and standby data servers are called The active and standby devices, and the transmission devices in the active and standby devices are respectively called the active and standby nodes.

In this embodiment, the network management center completes the configuration, performance, alarm monitoring and management of the entire network in the entire system.

The network management center sets parameters for all nodes directly connected to the active and standby servers, such as node 1 and node 2 in Figure 4, and each node has the following parameters:

Node business processing mode (ProcessMode) parameter, which has two states: normal mode (normal) and passthrough mode (passthrough), wherein, normal mode (normal) means that the node performs uplink and downlink transmission of the passing data stream; The passthrough mode means that the data flow can only pass through the node, and the node cannot realize the control of the uplink and downlink transmission of the data flow, but the node can recognize the flag bit of the transmitted message and intercept the control node. message, while other messages are transparently transmitted.

Node business master mode (ProtectMode) parameter, this parameter has two states of work mode (work) and backup mode (back), among them, work mode (work) means that the node can process data flow normally, control data flow in this The node performs uplink and downlink transmission; the backup mode (back) means that the node works in the backup mode, and only completes the pass-through function for the data flow, and cannot control the data flow for uplink and downlink transmission.

The node working address (NodeNormalAddress) parameter refers to the marked ring node address on the data transmission ring network where the node is located during normal operation.

The node protection address (NodeProtectAddress[N]) parameter refers to the marked ring node address on the data transmission ring network to be protected when the node in the backup working mode switches to the data transmission ring network to be protected when the protection is implemented. Wherein, N is the number to be protected.

The backup node regularly sends management messages to the master node, which can check whether the working status of the master node is normal through a fixed channel or other methods. The management messages are similar to mutual heartbeat checks; the master node receives the management messages sent by the backup node Then give a response. In the above management message, the detection information may also include the data backup information of the master server, so that the data in the master server and the slave server are consistent. The standby node can identify the flag bit of the management message and process it without transparently transmitting it to the next node.

If the backup node and the master node can periodically receive mutual management messages, it is considered that the master server is normal. If the slave node does not receive the message response from the master node within the specified time, it is considered that the master server has failed, The standby node automatically changes its own parameter status to replace the master node to work normally. The data flow is no longer passed through, and the corresponding processing is carried out. In this embodiment, a DSLAM device is used as an example for the access server, and a BAS device is used as an example for the server, and the transmission device in the data transmission ring network is defined as a node for simplicity of description.

The backup scheme of the present invention can be divided into three cases: 1:1 backup scheme, 1:N backup scheme and M:N backup scheme, wherein, M is the number of backup equipment, N is the number of active equipment, and M must be less than It is equal to N (M<=N), and the values of M and N are determined according to specific design requirements, which will be described respectively below.

1. 1:1 backup scheme

FIG. 4 is a schematic diagram of the flow direction of service data applying the present invention. Nodes 1 to 4 in the figure are transmission devices with an RPR function or similar functions. The above-mentioned transmission devices form a RPR network, and the solid line 103 is a networking line. Assuming that node 1 is the master node and node 2 is the standby node, node 1 processes the data flow normally, and the control data flow is transmitted uplink and downlink on this node, and node 2 only completes the pass-through function without any processing on the data flow. Then, the specific configuration of all parameters of node 1 and node 2 is:

Parameter configuration of node 1:

ProcessMode＝normal ProtectMode＝work

NodeNormalAddress＝Node1 NodeProtectAddress＝Node2

Parameter configuration of node 2:

ProcessMode＝passthrough ProtectMode＝back

NodeNormalAddress＝Node2 NodeProtectAddress＝Node1

Under normal circumstances, the data flow 101 accessed through DSLAM1 is transmitted to the main BAS through node 4 and node 1, and the main BAS provides the service of accessing the Internet; the data flow 102 accessed through DSLAM2 passes through node 3, node 4, node 1 is transmitted to the main BAS, and the main BAS provides the service of accessing the Internet. At the same time, the backup node 2 sends management messages to the master node 1 regularly to check whether the working status of the master node 1 is normal. The management messages are similar to mutual heartbeat checks. Node 2 responds to the management message sent by it. The above management message includes not only the detection information but also the data backup information of the master BAS, so that the data in the master and backup BASs are consistent. The standby node 2 can identify the flag bit of the management message and process it without transparently transmitting it to the next node.

If the standby node 2 and the primary node 1 can periodically receive mutual management messages, it is considered that the primary server is normal. If the secondary node 2 does not receive the message response from the primary node 1 within the specified time, such as 100ms, Then it is considered that the main server has a failure. The failure can be the failure of the main BAS, or the failure of node 1, or the failure of the optical fiber between node 1 and the main BAS. At this time, node 2 automatically changes all its own parameter settings: set The working address is set to Node1, that is, NodeNormalAddress=Node1, the node active/standby mode is set to the working mode, that is, ProtectMode=work, the node business processing mode is set to the normal mode, that is, ProcessMode=normal, and meanwhile, the business processing mode parameter of node 1 is set to Passthrough state, that is, ProcessMode=passthrough, so that node 2 automatically handles the work that should be done by the master node 1, and the backup BAS connects the data flow to the Internet, thereby automatically completing the switchover between the master and backup servers. If the above-mentioned fault occurs at node 1, that is, the data flow pass-through function cannot be completed due to node 1 being damaged, since the resilient packet ring network can detect errors and enable each affected node to start the fail-over function, the The function can redirect a data flow from the failure path to the protection path and retransmit it, so the backup node 2 can handle all the work that should be done by the master node 1.

The networking shown in Figure 4 can also be in the form of dual-homing. It is defined that the active BAS is in the working state, the standby BAS is in the backup state, node 1 is in the working mode, and node 2 is in the pass-through mode. Node 1 and Node 2 are two geographically separated nodes, and their node numbers can be the same, and they are distinguished by their working status. Under normal operation, the data of nodes 3 and 4 are aggregated to the main BAS, and node 2 is in the pass-through working mode, does not receive or send data, but is in hot backup work. When the main BAS fails, or node 1 fails, or the connection line between the main BAS and node 1 fails, node 1 can set its own working state to pass-through state, and node 2 can set its own working state to working state , and change its corresponding parameters, the data flow on the ring network is converged from node 2 to the standby BAS, thus realizing the 1:1 dual-homing protection on the RPR ring network.

2. 1:N active/standby protection scheme

This scheme can realize the situation that one set of standby servers protects N sets of active servers at the same time. The specific processing method is as follows:

The nodes directly connected to the standby server send management messages to the nodes directly connected to the N sets of active servers at the same time, and the nodes directly connected to the N sets of active servers respectively send management message response messages to the nodes directly connected to the standby servers.

If the node directly connected to the standby server can periodically receive the response messages of all management messages within the specified time, it is considered that the active server is normal; otherwise, the active server is considered to be faulty, and the fault can be the If a failure occurs, or the node directly connected to the active server fails, or the optical fiber between the active server and the node fails, the node directly connected to the standby server judges the failed active node according to the unreceived response message address, and set the working address of the standby node as the address of the node to be protected, that is, NodeNormalAddress[X], where X is the number of the active node that fails, and set the business processing mode parameter of the standby node to normal mode, and the active node The standby mode parameter is the working mode. In this way, the standby node automatically handles the work that should be done by the primary node X, and the standby server connects the data flow to the Internet, thereby automatically completing the switchover between the active and standby servers. If the above-mentioned failure occurs on node X, that is, the data flow pass-through function cannot be completed due to node X being damaged, the backup node can still handle all the data that should be completed by the master node X due to the fail-over function of the resilient packet ring network. work.

If multiple active nodes fail at the same time during work, you can pre-set the priority level of the active node, and perform active-standby switchover according to the priority level.

3. M:N active/standby protection scheme

This scheme can realize the situation that M sets of standby servers protect N sets of active servers at the same time. The specific processing method is as follows:

Nodes directly connected to M sets of standby servers send management messages to nodes directly connected to N sets of active servers at the same time, and nodes directly connected to N sets of active servers send management messages to M sets of standby servers according to the received messages. The node sends a response message. At the same time, M sets of backup servers can communicate with each other through service messages to understand the usage of the backup servers, such as one or several backup servers have replaced the main server for work, or one or several backup servers Still in the standby state, this message is similar to the management message, and the standby node can identify the flag bit of the service message and process it without transparently transmitting it to the next node.

If the nodes directly connected to M sets of standby servers can periodically receive the response messages of all management messages within the specified time, it is considered that the active server is normal; otherwise, it is considered that the active server fails, and the failure can be the main If the active server fails, or the node directly connected to the active server fails, or the optical fiber between the active server and the node fails, the node directly connected to M sets of standby servers judges that there is a failure based on the unreceived response message The address of the active node, and select a set of standby servers according to the pre-strategy. The pre-strategy can be round robin, table look-up, or priority setting. The working address of the node directly connected to the set of standby servers is set is the address to be protected, that is, NodeNormalAddress[X], where X is the number of the active node that has failed, and at the same time set the business processing mode parameter of the backup node to normal mode, and the node master/standby mode parameter to work mode. In this way, the standby node The node automatically handles the work that should be done by the primary node X, and the standby server connects the data flow to the Internet, thereby automatically completing the switchover between the primary and secondary servers. If the above-mentioned failure occurs on node X, that is, the data flow pass-through function cannot be completed due to node X being damaged, the backup node can still handle all the data that should be completed by the master node X due to the fail-over function of the resilient packet ring network. work.

Another embodiment of the present invention is: the transmission equipment with the characteristics of the elastic packet ring network is used as a module, a plurality of modules with the characteristics of the elastic packet ring network are located in the same device and form a ring, and the control module in the device controls each Each module with the characteristics of the elastic packet ring network is controlled, and a parameter identifying the status of the module is set for each module with the characteristics of the elastic packet ring network, and its working status is determined. The transmission equipment in the working state forming the ring is called the main equipment, that is, the master node, and the transmission equipment in the non-working state forming the ring is called the standby equipment, that is, the standby node. When the master device fails, switch the parameter status of the backup node and the parameter status of the master node, so that the backup device replaces the failed master device and continues to work normally.

The main and standby devices can be connected to one external device at the same time, or they can be connected to different external devices respectively.

The above is only a preferred embodiment implemented by using the elastic packet ring network, and the data ring network may also be a data packet ring network, or an ordinary data ring network with a data packet function.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of the present invention. within the scope of protection.

Claims (10)

1, a kind of method of master/slave device automatic switchover is characterized in that this method may further comprise the steps:

A, for all master/slave devices be provided with the sign this master/slave device state parameter;

B, according to the operating state of the main equipment that is obtained, judge whether main equipment is working properly, if, then main equipment continues the data flow that normal process is received, be equipped with equipment transparent transmission user's data stream, otherwise master/slave device changes the state of self parameter automatically, be equipped with equipment and switch to the data flow that the main equipment processing is received, main equipment switches to equipment transparent transmission user's data stream fully.

2, method according to claim 1, it is characterized in that, described main equipment is for the main locking equipment that articulates with looped network with as the combination of the transmission equipment of looped network node, described be equipped with equipment for looped network articulate be equipped with locking equipment and as the combination of the transmission equipment of looped network node, the looped network node that is comprised in the master/slave device is respectively main-standby nodes

Described step a is provided with parameter state and carries out initial configuration for all main-standby nodes by network management center;

Described step b is: slave node regularly sends administrative messag to all host nodes, and judges according to the response message of whether receiving host node in official hour whether host apparatus is working properly.

3, method according to claim 1 is characterized in that, described master/slave device is for constituting the node transmission equipment of ring, master/slave device is respectively main-standby nodes, and main-standby nodes is arranged in the same equipment that comprises control module, by control module designated host and slave node

Control module in this equipment of described step a is provided with parameter state and carries out initial configuration for all main-standby nodes;

Described step b is: determine directly by control module whether host apparatus is working properly.

4, according to claim 2 or 3 described methods; it is characterized in that; described parameter comprises node traffic tupe parameter, the active and standby mode parameter of node traffic, node work address parameter and Node Protection address parameter at least, and described parameter is provided with further and comprises:

The node traffic tupe parameter that host node is set is a normal mode, and the active and standby mode parameter of node traffic is a mode of operation;

The node traffic tupe parameter that slave node is set is the break-through pattern, and the active and standby mode parameter of node traffic is a backup mode.

5, method according to claim 4 is characterized in that, the described master/slave device of step b changes self parameter state automatically and further comprises:

The node traffic tupe parameter that host node is provided with himself is the break-through pattern, and the active and standby mode parameter of node traffic is a backup mode;

The node traffic tupe parameter that slave node is provided with himself is a normal mode, and the active and standby mode parameter of node traffic is a mode of operation, and the node work address that self is set is address of node to be replaced.

6, method according to claim 2, it is characterized in that, described step b further comprises: slave node is according to unreceived response message, determines the address of the host node that breaks down, replaces this main equipment that breaks down and handles the data flow of being received being equipped with equipment.

7, method according to claim 6 is characterized in that, this method further comprises: judged whether that an above main equipment breaks down simultaneously, if, then carry out active and standby switching according to the priority level of predefined main equipment, otherwise, directly carry out active and standby switching.

8, method according to claim 2 is characterized in that, exists in the system to be equipped with equipment more than one, and this method further comprises: be equipped with and pass through the mutual of service message between the locking equipment, determine the user mode that the other side is current.

9, method according to claim 3, it is characterized in that, this method further comprises: judged whether an above main equipment, or the locking equipment that directly links to each other with main equipment, or main equipment and break down with optical cable between the locking equipment that main equipment directly links to each other, if then carry out active and standby switching according to the priority level of predefined main equipment, otherwise, directly carry out active and standby switching.

10, method according to claim 1 is characterized in that, described looped network is a RPR network, or the packet looped network, or has the common looped network of packet function.

Images