CN1980224A - Method and system for recovering business after apparatus state switching-over based on main and spare netgate - Google Patents

Abstract

The present invention discloses a method and system for service restoration based on switching of active and standby gateway devices. The core of the invention is that when a BGP protocol connection is established between a user device and a gateway device, the gateway device uses a VRRP virtual address as the source of the BGP protocol. address, after the switchover of the active and standby state of the gateway device, the new gateway active device establishes a BGP protocol connection with the user equipment through GR to ensure that the user equipment and the new gateway active device establish a protocol connection smoothly. Through the present invention, in the scenario where BGP protocol neighbors are established between the user equipment and the gateway equipment, when the main and standby equipment of the gateway is switched, the GR method can effectively shorten the service interruption time caused by the switching of the gateway state, and greatly satisfy the needs of users. Requirements for network reliability.

Description

Method and system for service recovery after active/standby gateway device state switching

technical field

The present invention relates to the field of communications, in particular to a method and system for service restoration based on state switching of active and standby gateway devices.

Background technique

With the development of network technology, users have increasingly higher requirements for network reliability. User equipment usually requires access to dual gateways, and the Virtual Router Redundancy Protocol (VRRP) runs between the gateways, so that the switchover It is transparent to user equipment. In some scenarios, the user equipment is a layer-3 device such as a router, which runs the BGP protocol with the gateway device that allows the VRRP protocol to exchange routes with each other. In this application scenario, if the failure of the active device causes the state of the gateway device to switch, the user equipment needs to remove the Border Gateway Protocol (BGP) neighbor relationship established with the previous active gateway device, and then communicate with the The new gateway master device establishes the neighbor relationship of the BGP protocol. There must be a process of deleting and recreating the forwarding table entries generated by the BGP protocol, which will take a long time (several seconds, ten seconds or even tens of seconds) business interruption, it is difficult to meet customer requirements for network reliability.

In the existing network system, the gateway device does not use the VRRP function, and the interface or loopback address of the gateway device is used to establish a BGP protocol connection between the user device and the dual gateway device. The user device uses the optimal technology of the BGP protocol itself to select a gateway device as the For the service uplink egress, when the selected gateway device fails, the user equipment detects the failure by using the convergence technology of the BGP protocol itself, and selects another gateway device as the service uplink egress. In this prior art, when the user equipment perceives the failure of the active gateway device and modifies the forwarding entry pointing to the active gateway device to point to the standby gateway, the service is interrupted, and when the number of forwarding entries is large, the service is interrupted The time is still relatively long, and it is difficult to meet the user's requirements for network reliability.

Contents of the invention

In view of this, the main purpose of the present invention is to provide a method for service restoration based on the state switching of the active and standby gateway devices. When the switchover occurs between the main gateway device and the standby device, a new BGP protocol connection between the gateway master device and the user equipment can be established without causing service interruption by means of a graceful restart (GR).

The present invention provides a method for service recovery based on the state switching of active and standby gateway equipment, which is characterized in that it includes the following steps:

A. When a Border Gateway Protocol (BGP) connection is established between the user equipment and the gateway device, the gateway device uses a virtual redundant routing protocol (VRRP) virtual address as the source address of the BGP protocol;

B. After the switchover of the active and standby state of the gateway device, the new gateway active device establishes a Border Gateway Protocol (BGP) connection with the user equipment in a graceful restart (GR) manner to ensure smooth establishment of the user equipment and the new gateway active device protocol connection.

The step B specifically includes adding GR technology to the BGP protocol to expand the functions of the BGP protocol, so that the GR technology can be implemented between the user equipment and the gateway equipment.

The step B also includes,

B1. After the gateway device status switch, the VRRP module in the new gateway master device notifies the gateway device status switch message to the BGP module of the gateway device;

B2. After the BGP module of the gateway device receives the gateway device state switching message sent by the new gateway master device, it actively sends a connection request message of the BGP protocol to the BGP module of the user device;

B3. The BGP module of the user equipment processes the received BGP connection request message according to the GR technology.

The B2 specifically includes that the BGP module of the gateway device puts a restart (Restart) mark in the neighbor establishment message (OPEN) to initiate a protocol connection request message with the user equipment.

After step B, also specifically include:

C. Establish an IBGP protocol connection between the new gateway master device and the previous gateway master device, so that both the new gateway master device and the previous gateway master device have routes published by the user equipment;

D. After the state of the gateway device is switched, the new gateway master device and the previous gateway master device do not actively delete the route learned by the previous gateway master device from the user equipment, and the new gateway master device uses the route to Guide the forwarding of the traffic coming back from the bearer network to the user equipment, so as to ensure that the service returning to the user equipment is not interrupted.

Described step D specifically comprises,

D1. During the GR process between the new gateway master device and the user equipment, the new gateway master device will relearn from the user equipment the route from the previous gateway master device pointing to the user equipment;

D2. The new gateway master device forwards the route to the previous gateway master device through the IBGP protocol, and the previous gateway master device retains the received route and deletes its slave user according to the received route The corresponding route directly learned by the device;

D3. After the GR between the new gateway master device and the user equipment ends, the new gateway master device deletes all the routes pointing to the user equipment learned from the previous gateway master device, and at the same time, the new gateway master device sends the previous The gateway active device notifies the end of GR, and the previous gateway active device deletes all the routes obtained directly from the user equipment.

In step D3, the new gateway active device notifies the previous gateway active device of the end of GR through an extended VRRP protocol message.

In order to avoid GR failure or GR recovery time becoming longer, the source port number of the connection established by the new gateway master device is different from the port number of the previous gateway master device connection.

When the gateway active device establishes a connection with the user equipment, the gateway active device notifies the gateway backup device of the source port number used by it to establish a BGP neighbor relationship with the user equipment through the extended VRRP protocol.

The present invention also provides a system based on the service recovery after the state switching of the active and standby gateway equipment, including the gateway equipment, the user equipment, the BGP module, the VRRP module, the VRRP switching perception module and the switching notification module,

A virtual redundant routing protocol is run between the gateway devices through a VRRP module;

Both the gateway device and the user equipment are configured with a graceful restart (GR) technology, and a Border Gateway Protocol connection is established through a BGP module;

The switching notification module is used to notify the gateway device state switching message to the VRRP switching perception module:

The VRRP switching perception module initiates a BGP protocol connection request message to the BGP module of the user equipment according to the received gateway device state switching message;

The BGP module of the user equipment performs graceful restart (GR) according to the received BGP protocol connection request message.

The switching notification module is embedded in the VRRP module.

The VRRP switching perception module is embedded in the BGP module of the gateway device.

Description of drawings

Fig. 1 is a schematic diagram of the basic networking scheme of the present invention;

Fig. 2 is the method flowchart of the present invention;

FIG. 3 is a schematic diagram of a system according to an embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention more clearly, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Fig. 1 is a schematic diagram of the basic networking scheme of the present invention, referring to Fig. 1, assuming that gateway A is the default gateway master device, gateway B is the backup device, and the signaling of the gateway is in gateway A—switch A—switch B— - carried on the link between gateway B.

The state switchover of gateway master and backup devices includes the following situations, A. When gateway A fails, gateway B detects the failure of gateway A through VRRP signaling or other methods, and switches to the gateway master device; B. Gateway A and gateway B The priority between them is reversed, gateway B is configured in the preemption mode, and gateway B actively switches to the gateway active device; C. Modify the configuration of gateway A so that it leaves the gateway backup group, and gateway B receives the corresponding signaling message After that, actively switch to the main device of the gateway;

When the main and standby state of the gateway is switched, the present invention provides a method for business recovery based on the state switching of the active and standby gateway equipment. To achieve the establishment of a new BGP connection between the main gateway device and the user equipment without causing service interruption.

The specific technical solution of the present invention is as follows, including: gateway equipment, user equipment, BGP module and VRRP module. The virtual redundant routing protocol is run between the gateway devices; the gateway device and the user equipment are configured with graceful restart (GR) technology; the gateway device uses a VRRP virtual address as the source address of the BGP protocol connection; the User equipment using VRRP virtual address as the destination address of the BGP protocol connection;

Configure the binding relationship between VRRP protocol and BGP protocol on the gateway device (including the active device and the backup device). Only the active gateway device can normally receive and send protocol messages with the user equipment, while the backup gateway device neither receives nor sends such messages. In this way, the active gateway device can establish a normal protocol with the user equipment. protocol neighbors, but gateway backup devices cannot establish such neighbor relationships. From the point of view of the user equipment, it does not care whether its protocol neighbor is gateway A or gateway B. The neighbor address of its BGP protocol is the gateway address (VRRP virtual address). Correspondingly, the next The hop address is also the address of the gateway (VRRP virtual address). The destination MAC address of the message forwarded by these forwarding items is the MAC address corresponding to the VRRP virtual address. According to the principle of the VRRP protocol, it can only be received by the main gateway device and performed three layer forwarding;

When the state of the gateway device is switched due to various reasons described above, according to the method provided by the present invention, as shown in Figure 2,

Step 201, the state of the gateway device is switched;

For example, gateway A was previously the active gateway device, and when the status of the gateway device is switched, gateway B is upgraded to be the new active gateway device.

Step 202, switching notification module gateway device state switching occurs;

When the gateway device state is switched, the switch notification module notifies the gateway device state switch message to the VRRP switch perception module;

As shown in FIG. 3 , the switching notification module can be embedded in the VRRP module, and the VRRP switching awareness module can be embedded in the BGP module of the gateway device.

Step 203, the VRRP switching perception module initiates a BGP protocol connection request message to the BGP module of the user equipment;

According to our previous description, the gateway B device is configured with the BGP GR capability, and the BGP module of gateway B interprets the gateway device state switching message as a GR event, and actively initiates a protocol connection with the user equipment; Build) put Restart (restart) mark in the message.

Step 204, the BGP module of the user equipment performs graceful restart according to the received connection request message;

Since the user equipment is also configured with the BGP GR capability, and it cannot distinguish from the message whether it is gateway A or gateway B to establish a connection with it, when the BGP module of the gateway device initiates a new protocol connection, a trigger The GR capability of the user equipment, so that the user equipment can smoothly establish a protocol neighbor with the gateway B device, and, during the establishment of the BGP protocol, the established forwarding items will not be deleted, ensuring that the uplink service of the user equipment will not be interrupted;

In order to solve the problem of long-term interruption of the flow returned from the bearer network, the method of the present invention is to establish an IBGP neighbor between the gateway active device and the gateway backup device, for example, the gateway active device is gateway A, and the gateway backup device is gateway B In this way, gateway A will forward the route learned from the user equipment to gateway B, and its next-hop address is set to not be modified, and it still points to the user equipment. In this way, gateway B also has the route published by the user equipment. When the state of the gateway device switches, gateway A does not immediately delete the routes learned from the BGP neighbor established with the user equipment, and gateway B does not immediately delete the routes to the user equipment learned from gateway A. The traffic returned from the network is forwarded to the user equipment to ensure that the business returning to the user equipment is not interrupted;

During the GR process between gateway B and the user equipment, gateway B will relearn the routes pointing to the user equipment. Gateway B prefers the routes learned from the user equipment and forwards them to gateway A through the IBGP neighbor. Gateway A prefers gateway B. and delete the routes directly learned from the user equipment and the corresponding routes sent by gateway B; At the same time, by extending the VRRP protocol message or other protocol messages such as IBGP, the gateway A is notified that the GR is over, and the gateway A deletes all the routes directly obtained from the user equipment. All the routes on gateway A are directly obtained from the BGP neighbor established with the user equipment, and all the routes directed to the user equipment on gateway A are learned from the IBGP neighbor established with gateway B. In this process, neither gateway A nor gateway B actively deletes the forwarding items directed to the user equipment, but uses a combination of update and deletion technology to effectively ensure that the traffic returning to the user equipment will not be interrupted due to the switching of the gateway equipment status . When the user is not sensitive to the interruption time of the traffic returned by the bearer network, or uses other technologies such as deploying static routes or IBGP on gateway A and gateway B to guide the forwarding of traffic returned from the bearer network, between gateway A and gateway B It is not necessary to deploy the IBGP protocol among them.

In addition, from the perspective of GR, it is generally required that the source port number of the newly established connection is different from the port number of the previous connection, otherwise TCP may ignore or mishandle the new connection request, resulting in GR failure or GR recovery time. In order to solve this problem, the active gateway device can notify the standby gateway device of the source port number used by the BGP neighbor established with the user equipment through the extended VRRP protocol or other protocols such as IBGP. In this way, after gateway B becomes the new active gateway device , when establishing a BGP neighbor relationship with the user equipment, this port number is purposely not used as the source port number.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person familiar with the technology can easily think of changes or replacements within the technical scope disclosed in the present invention. , should be covered within the protection scope of the present invention.

Claims (12)

1. A method for business recovery based on active and standby gateway device state switching, characterized in that, comprising the steps: A. When a Border Gateway Protocol (BGP) connection is established between the user equipment and the gateway device, the gateway device uses a virtual redundant routing protocol (VRRP) virtual address as the source address of the BGP protocol; B. After the switchover of the active and standby state of the gateway device, the new gateway active device establishes a Border Gateway Protocol (BGP) connection with the user equipment in a graceful restart (GR) manner to ensure smooth establishment of the user equipment and the new gateway active device protocol connection. 2. The method according to claim 1, wherein the step B specifically comprises adding GR technology to the BGP protocol to expand the functions of the BGP protocol, so that the GR technology can be realized between the user equipment and the gateway equipment. 3. The method according to claim 2, characterized in that comprising a VRRP module and a BGP module, and said step B specifically includes, B1. After the gateway device status switch, the VRRP module in the new gateway master device notifies the gateway device status switch message to the BGP module of the gateway device; B2. After the BGP module of the gateway device receives the gateway device state switching message sent by the new gateway master device, it actively sends a connection request message of the BGP protocol to the BGP module of the user device; B3. The BGP module of the user equipment processes the received BGP connection request message according to the GR technology. 4. the method according to right 3, it is characterized in that, described B2 specifically comprises that the BGP module of gateway equipment puts restarting (Restart) mark in neighbor establishment message (OPEN), initiates the protocol connection with user equipment request message. 5. The method according to claim 1, characterized in that, after step B, further comprising: C. Establish an IBGP protocol connection between the new gateway master device and the previous gateway master device, so that both the new gateway master device and the previous gateway master device have routes published by the user equipment; D. After the state of the gateway device is switched, the new gateway master device and the previous gateway master device do not actively delete the route learned by the previous gateway master device from the user equipment, and the new gateway master device uses the route to Guide the forwarding of the traffic coming back from the bearer network to the user equipment, so as to ensure that the service returning to the user equipment is not interrupted. 6. method according to claim 5, is characterized in that, described step D specifically comprises, D1. During the GR process between the new gateway master device and the user equipment, the new gateway master device will relearn from the user equipment the route from the previous gateway master device pointing to the user equipment; D2. The new gateway master device forwards the route to the previous gateway master device through the IBGP protocol, and the previous gateway master device retains the received route and deletes its slave user according to the received route The corresponding route directly learned by the device; D3. After the GR between the new gateway master device and the user equipment ends, the new gateway master device deletes all the routes pointing to the user equipment learned from the previous gateway master device, and at the same time, the new gateway master device sends the previous The gateway active device notifies the end of GR, and the previous gateway active device deletes all the routes obtained directly from the user equipment. 7. The method according to claim 6, characterized in that, in step D3, the new gateway master device notifies the previous gateway master device that the GR is over through an extended VRRP protocol message. 8. The method according to claim 1, characterized in that, in order to avoid GR failure or GR recovery time becoming longer, the source port number of the connection established by the new gateway master device is the same as the port connected to the previous gateway master device number is different. 9. The method according to claim 8, wherein when the gateway master device establishes a connection with the user equipment, the gateway master device notifies the gateway backup device of the source used by it to establish a BGP neighbor relationship with the user equipment through the extended VRRP protocol. The port number. 10. A system based on active and standby gateway device state switching for service recovery, characterized in that it includes a gateway device, a user device, a BGP module, a VRRP module, a VRRP switching perception module and a switching notification module, A virtual redundant routing protocol is run between the gateway devices through a VRRP module; Both the gateway device and the user equipment are configured with a graceful restart (GR) technology, and a Border Gateway Protocol connection is established through a BGP module; The switching notification module is used to notify the gateway device state switching message to the VRRP switching perception module; The VRRP switching perception module initiates a BGP protocol connection request message to the BGP module of the user equipment according to the received gateway device state switching message; The BGP module of the user equipment performs graceful restart (GR) according to the received BGP protocol connection request message. 11. The system according to claim 10, wherein the switching notification module is embedded in the VRRP module. 12. The system according to claim 10, wherein the VRRP switching perception module is embedded in the BGP module of the gateway device.