CN108337118B

CN108337118B - Message forwarding method and device

Info

Publication number: CN108337118B
Application number: CN201810098794.8A
Authority: CN
Inventors: 汪欢
Original assignee: Hangzhou H3C Technologies Co Ltd
Current assignee: Hangzhou H3C Technologies Co Ltd
Priority date: 2018-01-31
Filing date: 2018-01-31
Publication date: 2021-01-26
Anticipated expiration: 2038-01-31
Also published as: CN108337118A

Abstract

The disclosure relates to a message forwarding method and device. The method comprises the following steps: detecting a tunnel state of the third VXLAN tunnel through the BFD session; when the tunnel state of the third VXLAN tunnel is detected to be changed from effective UP to ineffective Defect through the BFD session, the tunnel state of the first VXLAN tunnel is set to be ineffective Defect, so that the aggregation switch sets the use state of the second VXLAN tunnel to be Active when the aggregation switch detects that the tunnel state of the first VXLAN tunnel is changed from effective UP to ineffective Defect, thereby realizing that in the process that the aggregation switch provides service support for user equipment through DC1, if the VXLAN tunnel between the server in DC1 and the resource pool gateway has a fault, the aggregation switch can sense the fault of the VXLAN tunnel, automatically switch the service message from DC1 to DC2, and ensure the normal forwarding of the service message.

Description

Message forwarding method and device

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to a tunnel switching method and apparatus.

Background

In the related art, a network of a Spine-Leaf architecture (Spine-Leaf) is widely used in a metropolitan area network DC (Data Center) resource Pool, and under the architecture, a virtual Broadband Remote Access Server (vrbras, english: Virtualized Broadband Access Server, abbreviated as vbars) may be directly connected to a Pool-GW (Pool Gateway). VXLAN tunnel is established between vBRAS server and Pool-GW, and VXLAN tunnel is established between HJSW (HuiJu Switch) and Pool-GW. In a networking with two DCs as backups, two VXLAN tunnels configured in the HJSW uplink direction point to the Pool-GW of the two DCs respectively.

Fig. 1 shows a schematic diagram of a networking in which two DCs backup each other in the related art. As shown in fig. 1, when the DC1 works normally, the VSI (Virtual Switch Instance) of Pool-GW2 in the DC2 is manually turned off, so that the service support is provided for the user equipment through the DC 1. When a VXLAN tunnel between a vBRAS server and Pool-GW in DC1 fails, the problem that service messages cannot be switched to DC2 in time and service message forwarding is interrupted due to the fact that the HJSW cannot sense the failure of the VXLAN tunnel at the moment.

Disclosure of Invention

In view of this, the present disclosure provides a message forwarding method and apparatus, so as to solve the problem that service message forwarding is interrupted due to that a service message cannot be switched between multiple DC resource pools in time.

According to an aspect of the present disclosure, a method for forwarding a packet is provided, where a current use state of a first VXLAN tunnel is Active, a current use state of a second VXLAN tunnel is Inactive, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the method is used for a first resource pool gateway and comprises the following steps:

detecting a tunnel state of a third VXLAN tunnel through a BFD session, wherein the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server;

when detecting that the tunnel state of the third VXLAN tunnel is changed from effective UP to ineffective Defect through a BFD session, setting the tunnel state of the first VXLAN tunnel to ineffective Defect, so that the aggregation switch sets the use state of the second VXLAN tunnel to Active when detecting that the tunnel state of the first VXLAN tunnel is changed from effective UP to ineffective Defect.

According to another aspect of the present disclosure, a message forwarding method is provided, where a current use state of a first VXLAN tunnel is Active, a current use state of a second VXLAN tunnel is Inactive, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the method is used for a convergence switch and comprises the following steps:

detecting a tunnel state of the first VXLAN tunnel through the BFD session;

when detecting that the tunnel state of the first VXLAN tunnel is changed from effective UP to ineffective Defect through a BFD session, setting the use state of the second VXLAN tunnel to Active;

and forwarding the service message through the second VXLAN tunnel.

According to another aspect of the present disclosure, a message forwarding apparatus is provided, where a current use state of a first VXLAN tunnel is Active, a current use state of a second VXLAN tunnel is Inactive, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the apparatus is for a first resource pool gateway, comprising:

a first detection module, configured to detect a tunnel state of a third VXLAN tunnel through a BFD session, where the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server;

a first setting module, configured to set the tunnel state of the first VXLAN tunnel to invalid when detecting that the tunnel state of the third VXLAN tunnel is changed from valid UP to invalid, so that the aggregation switch sets the usage state of the second VXLAN tunnel to Active when detecting that the tunnel state of the first VXLAN tunnel is changed from valid UP to invalid.

According to another aspect of the present disclosure, a message forwarding apparatus is provided, where a current use state of a first VXLAN tunnel is Active, a current use state of a second VXLAN tunnel is Inactive, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the apparatus is for a converged switch, comprising:

the second detection module is used for detecting the tunnel state of the first VXLAN tunnel through the BFD session;

a second setting module, configured to set, when detecting that a tunnel state of the first VXLAN tunnel is changed from Active UP to invalid default through a BFD session, a use state of the second VXLAN tunnel to Active;

and the message forwarding module is used for forwarding the service message through the second VXLAN tunnel.

According to another aspect of the present disclosure, a message forwarding method is provided, where the method is used in a dual DC system, where the dual DC system includes an aggregation switch, a first resource pool gateway, a second resource pool gateway, and a first server, a first VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the first resource pool gateway, a second VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the second resource pool gateway, and a third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server; the method comprises the following steps:

the first resource pool gateway detects the tunnel state of the third VXLAN tunnel through a BFD session;

when detecting that the tunnel state of the third VXLAN tunnel is changed from effective UP to ineffective Defect through a BFD session, the first resource pool gateway sets the tunnel state of the first VXLAN tunnel to ineffective Defect;

the convergence switch detects the tunnel state of the first VXLAN tunnel through a BFD session;

when detecting that the tunnel state of the first VXLAN tunnel is changed from effective UP to ineffective Defect through a BFD session, the aggregation switch sets the use state of the second VXLAN tunnel to Active;

and the aggregation switch forwards the service message through the second VXLAN tunnel.

The message forwarding method and device disclosed by the invention provide a tunnel monitoring mode, and tunnel monitoring refers to a technology for keeping consistent tunnel states of a plurality of VXLAN tunnels belonging to the same monitoring tunnel group. Therefore, in the process that the aggregation switch provides service support for the user equipment through the DC1, if a VXLAN tunnel between a server in the DC1 and a resource pool gateway has a fault, the aggregation switch can sense the fault of the VXLAN tunnel and automatically switch a service message from the DC1 to the DC2, so that normal forwarding of the service message is ensured.

Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

Fig. 1 shows a schematic diagram of a networking in which two DCs backup each other in the related art.

Fig. 2a shows a schematic diagram of a dual DC networking according to an embodiment of the present disclosure.

Figure 2b shows a schematic diagram of a dual DC networking according to an embodiment of the present disclosure.

Fig. 2c shows a schematic diagram of a dual DC networking according to an embodiment of the present disclosure.

Fig. 2d shows a schematic diagram of a dual DC networking according to an embodiment of the present disclosure.

Fig. 2e shows a schematic diagram of a dual DC networking according to an embodiment of the present disclosure.

Fig. 2f shows a schematic diagram of a dual DC networking according to an embodiment of the present disclosure.

Fig. 3 shows a flowchart of a packet forwarding method according to an embodiment of the present disclosure.

Fig. 4 shows a flowchart of a packet forwarding method according to an embodiment of the present disclosure.

Fig. 5 shows a flowchart of a packet forwarding method according to an embodiment of the present disclosure.

Fig. 6 shows a block diagram of a message forwarding device according to an embodiment of the present disclosure.

Fig. 7 shows a block diagram of a message forwarding device according to an embodiment of the present disclosure.

Fig. 8 is a block diagram illustrating a message forwarding device 900 according to an example embodiment.

Detailed Description

Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers can indicate functionally identical or similar elements. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

Furthermore, in the following detailed description, numerous specific details are set forth in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements and circuits that are well known to those skilled in the art have not been described in detail so as not to obscure the present disclosure.

In the related art, BFD (Bidirectional Forwarding Detection) provides a general, standardized, media independent, and protocol independent fast failure Detection mechanism. BFD may quickly detect failures of the bidirectional forwarding path between two devices for various upper layer protocols (e.g., routing protocols, VXLAN protocols). After establishing the neighbor relation, the upper layer protocol informs the BFD of the neighbor parameters and the detection parameters (including destination address, source address and the like), and the BFD establishes a BFD session according to the received parameters. When the network fails, the BFD removes the BFD session after detecting the network failure and informs the upper layer protocol neighbor of being unreachable.

In the message forwarding method disclosed by the invention, BFD (bidirectional forwarding detection) control messages are periodically sent between devices at two ends of a VXLAN (virtual extensible local area network) tunnel to establish a BFD session, and the tunnel state of the VXLAN tunnel is detected through the BFD session.

In order to make those skilled in the art better understand the technical solutions in the embodiments of the present invention, a brief description will be given below of some technical terms involved in the embodiments of the present invention.

The state of the BFD session may include: establish UP or turn Down.

The tunnel status of the VXLAN tunnel may include: active UP or inactive delete.

When the tunnel status of the VXLAN tunnel is valid UP, the usage status of the VXLAN tunnel may include: active or Inactive.

If the BFD control message can be normally received and sent between the devices, the state of the BFD session between the devices is to establish UP. If the state of the BFD session between the devices is to establish UP, it can be said that the tunnel state of the VXLAN tunnel between the devices is valid UP.

And if the BFD control message cannot be normally received and transmitted between the devices, the state of the BFD session between the devices is closing Down. If the state of the BFD session between the devices is Down, it can be described that the tunnel state of the VXLAN tunnel between the devices is invalid.

It can be understood that whether the devices can normally transmit and receive the BFD control packet, the state of the BFD session between the devices, and the tunnel state of the VXLAN tunnel have an association relationship, and they are mutually influenced.

When the tunnel status of the VXLAN tunnel is valid UP, the usage status of the VXLAN tunnel with higher Priority (PRI) may be Active, and the usage status of the VXLAN tunnel with lower Priority may be Inactive.

Further, the service message can be set to be forwarded preferentially through the VXLAN tunnel which activates Active. When the convergence switch detects that the tunnel state of the VXLAN tunnel in the Active state is changed from Active UP to invalid Defect, the service message can be switched to the VXLAN tunnel which is newly changed into the Active state for forwarding, so that automatic switching between the VXLAN tunnels is realized, and normal forwarding of the service message is ensured.

After the tunnel status of the Active VXLAN tunnel is changed from Active UP to Inactive, the usage status of the VXLAN tunnel with higher priority among the Inactive VXLAN tunnels is changed from Inactive Active to Active.

It should be noted that, as those skilled in the art can understand, there are various reasons for causing the tunnel status of the VXLAN tunnel to be invalid. For example, VXLAN tunnel becomes invalid after failure. As another example, the VXLAN tunnel is set to invalid delete. There may be various reasons for VXLAN tunnel failure, such as device problem, link problem, or policy configuration problem, which is not limited by this disclosure.

It should be noted that, when the tunnel state of the VXLAN tunnel is valid UP, it indicates that the VXLAN tunnel can forward the service packet. When a plurality of VXLAN tunnels with effective UP tunnel states exist, the plurality of VXLAN tunnels can all forward the service message. Therefore, for the VXLAN tunnel with the tunnel state of valid UP, the significance of distinguishing the Active tunnel from the Inactive Inactive tunnel is as follows: whether a VXLAN tunnel is adopted to forward the service message or not.

When the tunnel state of the VXLAN tunnel is invalid, the VXLAN tunnel cannot forward the service message at all, so that the Active and Inactive are not distinguished for the VXLAN tunnel with the tunnel state of invalid, but the Active and Inactive are distinguished for the VXLAN tunnel with the tunnel state of valid UP.

In one implementation, a redundant-Tunnel (redundant-Tunnel) may refer to a technique in which multiple VXLAN tunnels belonging to the same redundant Tunnel group are backup VXLAN tunnels to each other. The setting of the use state of the VXLAN tunnel can be realized through the redundant tunnel.

Wherein, the state of the VXLAN tunnel in the redundant tunnel group is the use state of the VXLAN tunnel. The status of the VXLAN tunnel in the redundant tunnel group may thus include: active or Inactive. When the tunnel state of the VXLAN tunnel is effective UP, the VXLAN tunnel with higher priority in the redundant tunnel group is Active, and the VXLAN tunnel with lower priority is Inactive.

Fig. 2 a-2 f show schematic diagrams of a dual DC networking according to an embodiment of the present disclosure. The dual DC networking shown in fig. 2a to 2f may be used to exemplify the flow diagrams shown in fig. 3, 4 and 5.

As shown in fig. 2a, the DC resource Pool 1 includes a first resource Pool gateway Pool-GW1 and a first Server 1. The DC resource Pool 2 includes a second resource Pool gateway Pool-GW2 and a second Server 2. The DC resource pool 1 and the DC resource pool 2 are connected by HJSW. A first VXLAN Tunnel1 is established between HJSW and Pool-GW1, a second VXLAN Tunnel2 is established between HJSW and Pool-GW2, a third VXLAN Tunnel3 is established between Pool-GW1 and Server1, and a fourth VXLAN Tunnel4 is established between Pool-GW2 and Server 2.

The first server and the second server may be VBARS servers, the VBRAS servers may include virtual machines for processing various service types, and the service types may include an ITMS (Integrated Terminal Management System), an IPTV (Internet Protocol television), a PPPOE (Point to Point Protocol over Ethernet ), a VOIP (Voice over Internet Protocol, network phone), and the like.

In this embodiment, the Pool-GW has no difference in function from the conventional gateways in digital communication, and is connected to a server carrying multiple virtual machines as an egress device for connecting the virtual machines to other networks. The software and hardware facilities provided for the plurality of virtual machines can be understood as a resource pool, and the gateway is named as a resource pool gateway in the present embodiment to correspond to the concept of the resource pool.

As shown in fig. 2a, a BFD session is established between the HJSW and Pool-GW1, and a Tunnel state of Tunnel1 is detected through the BFD session. And establishing a BFD session between the HJSW and the Pool-GW2, and detecting the Tunnel state of the Tunnel2 through the BFD session. A BFD session is established between Pool-GW1 and Server1, and a Tunnel status of Tunnel3 is detected through the BFD session. And establishing a BFD session between the Pool-GW2 and the Server2, and detecting the Tunnel state of the Tunnel4 through the BFD session.

In the related art, as shown in fig. 1, when a VXLAN tunnel between a server in DC1 and a Pool-GW fails, at this time, the HJSW cannot sense the failure of the VXLAN tunnel, and needs to manually determine whether the VXLAN tunnel fails. And when the VXLAN tunnel fault is judged manually, the VSI of the Pool-GW2 in the DC1 is closed manually, and the VSI of the Pool-GW2 in the DC2 is opened, so that the service message can be switched to the DC 2. In the process, a special network administrator is needed to judge and modify the VXLAN tunnel fault, so that the operation is complicated and inconvenient. In addition, the service message cannot be switched into the DC2 in time, which causes the problem of service message forwarding interruption.

In the following packet forwarding method of the present disclosure, a way of monitoring a Tunnel (Monitor-Tunnel) is provided to solve the above problem, and Tunnel monitoring may refer to a technology in which Tunnel states of a plurality of VXLAN tunnels belonging to the same monitoring Tunnel group are kept consistent. The tunnel state of the VXLAN tunnel can be set through tunnel monitoring.

The monitoring tunnel can realize the linkage of the monitoring of the uplink VXLAN tunnel and the tunnel state of the downlink VXLAN tunnel. Specifically, fig. 3 shows a flowchart of a message forwarding method according to an embodiment of the present disclosure. The current use state of the first VXLAN tunnel is Active, the current use state of the second VXLAN tunnel is Inactive Inactive, the first VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the second resource pool gateway. The method is used for establishing a tunnel resource pool gateway with a convergence switch, and the state of establishing a tunnel between the resource pool gateway and the convergence switch is Active. In this embodiment, because the first VXLAN tunnel is currently Active, and the first VXLAN tunnel in Active is a tunnel established between the first resource pool gateway and the aggregation switch, the method in this embodiment is described by taking application to the first resource pool gateway as an example, and as shown in fig. 3, the message forwarding method includes:

in step S31, a tunnel status of a third VXLAN tunnel is detected through the BFD session, the third VXLAN tunnel being a VXLAN tunnel between the first resource pool gateway and the first server.

In step S32, upon detecting that the tunnel status of the third VXLAN tunnel is changed from Active UP to inactive delete through the BFD session, the tunnel status of the first VXLAN tunnel is set to inactive delete, so that the aggregation switch sets the usage status of the second VXLAN tunnel to Active upon detecting that the tunnel status of the first VXLAN tunnel is changed from Active UP to inactive delete.

As an example, as shown in fig. 2b, Pool-GW1 detects the Tunnel status of Tunnel3 through a BFD session. Tunnel3 fails, whereby Pool-GW1 detects that the Tunnel status of Tunnel3 changes from active UP to inactive default through a BFD session, and since Tunnel monitoring is configured between Tunnel1 and Tunnel3 in Pool-GW1, Pool-GW1 sets the Tunnel status of Tunnel1 to inactive default. Thus HJSW can detect through the BFD session that the Tunnel status of Tunnel1 changes from Active UP to inactive delete, then HJSW sets the usage status of Tunnel2 to Active.

In the message forwarding method disclosed by the invention, the Tunnel state of the Tunnel1 is linked with the Tunnel state of the Tunnel3 through Tunnel monitoring, so that if the Tunnel state of the Tunnel3 is changed from effective UP to ineffective Defect, the Tunnel state of the Tunnel1 is changed into ineffective Defect along with the Tunnel3, thereby realizing that HJSW can sense the fault of the remote Tunnel3, switching the service message to DC2 in time for forwarding, and ensuring the normal forwarding of the service message.

It should be noted that, after the use state of Tunnel2 is set to Active, the second resource pool gateway may still implement Tunnel linkage by the method provided in this embodiment.

In one implementation, detecting a tunnel status of the third VXLAN tunnel through the BFD session includes: sending a BFD control message to a first server at intervals of a first time; when a response message returned by the first server based on the BFD control message is received within a second time length, determining that the tunnel state of the third VXLAN tunnel is effective UP; or when a response message returned by the first server based on the BFD control message is not received within the second time, determining that the tunnel state of the third VXLAN tunnel is invalid.

The first time period may be a sending period of the BFD control packet, for example, 500 milliseconds. The second duration BFD controls a receiving period of the packet, for example, 800 ms. It is understood that the second duration is greater than or equal to the first duration.

As an example, as shown in fig. 2a, Pool-GW1 sends BFD control messages to Server1 every 500 milliseconds, and Server1 sends BFD control messages to Pool-GW1 every 500 milliseconds. Therefore, if the Pool-GW1 receives the BFD control packet from the Server1 within 800 ms, the Pool-GW1 determines that the Tunnel status of the Tunnel3 is valid UP. If the Pool-GW1 does not receive the BFD control message from the Server1 within 800 ms, the Pool-GW1 determines that the Tunnel state of the Tunnel3 is invalid.

In one implementation, after detecting that the tunnel status of the third VXLAN tunnel is changed from active UP to inactive default through the BFD session, the method further comprises: detecting a tunnel state of the third VXLAN tunnel through the BFD session; and when the tunnel state of the third VXLAN tunnel is detected to be changed from invalid Defect to valid UP through the BFD session, setting the tunnel state of the first VXLAN tunnel to be valid UP.

As an example, as shown in fig. 2d, after the Pool-GW1 detects that the Tunnel state of Tunnel3 changes from active UP to invalid default through a BFD session, the Pool-GW1 sends BFD control packets to the Server1 every time a preset number (for example, 5) of BFD control packet sending cycles pass. If Pool-GW1 receives the BFD control packet sent by Server1, Pool-GW1 determines that the Tunnel state of Tunnel3 is changed from invalid default to valid UP, and Pool-GW1 sets the Tunnel state of Tunnel1 to valid UP because Tunnel monitoring is configured between Tunnel1 and Tunnel3 in Pool-GW 1.

In one implementation, when detecting that the tunnel status of the third VXLAN tunnel is changed from invalid default to valid UP through the BFD session, setting the tunnel status of the first VXLAN tunnel to valid UP includes: and when the tunnel state of the third VXLAN tunnel is detected to be changed from invalid Defect to valid UP through the BFD session and the valid UP is kept to exceed the third time length, setting the tunnel state of the second VXLAN tunnel to be valid UP.

The message forwarding method provided by this embodiment reduces frequent switching of the tunnel state by using a delayed switchback manner. The delayed switchback may refer to that after the tunnel state of the uplink VXLAN tunnel in the tunnel monitoring group is changed from invalid default to valid UP and maintained for a third duration, the tunnel state of the downlink VXLAN tunnel is changed to valid UP along with the invalid default. Wherein the third time length is the back-off delay in the tunnel monitoring, and is, for example, 5 seconds.

It can be understood that the delayed switchback can avoid frequent change of the tunnel state of the downlink VXLAN tunnel due to frequent change of the tunnel state of the uplink VXLAN tunnel in the tunnel monitoring group, and reduce network oscillation.

As an example, as shown in fig. 2d, after the Pool-GW1 detects that the Tunnel state of Tunnel3 changes from active UP to invalid default through a BFD session, the Pool-GW1 sends BFD control packets to the Server1 every time a preset number (for example, 5) of BFD control packet sending cycles pass. If Pool-GW1 receives the BFD control packet sent by Server1, Pool-GW1 determines that the Tunnel state of Tunnel3 is changed from invalid default to valid UP. When the Tunnel state of Tunnel3 is an effective UP and the effective UP is maintained for more than 5 seconds, Pool-GW1 sets the Tunnel state of Tunnel1 to be an effective UP since Tunnel monitoring is configured between Tunnel1 and Tunnel3 in Pool-GW 1.

In the message forwarding method disclosed by the invention, the Tunnel state of the Tunnel1 is linked with the Tunnel state of the Tunnel3 through Tunnel monitoring, so that the HJSW can sense the fault of the remote Tunnel3 and can sense the fault repair of the remote Tunnel3, support is provided for the HJSW to switch forwarding of service messages in time, and the normal forwarding of the service messages is ensured.

Fig. 4 shows a flowchart of a packet forwarding method according to an embodiment of the present disclosure. The current use state of the first VXLAN tunnel is Active, the current use state of the second VXLAN tunnel is Inactive Inactive, the first VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the second resource pool gateway. The method is used for an aggregation switch. As shown in fig. 4, the message forwarding method includes:

in step S41, a tunnel status of the first VXLAN tunnel is detected through the BFD session.

In step S42, upon detecting that the tunnel status of the first VXLAN tunnel is changed from Active UP to inactive delete through the BFD session, the usage status of the second VXLAN tunnel is set to Active.

In step S43, the service packet is forwarded through the second VXLAN tunnel.

As an example, as shown in fig. 2b, the HJSW detects the Tunnel status of Tunnel1 through the BFD session.

In one implementation, when the tunnel status of the first VXLAN tunnel is changed from active UP to inactive default, which is detected by the BFD session, two cases may be included:

one situation is: the tunnel state of the first VXLAN tunnel is invalid Defect, which is set for the first VXLAN tunnel by the first resource pool gateway when the tunnel state of the third VXLAN tunnel is invalid Defect, and the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server.

As an example, as shown in fig. 2b, Pool-GW1 detects the Tunnel status of Tunnel3 through a BFD session. Tunnel3 fails, whereby Pool-GW1 detects that the Tunnel status of Tunnel3 changes from active UP to inactive default through a BFD session, and since Tunnel monitoring is configured between Tunnel1 and Tunnel3 in Pool-GW1, Pool-GW1 sets the Tunnel status of Tunnel1 to inactive default. The HJSW detects the Tunnel status of Tunnel1 through the BFD session. Thus HJSW detects that the Tunnel state of Tunnel1 changes from Active UP to inactive delete through the BFD session, then HJSW sets the usage state of Tunnel2 to Active. And the HJSW forwards the service message through the Tunnel2 in the Active state.

Another case is: the tunnel status of the first VXLAN tunnel being invalid Defect is caused by a failure of the first VXLAN tunnel itself. For example, Tunnel1 failed, whereby the HJSW detects through the BFD session that the Tunnel status of Tunnel1 changed from Active UP to inactive delete, the HJSW sets the usage status of Tunnel2 to Active. And the HJSW forwards the service message through the Tunnel2 in the Active state.

In one implementation, when the aggregation switch detects that the tunnel status of the first VXLAN tunnel is changed from valid UP to invalid default through the BFD session, the priority of the first VXLAN tunnel or the second VXLAN tunnel is adjusted so that the priority of the first VXLAN tunnel is smaller than the priority of the second VXLAN tunnel.

It can be understood that, as shown in fig. 2b, when the HJSW detects that the Tunnel status of Tunnel1 changes from active UP to invalid default through the BFD session, Tunnel2 becomes the higher priority VXLAN Tunnel in active UP. The use state of Tunnel2 is changed from Inactive to Active. And the HJSW forwards the service message through the Tunnel2 in the Active state.

As shown in fig. 2b, if HJSW keeps the priorities of Tunnel1 and Tunnel2 unchanged, when the Tunnel state of Tunnel1 is restored from invalid default to valid UP, since the priority of Tunnel2 is lower than that of Tunnel1, the use state of Tunnel1 is restored to Active, and the use state of Tunnel2 is restored to Inactive. If the Tunnel states of Tunnel1 are frequently switched, the service messages are frequently switched between DC1 and DC2, which causes network oscillation.

As shown in fig. 2c, when the HJSW detects that the Tunnel state of Tunnel1 changes from Active UP to Inactive delete through the BFD session, the HJSW adjusts the priority of Tunnel1 or Tunnel2, so that the priority of Tunnel2 is greater than the priority of Tunnel1, thereby when the Tunnel state of Tunnel1 is restored from Inactive delete to Active UP, since the priority of Tunnel1 is smaller than the priority of Tunnel2, the use state of Tunnel1 is restored to Inactive while the use state of Tunnel2 is still maintained to Active, and the HJSW still performs service packet forwarding through Tunnel2 which is Active, thereby reducing network oscillation.

It should be noted that, specifically, the priority of Tunnel1 or Tunnel2 may be set according to an actual application scenario by adjusting the priority, as long as the highest priority of the VXLAN Tunnel for forwarding the service packet is ensured, and the method for setting the priority is not limited by the present disclosure.

In the message forwarding method disclosed by the invention, the Tunnel1 and the Tunnel2 are mutually backed UP through the redundant Tunnel, the message forwarding is carried out through the Tunnel1 which activates the Active under the default condition, and when the Tunnel state of the Tunnel1 is changed from the effective UP to the ineffective Defect, the using state of the Tunnel2 is set to the activated Active by the HJSW, so that the HJSW carries out message forwarding through the Tunnel2 which newly activates the Active, the HJSW can timely switch the forwarding of the service message, and the normal forwarding of the service message is ensured.

Fig. 5 shows a flowchart of a packet forwarding method according to an embodiment of the present disclosure. The method is used in a double-DC system, the double-DC system comprises an aggregation switch, a first resource pool gateway, a second resource pool gateway and a first server, a first VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the first resource pool gateway, a second VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the second resource pool gateway, and a third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server. As shown in fig. 5, the message forwarding method includes:

in step S51, the first resource pool gateway detects a tunnel status of the third VXLAN tunnel through the BFD session.

In step S52, upon detecting through the BFD session that the tunnel status of the third VXLAN tunnel is changed from active UP to inactive delete, the first resource pool gateway sets the tunnel status of the first VXLAN tunnel to inactive delete.

In step S53, the aggregation switch detects the tunnel status of the first VXLAN tunnel through the BFD session.

In step S54, upon detecting through the BFD session that the tunnel status of the first VXLAN tunnel changes from Active UP to inactive default, the aggregation switch sets the usage status of the second VXLAN tunnel to Active.

In step S55, the aggregation switch forwards the service packet through the second VXLAN tunnel.

As shown in fig. 2a, when initialized in HJSW, may be configured with: { VSI VPN 1333, VXLAN 1333, Reduntant-Tunnel 1, Reduntant-Tunnel 1: tunnel1(PRI 100, Active), UP; tunnel2(PRI 10, Inactive), UP }. In the configuration described above, the VSI in which the aggregation switch is configured, and the belonging VXLAN, reductant-Tunnel 1 (first redundant Tunnel group) include

tunnels

1 and 2. Wherein, the Tunnel states of Tunnel1 and Tunnel2 are both effective UP. The priority of Tunnel1 is higher, so the usage status is Active. The priority of Tunnel2 is low, and the use status is Inactive. And the HJSW forwards the service message through the Tunnel1 in the Active state.

As shown in fig. 2a, when initialized in Pool-GW1, may be configured with: { Monitor Tunnel 1: tunnel1, Tunnel3 }. In the configuration described above, Monitor Tunnel1 (first monitoring Tunnel group) includes

tunnels

1 and 3. Thus, when the Tunnel state of Tunnel3 changes from active UP to inactive delete, Pool-GW1 sets the Tunnel state of Tunnel1 to inactive delete. If the Tunnel state of Tunnel3 is changed from invalid default to valid UP, Pool-GW1 sets the Tunnel state of Tunnel1 to valid UP. The Tunnel state of the Tunnel1 is linked with the Tunnel state of the Tunnel 3.

As shown in fig. 2a, when initialized in Pool-GW2, may be configured with: { Monitor Tunnel 2: tunnel2, Tunnel4 }. In the configuration described above, Monitor Tunnel2 (second monitoring Tunnel group) includes Tunnel2 and Tunnel 4. If the Tunnel state of Tunnel4 is changed from active UP to inactive delete, Pool-GW2 sets the Tunnel state of Tunnel2 to inactive delete. If the Tunnel state of Tunnel4 is changed from invalid default to valid UP, Pool-GW2 sets the Tunnel state of Tunnel2 to valid UP. The Tunnel state of the Tunnel2 is linked with the Tunnel state of the Tunnel 4.

As shown in fig. 2b, Pool-GW1 detects the Tunnel status of Tunnel3 through the BFD session. Tunnel3 fails, whereby Pool-GW1 detects that the Tunnel status of Tunnel3 changes from active UP to inactive default through a BFD session, since in Pool-GW1 { Monitor Tunnel 1: tunnel1, Tunnel3}, then Pool-GW1 sets the Tunnel status of Tunnel1 to invalid default. The HJSW detects the Tunnel status of Tunnel1 through the BFD session. Thus, if the HJSW detects that the Tunnel state of Tunnel1 changes from active UP to invalid delete through the BFD session, only Tunnel2 in the Reductant-Tunnel 1 set in the HJSW is active UP. Tunnel2 becomes the higher priority among VXLAN tunnels for active UP. The use state of Tunnel2 is changed from Inactive to Active. And the HJSW forwards the service message through the Tunnel2 in the Active state.

For example, as shown in fig. 2c, in HJSW may be configured: { VSI VPN 1333, VXLAN 1333, Reduntant-Tunnel 1, Reduntant-Tunnel 1: tunnel1(PRI 100), Defect; tunnel2(PRI 105, Active), UP }. The Tunnel state of Tunnel1 is invalid, and there is no use state. The Tunnel state of Tunnel2 is Active UP and the use state is Active.

As shown in fig. 2d, in the case of fault repair of Tunnel3, the Tunnel status of Tunnel3 is changed from invalid default to valid UP, and since the Pool-GW1 is configured with { Monitor Tunnel 1: tunnel1, Tunnel3}, Pool-GW1 sets the Tunnel state of Tunnel1 to active UP. When the Tunnel state of Tunnel1 becomes active UP, the use state of Tunnel1 is restored to Inactive since the priority of Tunnel1 is lower than the priority of Tunnel 2.

For example, as shown in fig. 2d, in HJSW may be configured: { VSI VPN 1333, VXLAN 1333, Reduntant-Tunnel 1, Reduntant-Tunnel 1: tunnel1(PRI 100, Inactive), UP; tunnel2(PRI 105, Active), UP }. The Tunnel state of Tunnel1 is UP, and the active state is Inactive. The Tunnel state of Tunnel2 is Active UP and the use state is Active.

As shown in fig. 2e, Pool-GW2 detects the Tunnel status of Tunnel4 through the BFD session. Tunnel4 fails, whereby Pool-GW2 detects that the Tunnel status of Tunnel4 changes from active UP to inactive default through a BFD session, since in Pool-GW2 { Monitor Tunnel 2: tunnel2, Tunnel4}, then Pool-GW2 sets the Tunnel status of Tunnel2 to invalid default. The HJSW detects the Tunnel status of Tunnel2 through the BFD session. Thus, if the HJSW detects that the Tunnel state of Tunnel2 changes from active UP to invalid delete through the BFD session, only Tunnel1 in the Reductant-Tunnel 1 set in the HJSW is active UP. Tunnel1 becomes the higher priority among VXLAN tunnels for active UP. The use state of Tunnel1 is changed from Inactive to Active. And the HJSW forwards the service message through the Tunnel1 in the Active state.

As shown in fig. 2f, when the HJSW detects that the Tunnel state of Tunnel2 changes from Active UP to Inactive delete through the BFD session, the HJSW adjusts the priority of Tunnel1 or Tunnel2, so that the priority of Tunnel1 is greater than the priority of Tunnel2, thereby when the Tunnel state of Tunnel2 is restored from Inactive delete to Active UP, since the priority of Tunnel2 is smaller than the priority of Tunnel1, the use state of Tunnel2 is restored to Inactive while the use state of Tunnel1 is still maintained to Active, and the HJSW still performs service packet forwarding through Tunnel1 which is Active, thereby reducing network oscillation.

For example, as shown in fig. 2f, in HJSW may be configured: { VSI VPN 1333, VXLAN 1333, Reduntant-Tunnel 1, Reduntant-Tunnel 1: tunnel1(PRI 100, Active), UP; tunnel2(PRI ═ 8), Defect }. Wherein, the Tunnel state of Tunnel1 is Active UP, and the use state is Active. The Tunnel state of Tunnel2 is invalid Defect, and there is no use state.

In the message forwarding method disclosed by the invention, the Tunnel state of the Tunnel1 is linked with the Tunnel state of the Tunnel3 through Tunnel monitoring, so that the HJSW can sense the fault of the remote Tunnel3 and can also sense the fault repair of the remote Tunnel 3. The method comprises the steps that mutual backup of Tunnel1 and Tunnel2 is realized through a redundant Tunnel, message forwarding is carried out through Tunnel1 of Active activation under the default condition, when the Tunnel state of Tunnel1 is changed from effective UP to ineffective Defect, the using state of Tunnel2 is set to Active by HJSW, and therefore the HJSW carries out message forwarding through Tunnel2 of the newly activated Active, forwarding of service messages can be timely switched by the HJSW, and normal forwarding of the service messages is guaranteed.

Fig. 6 shows a block diagram of a message forwarding device according to an embodiment of the present disclosure. The current use state of a first VXLAN tunnel is an activated Active state, the current use state of a second VXLAN tunnel is an inactivated Inactive state, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the apparatus is for a first resource pool gateway. As shown in fig. 6, the packet forwarding apparatus includes:

a first detecting module 61, configured to detect a tunnel state of a third VXLAN tunnel through a BFD session, where the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server; a first setting module 62, configured to set the tunnel state of the first VXLAN tunnel to invalid when detecting that the tunnel state of the third VXLAN tunnel is changed from valid UP to invalid delete through a BFD session, so that the aggregation switch sets the usage state of the second VXLAN tunnel to Active when detecting that the tunnel state of the first VXLAN tunnel is changed from valid UP to invalid delete.

In one implementation, the first setting module 62 is further configured to: and when detecting that the tunnel state of the third VXLAN tunnel is changed from invalid Defect to valid UP through the BFD session, setting the tunnel state of the first VXLAN tunnel to be valid UP.

In one implementation, the first detection module 61 is configured to: sending a BFD control message to the first server at intervals of a first time length; when a response message returned by the first server based on the BFD control message is received within a second time length, determining that the tunnel state of the third VXLAN tunnel is effective UP; or when a response message returned by the first server based on the BFD control message is not received within a second time length, determining that the tunnel state of the third VXLAN tunnel is invalid.

In one implementation, the first setting module 62 is further configured to: and when the tunnel state of the third VXLAN tunnel is detected to be changed from invalid Defect to valid UP through the BFD session and the valid UP is kept to exceed a third time length, setting the tunnel state of the second VXLAN tunnel to be valid UP.

Fig. 7 shows a block diagram of a message forwarding device according to an embodiment of the present disclosure. The current use state of a first VXLAN tunnel is an activated Active state, the current use state of a second VXLAN tunnel is an inactivated Inactive state, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the apparatus is for a convergence switch. As shown in fig. 7, the packet forwarding apparatus includes:

a second detecting module 71, configured to detect a tunnel state of the first VXLAN tunnel through a BFD session; a second setting module 72, configured to set, when detecting that the tunnel status of the first VXLAN tunnel changes from Active UP to invalid default through a BFD session, a use status of the second VXLAN tunnel to Active; and the message forwarding module 73 is configured to forward the service message through the second VXLAN tunnel.

In one implementation, when detecting that the tunnel status of the first VXLAN tunnel changes from active UP to invalid default through a BFD session, the method includes: the tunnel state of the first VXLAN tunnel is invalid Defect, which is set for the first VXLAN tunnel by the first resource pool gateway when the tunnel state of the third VXLAN tunnel is invalid Defect, and the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server.

The message forwarding device provides a tunnel monitoring mode, and the tunnel monitoring refers to a technology for keeping consistent tunnel states of a plurality of VXLAN tunnels belonging to the same monitoring tunnel group. Therefore, in the process that the aggregation switch provides service support for the user equipment through the DC1, if a VXLAN tunnel between a server in the DC1 and a resource pool gateway has a fault, the aggregation switch can sense the fault of the VXLAN tunnel and automatically switch a service message from the DC1 to the DC2, so that normal forwarding of the service message is ensured.

Fig. 8 is a block diagram illustrating a message forwarding device 900 according to an example embodiment. Referring to fig. 8, the apparatus 900 may include a processor 901, a machine-readable storage medium 902 having stored thereon machine-executable instructions. The processor 901 and the machine-readable storage medium 902 may communicate via a system bus 903. And, the processor 901 executes the message forwarding method executed by the aggregation switch by reading the machine-executable instruction corresponding to the message forwarding logic in the machine-readable storage medium 902.

The present disclosure also provides another message forwarding device that may include a processor, and a machine-readable storage medium having stored thereon machine-executable instructions. The processor and the machine-readable storage medium may communicate via a system bus. And the processor reads the machine-executable instruction corresponding to the message forwarding logic in the machine-readable storage medium to execute the message forwarding method executed by the resource pool gateway.

A machine-readable storage medium as referred to herein may be any electronic, magnetic, optical, or other physical storage device that can contain or store information such as executable instructions, data, and the like. For example, the machine-readable storage medium may be: a RAM (random Access Memory), a volatile Memory, a non-volatile Memory, a flash Memory, a storage drive (e.g., a hard drive), a solid state drive, any type of storage disk (e.g., an optical disk, a dvd, etc.), or similar storage medium, or a combination thereof.

Having described embodiments of the present disclosure, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein were chosen in order to best explain the principles of the embodiments, the practical application, or technical improvements to the techniques in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A message forwarding method is characterized in that the current use state of a first VXLAN tunnel is an Active state, the current use state of a second VXLAN tunnel is an Inactive state, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the method is used for a first resource pool gateway and comprises the following steps:

when detecting that the tunnel state of the third VXLAN tunnel is changed from Active UP to invalid Defect through a BFD session, setting the tunnel state of the first VXLAN tunnel to invalid Defect, so that the aggregation switch sets the use state of the second VXLAN tunnel to Active when detecting that the tunnel state of the first VXLAN tunnel is changed from Active UP to invalid Defect;

the tunnel state of the VXLAN tunnel includes: valid UP and invalid delete;

when the tunnel state of the VXLAN tunnel is valid UP, the using state of the VXLAN tunnel includes: active and Inactive Inactive;

and when the tunnel state of the VXLAN tunnel is invalid, the VXLAN tunnel cannot forward the service message.

2. The method of claim 1, wherein after detecting, via a BFD session, that a tunnel status of the third VXLAN tunnel changes from a valid UP to an invalid default, the method further comprises:

detecting a tunnel state of the third VXLAN tunnel through a BFD session;

and when detecting that the tunnel state of the third VXLAN tunnel is changed from invalid Defect to valid UP through the BFD session, setting the tunnel state of the first VXLAN tunnel to be valid UP.

3. The method of claim 1, wherein detecting a tunnel status of the third VXLAN tunnel through a BFD session comprises:

sending a BFD control message to the first server at intervals of a first time length;

when a response message returned by the first server based on the BFD control message is received within a second time length, determining that the tunnel state of the third VXLAN tunnel is effective UP; or

And when a response message returned by the first server based on the BFD control message is not received within a second time, determining that the tunnel state of the third VXLAN tunnel is invalid.

4. The method of claim 2, wherein setting the tunnel state of the first VXLAN tunnel to active UP upon detecting, via a BFD session, that the tunnel state of the third VXLAN tunnel changes from inactive Defect to active UP comprises:

and when the tunnel state of the third VXLAN tunnel is detected to be changed from invalid Defect to valid UP through the BFD session and the valid UP is kept to exceed a third time length, setting the tunnel state of the second VXLAN tunnel to be valid UP.

5. A message forwarding method is characterized in that the current use state of a first VXLAN tunnel is an Active state, the current use state of a second VXLAN tunnel is an Inactive state, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the method is used for a convergence switch and comprises the following steps:

detecting a tunnel state of the first VXLAN tunnel through the BFD session;

forwarding the service message through the second VXLAN tunnel;

the tunnel state of the VXLAN tunnel includes: valid UP and invalid delete;

6. The method of claim 5, wherein upon detecting, via a BFD session, that a tunnel status of the first VXLAN tunnel changes from active UP to inactive default, comprising: the tunnel state of the first VXLAN tunnel is invalid Defect, which is set for the first VXLAN tunnel by the first resource pool gateway when the tunnel state of the third VXLAN tunnel is invalid Defect, and the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server.

7. A message forwarding method is characterized in that the method is used for a dual DC system, the dual DC system comprises an aggregation switch, a first resource pool gateway, a second resource pool gateway and a first server, a first VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the first resource pool gateway, a second VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the second resource pool gateway, and a third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server; the method comprises the following steps:

the convergence switch forwards the service message through the second VXLAN tunnel;

the tunnel state of the VXLAN tunnel includes: valid UP and invalid delete;

8. A message forwarding device is characterized in that the current use state of a first VXLAN tunnel is an Active state, the current use state of a second VXLAN tunnel is an Inactive state, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the apparatus is for a first resource pool gateway, comprising:

a first setting module, configured to set a tunnel state of the first VXLAN tunnel to invalid when detecting that a tunnel state of the third VXLAN tunnel is changed from valid UP to invalid default through a BFD session, so that the aggregation switch sets a usage state of the second VXLAN tunnel to Active when detecting that the tunnel state of the first VXLAN tunnel is changed from valid UP to invalid default;

the tunnel state of the VXLAN tunnel includes: valid UP and invalid delete;

9. The apparatus of claim 8, wherein the first setup module is further configured to:

10. The apparatus of claim 8, wherein the first detection module is configured to:

11. The apparatus of claim 9, wherein the first setup module is further configured to:

12. A message forwarding device is characterized in that the current use state of a first VXLAN tunnel is an Active state, the current use state of a second VXLAN tunnel is an Inactive state, the first VXLAN tunnel is a VXLAN tunnel between a convergence switch and a first resource pool gateway, and the second VXLAN tunnel is a VXLAN tunnel between the convergence switch and a second resource pool gateway; the apparatus is for a converged switch, comprising:

the message forwarding module is used for forwarding the service message through the second VXLAN tunnel;

the tunnel state of the VXLAN tunnel includes: valid UP and invalid delete;

13. The apparatus of claim 12, wherein upon detecting, via a BFD session, that a tunnel status of the first VXLAN tunnel changes from active UP to inactive default, comprising: the tunnel state of the first VXLAN tunnel is invalid Defect, which is set for the first VXLAN tunnel by the first resource pool gateway when the tunnel state of the third VXLAN tunnel is invalid Defect, and the third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server.

14. A message forwarding apparatus, wherein the apparatus is used in a dual DC system, the dual DC system includes an aggregation switch, a first resource pool gateway, a second resource pool gateway, and a first server, a first VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the first resource pool gateway, a second VXLAN tunnel is a VXLAN tunnel between the aggregation switch and the second resource pool gateway, and a third VXLAN tunnel is a VXLAN tunnel between the first resource pool gateway and the first server; the device comprises:

the tunnel state of the VXLAN tunnel includes: valid UP and invalid delete;