A multi-protocol linkage method for fast switching network failures
By configuring the linkage of CFM, BFD and VRRP protocols in the Ethernet communication network, the problem of long failover time in the existing technology is solved, and fast network failure recovery is achieved, reducing the switching time.
Patent Information
- Application Number
- CN202210709911.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-22
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2042-06-22
AI Technical Summary
The existing Ethernet communication networking architecture has a long failover time in complex environments, which affects the normal operation of the network and is difficult to recover quickly.
By reasonably configuring a variety of protocols, such as CFM, BFD and VRRP in the network communication network, link detection and priority linkage is realized, replacing VRRP announcement linkage, and quickly switching network failures.
The failover time is greatly reduced, and the failover time between the data center and local unit core switching is no more than 500ms, and the failover time between the data center and PTN link is no more than 300ms, which improves the failure recovery rate.
Smart Images

Figure CN115190006B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of communication technology, and in particular to a fast switching network fault method for fast fault recovery in a network scenario with high reliability requirements. Background Art
[0002] Usually, Ethernet communication networking is more complex, and many redundant protocols are used in networking, such as Rapid Spanning Tree Protocol (RSTP protocol), Virtual Router Redundancy Protocol (VRRP protocol), Bidirectional Forwarding Detection Protocol (BFD protocol), OAM protocol (Link Detection Protocol) and Link Aggregation Control Protocol (LACP protocol). If only a single protocol fault switching is fast, most protocol switching times can be achieved at the MS level, but in actual use, when the environment is more complex, the communication fault lasts longer, affecting the normal operation of the network.
[0003] The Ethernet communication network architecture of the prior art is as follows Figure 4 As shown in the figure, the data center and local units in the network are connected through the active-standby switching network cable transmission, and the external transmission side uses the active-standby PTN equipment through optical fiber transmission. Normally, only VRRP protocol and RSTP protocol are used. VRRP maintains the status through negotiation. The default VRRP notification time is 1S, so the fault switching time will be in seconds. The specific fault situation is described as follows:
[0004] If the core master switch fails, the standby switch needs to switch to the Master state, and the RSTP state needs to be renegotiated to confirm that the standby switch is the root bridge. Data packets can only be forwarded from the core standby switch to the SPTN link, so the standby switch external interface needs to synchronously learn the ARP table entries, which takes at least 4 seconds to restore normal communication.
[0005] If the link between the core main switching device and the PTN device in the data center fails, the data message can only be forwarded from the core main switching to the core backup switching, and then forwarded from the link between the core backup switching and the SPTN. In this case, the backup switching external interface needs to synchronously learn the ARP table entries, which takes at least 2S to restore normal communication.
[0006] Therefore, in the existing Ethernet communication networking architecture mode, how to improve the fault recovery speed has always been a technical problem that has troubled technical personnel in this field. Summary of the invention
[0007] In view of the above problems, the present invention provides a multi-protocol linkage fast switching network fault method based on the existing networking architecture, which realizes fast fault recovery by reasonably configuring the communication protocol of each node.
[0008] The technical solution of the present invention is: running in a network communication network, the network communication network includes a terminal 1, the terminal 1 is connected to a data center-main and a data center-backup through an access 1,
[0009] The data center-master is then connected to several local master PTNs and local unit-masters in sequence through the data center master PTN, and then connected to access 2;
[0010] The data center-backup is then connected to several local backup PTNs and local unit-backups (local unit backup switches) in sequence through the data center backup PTN, and then connected to the access switching device 2;
[0011] The access switching device 2 is connected to the terminal 2;
[0012] Establishing a VRRP group between the local unit-master and the local unit-backup, and establishing a VRRP group between the data center-master and the data center-backup;
[0013] The communication between the terminal-1 and the terminal-2 is a static routing protocol used by the core layer;
[0014] The data center-master and data center master PTN, data center-backup and data center backup PTN, local unit-master and local unit master PTN, local unit-backup and local unit backup PTN respectively use CFM protocol to perform link detection;
[0015] The BFD protocol is used to perform link detection between the local unit-master and the local unit-backup, and between the data center-master and the data center-backup;
[0016] The VRPP protocols established between the local unit-master and the local unit-backup and between the data center-master and the data center-backup are all associated with the detection status of the CFM and BFD protocols;
[0017] Link detection is that the CFM and BFD protocols are linked to the VRRP protocol priority calculation, replacing the VRRP notification linkage to achieve fast link switching.
[0018] The terminal-1, access-1, data center-master, data center-backup, data center master PTN and data center backup PTN constitute a data center; the terminal-2, access-2, local unit-master, local unit-backup, local master PTN and local backup PTN constitute a local unit;
[0019] The data center and several local units use a layered network design including a core layer and an access layer. The core layer provides transmission within the optimal range, and the access layer provides user access to the network for multi-service applications and other network applications.
[0020] The data center at least includes terminal-2 and access-2; the access-2 connects the data center-master and the data center-backup.
[0021] The core layer adopts a three-layer switching environment to provide high-speed connections between various access layers of the network.
[0022] The steps to implement CFM and VRRP are as follows:
[0023] 1) Run the CFM protocol between the switching interfaces of data center-master / local unit-master, data center-backup / local unit-backup and the corresponding interfaces of each PTN to detect the link status;
[0024] 2) For VRRP groups between local unit-master / local unit-backup, data center-master / data center-backup, set the master-backup exchange VRRP instance priority respectively;
[0025] 3) Set the CFM protocol to be linked with the VRRP protocol and set the descending priority;
[0026] 4) Ensure that the instance priority of the VRRP master is less than the current backup priority after deducting the decreasing priority.
[0027] The steps to implement BFD and VRRP are as follows:
[0028] Step 1: For VRRP groups between local unit-master / local unit-backup, data center-master / data center-backup, set the master-backup exchange VRRP instance priority respectively;
[0029] Step 2: Configure the real address of the VRRP group between the local unit-master / local unit-backup and the data center-master / data center-backup to the BFD state address;
[0030] Step 3: Set the expected packet receiving rate and packet sending rate of the BFD state address parameter setting interface to 10 milliseconds, and set the detection interval multiplier to 3.
[0031] Step 4: Set the BFD protocol to be linked with the VRRP protocol, and set the descending priority.
[0032] Step 5: Ensure that the priority of the instance in VRRP Master state minus the decreasing priority is less than the priority of the current Backup.
[0033] The present invention establishes VRRP groups in local unit-master / local unit-backup and data center-master / data center-backup, uses static routing between the data center and the local unit core switch, uses CFM protocol to perform link detection between the data center, the local unit core switch and the PTN equipment, uses BFD protocol to perform link detection between the local unit master and backup core switches and the data center master and backup core switches, VRPP protocol of the core switch is associated with CFM and BFD detection status, link detection CFM and BFD protocol are linked to VRRP protocol priority calculation, and VRRP notification linkage is replaced to realize fast switching of links.
[0034] In this system networking scenario, the existence of a Layer 2 communication area will cause a broadcast storm. The loop networking breaking method adopted is shown in Table 1 below:
[0035] Table 1: Loop breaking protocols used by various devices in the network
[0036]
[0037] Since the VRRP protocol is only the master and backup protocol status, but the interface is still in the active state, the PTN equipment on the transmission side is a layer 2 ring network, which will still lead to an external ring network. The OAM-interface status is derived from this. When VRRP is Master, the OAM-interface status of the interface between the main switch and the transmission side is Forwarding. When VRRP is Backup, the OAM-interface status of the interface between the main switch and the transmission side is Block. Through the state control of the OAM-interface, the transmission side is loop-free, thus avoiding the risk of network storms.
[0038] The RSTP rapid spanning tree protocol is mainly used within the data center and local units. The CFM linkage VRRP protocol is mainly used between the data center core switch and the PTN equipment. The BFD linkage VRRP protocol is mainly used between the data center master and standby switches. In the above configuration, when the data center and local unit core master switch fails, the switching time is no more than 500ms. When the link between the data center and local unit core master switch and the PTN master device fails: the switching time is no more than 300ms.
[0039] This communication scenario greatly optimizes the time for communication recovery after a failure. The actual test failure switching time of this scenario is shown in Table 2 below:
[0040] Table 2: Failure switching test data of the present invention
[0041]
[0042] The test data in the table above shows that: when the core main switch of the data center and the local unit fails, the switching time is no more than 500ms. When the link between the core main switch of the data center and the local unit and the PTN main equipment fails: the switching time is no more than 300ms.
[0043] The present invention runs multiple link detection protocols on different components and switches in conjunction with the VRRP protocol, thereby greatly improving the rate of fault recovery. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 is a block diagram of a communication network group with high reliability requirements according to the present invention,
[0045] Figure 2 This is a principle block diagram of the present invention applied to large-scale networking.
[0046] Figure 3 This is a flowchart of multi-protocol linkage fast switching.
[0047] Figure 4 It is a schematic diagram of the background technology of the present invention. DETAILED DESCRIPTION
[0048] The technical solution of the present invention is as follows Figure 1-3 : In the front and back scenarios, the network switching equipment and the transmission-side PTN equipment are equipped with the Rapid Spanning Tree Protocol (RSTP protocol), Virtual Router Redundancy Protocol (VRRP protocol), Bidirectional Forwarding Detection Protocol (BFD protocol) and OAM protocol (link detection protocol).
[0049] A multi-protocol linkage fast switching network fault method of the present invention is operated in a network communication network, the network communication network (such as Figure 1 The data center includes terminal 1 (data center terminal device at the rear), which is connected to data center-main (data center main switch) and data center-backup (data center backup switch) through access 1 (i.e., access switching device-1).
[0050] The data center-main is then connected to several local main PTNs and local unit-mains (local unit main switches) in sequence through the data center main PTN, and then connected to access 2 (i.e. access switching device-2);
[0051] The data center-backup is then connected to several local backup PTNs and local unit-backups (local unit backup switches) in sequence through the data center backup PTN, and then connected to access switching device 2;
[0052] Access switch 2 connects to terminal 2 (the local unit terminal device at the rear);
[0053] A VRRP group is established between the local unit-master and the local unit-backup, and a VRRP group is established between the data center-master and the data center-backup;
[0054] The communication between terminal-1 and terminal-2 is through the static routing protocol used by the core layer;
[0055] The data center-master and data center master PTN, the data center-backup and data center backup PTN, the local unit-master and local unit master PTN, and the local unit-backup and local unit backup PTN use the CFM protocol to perform link detection.
[0056] The BFD protocol is used to detect links between the local unit-master and local unit-backup, and between the data center-master and data center-backup.
[0057] The VRPP protocols established between the local unit-master and the local unit-backup and between the data center-master and the data center-backup are all associated with the detection status of the CFM and BFD protocols;
[0058] Link detection is the process of CFM and BFD protocols being linked to VRRP protocol priority calculation, replacing VRRP notification linkage to achieve fast link switching.
[0059] Further, terminal-1, access-1, data center-master, data center-backup, data center master PTN and data center backup PTN constitute a data center; terminal-2, access-2, local unit-master, local unit-backup, local master PTN and local backup PTN constitute a local unit;
[0060] The layered network design used within the data center and some of the local units includes a core layer and an access layer. The core layer provides transmission within the optimal range, and the access layer provides user access to the network for multi-service applications and other network applications.
[0061] Furthermore, the data center at least includes a terminal-2 and an access-2; the access-2 connects the data center-master and the data center-backup. Figure 1 , 2 As shown, there will be multiple other isolated departments in the data center, and different departments can set up their own terminals and access switches respectively.
[0062] Furthermore, the core layer adopts a three-layer switching environment to provide high-speed connections between various access layers of the network.
[0063] The steps of implementing the CFM linkage VRRP protocol in the present invention are as follows:
[0064] 1) Run the CFM protocol between the switching interfaces of data center-master / local unit-master, data center-backup / local unit-backup and the corresponding interfaces of each PTN to detect the link status;
[0065] 2) For VRRP groups between local unit-master / local unit-backup, data center-master / data center-backup, set the master-backup exchange VRRP instance priority respectively;
[0066] 3) Set the CFM protocol to be linked with the VRRP protocol and set the descending priority;
[0067] 4) Ensure that the instance priority of the VRRP master is less than the current backup priority after deducting the decreasing priority.
[0068] The steps of implementing the BFD linkage VRRP protocol in the present invention are as follows:
[0069] Step 1: For VRRP groups between local unit-master / local unit-backup, data center-master / data center-backup, set the master-backup exchange VRRP instance priority respectively;
[0070] Step 2: Configure the real address of the VRRP group between the local unit-master / local unit-backup and the data center-master / data center-backup to the BFD state address;
[0071] Step 3: Set the expected packet receiving rate and packet sending rate of the BFD state address parameter setting interface to 10 milliseconds, and set the detection interval multiplier to 3.
[0072] Step 4: Set the BFD protocol to be linked with the VRRP protocol, and set the descending priority.
[0073] Step 5: Ensure that the priority of the instance in VRRP Master state minus the decreasing priority is less than the priority of the current Backup.
[0074] The fast switching mechanism implemented by the present invention is:
[0075] The local unit main switch and the local unit standby switch establish a VRRP group, the data center main switch and the data center standby switch establish a VRRP group, the data center and the local unit core switch use static routing, the data center and the local unit core switch and PTN equipment use the CFM protocol for link detection, the local unit main and standby core switches and the data center main and standby core switches use the BFD protocol for link detection, the core switch VRPP protocol is associated with CFM, BFD detection status, link detection CFM, BFD protocol linkage VRRP protocol priority calculation, instead of VRRP notification linkage to achieve fast link switching.
[0076] Since the VRRP protocol is only the master and backup protocol status, but the interface is still in the active state, the PTN equipment on the transmission side is a layer 2 ring network, which will still lead to an external ring network. The OAM-interface status is derived from this. When VRRP is Master, the OAM-interface status of the interface between the main switch and the transmission side is Forwarding. When VRRP is Backup, the OAM-interface status of the interface between the main switch and the transmission side is Block. Through the state control of the OAM-interface, the transmission side is loop-free, thus avoiding the risk of network storms.
[0077] The RSTP rapid spanning tree protocol is mainly used within the data center and local units. The CFM linkage VRRP protocol is mainly used between the core switch of the data center and the PTN equipment. The BFD linkage VRRP protocol is mainly used between the active and standby switches of the data center.
[0078] In addition, large-scale networking scenarios such as data centers and a large number of local units are being constructed. Figure 2 As shown in the figure, the two data are physically connected to a large number of local units through the transmission side PTN equipment, and communicate through routing. The internal scenes of each data center and local unit are still built according to the protocol mode described above, and a large number of local units are distinguished by different IP addresses. The protocol linkage of each data center and local unit does not affect each other, and the working process is the same as Figure 1 The data flow direction remains consistent.
[0079] The working process of the present invention is further described below in conjunction with embodiments.
[0080] Example 1: Under normal conditions, data flows as follows: Figure 1 The “normal traffic path” is shown by the dashed line.
[0081] Under normal conditions of all devices: data center terminal device traffic → access switch → data center main core switching device → data center main PTN device → local unit main PTN device → local unit main core switching device → access switch → local unit terminal;
[0082] Example 2: Data center and local unit core main switch failure downtime fault scenario: Figure 1 The double-dotted line in the figure shows the "data center main switch failure traffic path". Figure 3 Right side.
[0083] Step 1: The BFD protocol between the data center and the local unit's active and standby core switches detects a fault state;
[0084] Step 2: The VRRP role for data center and local unit standby switching is immediately switched to Master state;
[0085] Step 3: The OAM-Interface status between the data center and the local unit core switch and the PTN equipment is changed from Block to Forwarding;
[0086] Step 4: Request ARP from the standby core switch to synchronize the table entries, and the switching time is no more than 500ms.
[0087] In the event of a failure of the main core switching device in the data center: data center terminal device traffic → access switch → data center backup core switching device → data center backup PTN device → local unit backup PTN device → local unit backup core switching device → local unit main core switching device → access switch → local unit terminal;
[0088] Example 3: Link failure scenario between data center and local unit core main switch and PTN main equipment: Figure 1 In the figure, the dashed line shows the traffic path of the main link failure in the data center. Figure 3 Left side.
[0089] Step 1: The CFM protocol between the data center and the local unit main core switch and the PTN main equipment detects a fault state;
[0090] Step 2: The VRRP roles of the data center and the local unit are immediately switched to the Master state;
[0091] Step 3: The OAM-Interface status between the data center and the local unit core switch and the PTN equipment is changed from Block to Forwarding;
[0092] Step 4: The standby core switch requests ARP to synchronize entries, and the switching time is no more than 300ms.
[0093] Link failure status between the data center main core switching device and the main PTN device: data center terminal device traffic → access switch → data center main core switching device → data center backup core switching device → data center backup PTN device → local unit backup PTN device → local unit backup core switching device → local unit main core switching device → access switch → local unit terminal;
[0094] The present invention is applicable to scenarios with high reliability requirements, and all devices in the network support the required network protocols.
[0095] The backbone of the overall network communication equipment includes transmission PTN equipment and data service Ethernet, among which the data service Ethernet plays the role of terminal access, data aggregation, service gateway, service routing, link backup and other functions, and the transmission network plays the role of long-distance and rapid transmission of service data.
[0096] The network layered design used within data centers and local units includes a core layer and an access layer. The core layer provides transmission within the optimal range, and the access layer provides user access to the network for multi-service applications and other network applications.
[0097] The core layer is a high-speed switching backbone. The design is to minimize the time spent switching packets. It is common practice to use a layer 3 switching environment in the core layer, which means that VLANs and VLAN trunks do not appear in the core layer. This also means that spanning tree loops in the core layer can usually be avoided. The main function of the core layer is to provide high-speed connections between the various access layers of the network.
[0098] The access layer is the point where local end users are allowed to access the network. Usually, layer 2 switches play a very important role in the access layer. In the access layer, switches become edge devices because they are located at the boundary of the network. Layer 2 switches are data link layer devices that can receive MAC address information in device data packets, forward them according to MAC addresses, and record these MAC addresses and corresponding ports in their internal MAC address table entries.
[0099] The transmission side is carried out by PTN equipment. The main protocols deployed by data centers and local units are: Vlan, Trunk, NAT, CfM, ACL, VRRP, OAM, OSPF, spanning tree, static routing and other protocols.
[0100] The local unit main switch and the local standby switch establish a VRRP group. The static routing protocol is used between the data center core and transmission. The data center and local unit core switches and transmission PTN equipment use the CFM protocol for link detection. The local unit core switch VRPP protocol is associated with the CFM and BFD detection status. The link detection CFM and BFD protocols are linked to the VRRP protocol priority calculation, replacing the VRRP notification linkage to achieve fast link switching and optimize network transmission efficiency.
[0101] The steps to implement CFM and VRRP are as follows:
[0102] Step 1: Run the CFM protocol between the switching interface and the corresponding interface of the PTN to detect the link status;
[0103] Step 2: The primary and backup switches of the data center and local units form VRRP groups respectively, and set the priority of the primary and backup VRRP instances respectively;
[0104] Step 3: Set the CFM protocol to be linked with the VRRP protocol and set the descending priority.
[0105] Step 4: Ensure that the priority of the instance in VRRP Master state minus the decreasing priority is less than the priority of the current Backup.
[0106] The steps to implement BFD and VRRP are as follows:
[0107] Step 1: The primary and backup switches of the data center and local units form VRRP groups respectively, and set the priority of the primary and backup VRRP instances respectively;
[0108] Step 2: Configure the BFD state machine between the primary and standby switches of the data center and the local unit using the real address of the VRRP group;
[0109] Step 3: Set the expected packet receiving rate and packet sending rate of the BFD interface to 10 milliseconds, and set the detection interval multiplier to 3.
[0110] Step 4: Set the BFD protocol to be linked with the VRRP protocol, and set the descending priority.
[0111] Step 5: Ensure that the priority of the instance in VRRP Master state minus the decreasing priority is less than the priority of the current Backup.
[0112] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solution disclosed in the present invention, technicians in this field can make some substitutions and deformations to some technical features therein according to the disclosed technical content without creative labor, and these substitutions and deformations are all within the protection scope of the present invention.
Claims
1. A multi-protocol linkage fast switching network failure method, which runs in a network communication network, wherein the network communication network includes a terminal-1, and the terminal-1 is connected to a data center main switch and a data center backup switch through an access switching device-1. The data center main switch is then connected to several local main PTNs and local unit main switches in sequence through the data center main PTN, and then connected to the access switching device-2; The data center backup switch is then connected to several local backup PTNs and local unit backup switches in sequence through the data center backup PTN, and then connected to the access switching device-2; The access switching device-2 is connected to the terminal-2; It is characterized in that Establishing a VRRP group between the local unit main switch and the local unit backup switch, and establishing a VRRP group between the data center main switch and the data center backup switch; The communication between the terminal-1 and the terminal-2 is a static routing protocol used by the core layer; The data center main switch and the data center main PTN, the data center backup switch and the data center backup PTN, the local unit main switch and the local unit main PTN, the local unit backup switch and the local unit backup PTN respectively use the CFM protocol to perform link detection; The BFD protocol is used to perform link detection between the local unit main switch and the local unit backup switch, and between the data center main switch and the data center backup switch; The VRPP protocols established between the local unit main switch and the local unit backup switch and between the data center main switch and the data center backup switch are all associated with the detection status of the CFM and BFD protocols; Link detection is that the CFM and BFD protocols are linked to the VRRP protocol priority calculation, replacing the VRRP notification linkage to achieve fast link switching; The terminal-1, access switching device-1, data center main switch, data center backup switch, data center main PTN and data center backup PTN constitute a data center; the terminal-2, access switching device-2, local unit main switch, local unit backup switch, local main PTN and local backup PTN constitute a local unit; The data center and the local units use a layered network design including a core layer and an access layer, wherein the core layer provides transmission in the optimal interval, and the access layer provides user access to the network for multi-service applications and other network applications; The steps to implement CFM and VRRP are as follows: 1) Run the CFM protocol between the switching interfaces of the data center main switch / local unit main switch, data center backup switch / local unit backup switch and the corresponding interfaces of each PTN to detect the link status; 2) For the VRRP groups between the local unit main switch / local unit backup switch and the data center main switch / data center backup switch, set the priority of the primary and backup switching VRRP instances respectively; 3) Set the CFM protocol to be linked with the VRRP protocol and set the descending priority; 4) Ensure that the priority of the instance in VRRP Master state is less than the priority of the current Backup after deducting the decreasing priority; The steps to implement BFD and VRRP are as follows: Step 1: For the VRRP groups between the local unit master switch / local unit backup switch and the data center master switch / data center backup switch, set the master and backup switching VRRP instance priorities respectively; Step 2: The real address of the VRRP group between the local unit main switch / local unit backup switch and the data center main switch / data center backup switch is configured as the BFD state address; Step 3: Set the expected packet receiving rate and packet sending rate of the BFD state address parameter setting interface to 10 milliseconds, and set the detection interval multiplier to 3. Step 4: Set the BFD protocol to be linked with the VRRP protocol, and set the descending priority. Step 5: Ensure that the priority of the instance in VRRP Master state minus the decreasing priority is less than the priority of the current Backup.
2. According to claim 1, a multi-protocol linkage fast switching network failure method, It is characterized in that The data center at least includes a terminal-2 and an access switching device-2; the access switching device-2 is connected to a main switch of the data center and a backup switch of the data center.
3. According to claim 1, a multi-protocol linkage fast switching network failure method, It is characterized in that The core layer adopts a three-layer switching environment to provide high-speed connections between various access layers of the network.
