[go: up one dir, main page]

WO2018108149A1 - 数据链路切换方法和装置以及数据链路切换设备 - Google Patents

数据链路切换方法和装置以及数据链路切换设备 Download PDF

Info

Publication number
WO2018108149A1
WO2018108149A1 PCT/CN2017/116457 CN2017116457W WO2018108149A1 WO 2018108149 A1 WO2018108149 A1 WO 2018108149A1 CN 2017116457 W CN2017116457 W CN 2017116457W WO 2018108149 A1 WO2018108149 A1 WO 2018108149A1
Authority
WO
WIPO (PCT)
Prior art keywords
link
port
information
switching
switch
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/116457
Other languages
English (en)
French (fr)
Inventor
黄治文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Publication of WO2018108149A1 publication Critical patent/WO2018108149A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/22—Arrangements for detecting or preventing errors in the information received using redundant apparatus to increase reliability
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/28—Routing or path finding of packets in data switching networks using route fault recovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/24—Multipath

Definitions

  • the present invention relates to the field of optical communications and computers, and in particular, to a data link switching method and apparatus.
  • An optical line terminal (OLT) device usually needs to have a switching protection function.
  • OLT optical line terminal
  • the packet that should be sent to the failed port needs to be changed. It is sent to an alternate port that is used to replace the failed port to protect the normal use of the OLT device.
  • the port forwarding table needs to be updated synchronously between the peripheral switching chips of the OLT device, the update time of the port forwarding table is long, and another mechanism is needed to reduce the packet loss during the update. rate.
  • the present invention provides a data link switching method, a data link switching device, and a data link switching device, to solve the problem that the device bandwidth utilization is not high, the design is complicated, the versatility is low, and the multi-function cannot be targeted in the data link switching process. At least one of technical problems such as handling of a fault condition differently.
  • a data link switching method comprising the steps of: collecting link failure status information and link port information by a central processor; and transmitting the link failure status information by the central processor And the link port information is encapsulated into The link switch message is sent to the network processor, where the link switch message is used to instruct the network processor to perform link switch.
  • a data link switching method including the steps of: a network processor receiving a message, where the network processor extracts the message if the message is a link switch message Link failure status information and link port information carried in the message; in the case that the failure status indicated by the link failure status information is a handover allowed state, the network processor is based on the priority in the handover port database Selecting a target switching port; and the network processor switching a forwarding function of the port pointed to by the link port information to the target switching port.
  • a data link switching apparatus includes: a collection module, configured to collect link failure status information and link port information; and a packaging module, configured to: The link port information is encapsulated into a link switch message, and the sending module is configured to send the link switch message to the network processor.
  • a data link switching apparatus including: a receiving module, configured to receive a message, and in a case where the packet is a link switching message, extracting the packet Carrying link fault state information and link port information; selecting a module, in the case that the fault state indicated by the link fault state information is a permission to switch state, the selecting module selects according to a priority in the switch port database a target switching port; and a switching module, configured to switch a forwarding function of the port pointed to by the link port information to the target switching port selected by the selecting module.
  • a data link switching device includes a network processor and a central processor, wherein the central processor is configured to collect link failure status information and link port information of a faulty port Transmitting, to the network processor, a link switch message including the link fault status information and the link port information, and the network processor is configured to receive the message sent by the central processor, And in the case that the packet is a link switch message, extracting link fault state information and link port information carried in the packet, where the fault state indicated by the link fault state information is allowed to be switched.
  • the target switching port is selected according to the priority in the switching port database, and the forwarding function of the port pointed to by the link port information is switched to the target switching port.
  • FIG. 1 is a flow chart of a data link switching method in accordance with an embodiment of the present invention
  • FIG. 2 is a flowchart of a data link switching method according to another embodiment of the present invention.
  • FIG. 3 is a schematic block diagram of a data link switching device in accordance with an embodiment of the present invention.
  • FIG. 4 is a schematic block diagram of a data link switching apparatus according to another embodiment of the present invention.
  • FIG. 5 is a schematic block diagram of a network processor based line card in accordance with an embodiment of the present invention.
  • FIG. 6 is a schematic diagram of a link switch port conversion table according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of a link switching function switch table according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of a microcode link switching information table according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of a link switch message in accordance with an embodiment of the present invention.
  • FIG. 1 is a flow chart of a data link switching method in accordance with an embodiment of the present invention.
  • a data link switching method may include the steps of collecting link failure state information and link port information by a central processing unit (S101); and causing a link failure through a central processing unit.
  • the status information and link port information are encapsulated into a link switch message and sent to the network processor (S102).
  • the link switch message can be used to instruct the network unit to perform link switch.
  • synchronously updating the data status of all the boards may result in excessive link switching and excessive packet loss, and data link switching may be implemented in different manners in different scenarios. It can lead to complicated logic.
  • the data link switching method according to the present invention can effectively avoid or reduce the packet loss of the forwarding data during the data state synchronization process of the card during the link switching, and can solve the problem that the specific failure scenario cannot be differentiated. The solution to the problem.
  • a Central Processing Unit (CPU) is used as a collection unit of failure information of a collection port, and a Network Processor (NP) is used as a forwarding unit for forwarding data.
  • the CPU notifies the NP of the status information required for the switching of the various types of data links in the form of a message, and the NP searches the specific microcode table according to the received status information to obtain the link switching operation information to be taken, and according to the obtained chain.
  • the path switching operation information forwards the packet.
  • the CPU detects various types of status information in various fault scenarios that need to perform link switching processing, and generates a message according to the status information.
  • the packet header carries all the status information that may affect the link switching processing mode.
  • Link failure status information can be generated by abstracting scenarios of various link failures.
  • the link failure status information may include one of the following link failures: loss of optical network unit signal, port not establishing link, continuity check exception, bidirectional forwarding detection failure, and the like.
  • the link port information indicates the identity of the port and may include, but is not limited to, port type information and port identity information.
  • link failure state information can be divided into a plurality of types.
  • the CPU may generate one or more messages and send them to the NP for processing.
  • Each packet can trigger an operation of searching for a link switching information table, thereby implementing switching processing of multiple data links.
  • the link fault status information may be divided into two sets of serial and parallel by type, and the types may include, but are not limited to, optical network unit signal loss, port not established link, continuity check abnormality, bidirectional forwarding detection failure, and the like. In a serial set, if there are two or more types of link failure status information, multiple messages must be generated accordingly.
  • the NP after receiving the packet sent by the CPU, the NP searches only the link switching information table once. Parallel set If there are two or more types of link fault status information, only one packet can be generated and various link fault status information can be carried in the packet header at the same time. In this case, after receiving the packet sent by the CPU, the NP searches the link switch information table multiple times and performs the corresponding packet forwarding special processing according to the result of the lookup table of the link switch information table, thereby improving the chain. The response speed of the road switch.
  • the link failure state information may be encapsulated using a message having a variable header length, so that only one message carrying multiple types of link failure state information may be generated.
  • the NP After receiving the packet, the NP needs to extract the link fault status information of all types (for example, N types) from the packet header after detecting that the packet header is a variable length header. N parts for loopback processing.
  • Each packet receives the link fault status information at different offset positions of the packet header according to the sequence number of the packet, and performs an operation of searching for the link switching information table, and then performs corresponding operations according to the result of the lookup table. Link switching operation.
  • FIG. 2 is a flow chart of a data link switching method according to another embodiment of the present invention.
  • the data link switching method may include the following steps: the network processor receives the packet, and if the packet is a link switching packet, extracts the packet to carry Link failure status information and link port information (S201); and in the case where the failure status indicated by the link failure status information is the allowable handover status, the network processor selects the target switching port according to the priority in the handover port database. (S202).
  • the link failure state required for switching various types of data links received by the NP is utilized by utilizing the NP's Operation Management and Maintenance (OAM) module's constructing and packet sending functions.
  • OAM Operation Management and Maintenance
  • the information and link port information informs the microcode module in the form of a message.
  • the microcode module searches for a specific microcode table according to the link fault state information, obtains the link switching operation information that should be taken, and forwards the packet according to the obtained link switching operation information.
  • the OAM module forwards the received packet to the microcode module of the NP.
  • the microcode module parses the message to obtain various link fault state information and link port information for performing a link switching operation.
  • the microcode module searches for a specific microcode table (for example, a link switching information table) according to link fault state information and link port information to obtain a link cut for performing Change the operation information of the operation.
  • the microcode module performs corresponding processing on the packet that needs to change the forwarding path according to the operation information obtained by searching the link switching information table, so that the packet is not forwarded to the Forwarding Information Base (FIB) table.
  • the determined port is forwarded to the port obtained by looking up the link switch information table.
  • FIB Forwarding Information Base
  • the microcode module parses the packet sent by the OAM module.
  • the microcode module can process multiple types of packets sent by the OAM module, for example, MAC aging or fast aging packets, 1588 packets, and the like.
  • the type of the packet can be distinguished by a specific field in the header of the customized packet. Different types of packets can correspond to different parsing methods and processing codes.
  • the microcode code for implementing the link switching function is executed when the microcode module receives the link switch message.
  • the microcode module may further determine whether the header of the message is a fixed length header or a variable length header. . If it is a fixed-length header, the number of times the operation of finding the link switching information table needs to be performed is determined according to the number of link failure state information types included in the header; if it is a variable-length header, it will be reported. Copy multiple copies and perform loopback processing.
  • the link switch operation switch table needs to be searched to determine whether to perform the link corresponding to the current link fault state information. Road switching operation. If the link switching operation corresponding to the current link failure status information is forbidden, the microcode module directly exits the relevant processing flow, and does not actually perform any link switching operation. If the link switching operation corresponding to the current link failure state information is allowed to be performed, the microcode module continues to perform the related operations of the link switching process.
  • the microcode module searches for the link switching information table according to the obtained link fault state information
  • multiple optional link switching operations may be simultaneously obtained. For example, when port A cannot forward packets due to passive or active reasons, you can choose to switch the forwarding of packets to port B or switch to port C. At this time, it is necessary to determine the link switching operation to be actually performed according to the priority of each optional link switching operation. The link switching operation with the highest priority can be selected as the link switching operation to be actually performed next.
  • the priority of the link switching operation may be a result calculated by a function including two factors.
  • One factor is the default priority of the link switching operation. The higher the default priority of the link switching operation, the higher the final priority calculated by the link switching operation through the function.
  • Another factor is the number of times the link switching operation is performed within a fixed time interval (i.e., unit time), and the more the number of executions, the lower the final priority calculated by the function.
  • a value can be set as a priority for the link switching operation.
  • the value set can be a fixed value, ie the default priority of the link switch operation.
  • the set value may vary (eg, periodically) depending on the port processing capabilities to be switched.
  • the value set when the port processing capability to be switched is increased can be increased, and the value set when the port processing capability to be switched is lowered can be reduced.
  • the port to be switched having the largest set value can be selected as the target switch port, that is, the link switch operation having the highest priority can be selected as the link switch operation to be actually performed next.
  • the NP performs handover forwarding according to the obtained link failure state information and link port information.
  • the packet is actually forwarded to the port, and finally the link switching operation is implemented.
  • the destination port ie, the faulty port
  • the affected packet can be forwarded without waiting for a large number of MAC address synchronization operations of the boards in the distributed system to be completed. Switching to the new destination port quickly, which enables fast switching of the destination port for packet forwarding without significantly changing the general processing flow of Layer 2 packets. Therefore, while ensuring reliability, the performance of data link switching in a distributed system is effectively improved.
  • FIG. 3 is a schematic block diagram of a data link switching device in accordance with an embodiment of the present invention.
  • the data link switching device 300 may be The collection module 31, the encapsulation module 32, and the transmission module 33 are included.
  • the collection module 31 is configured to collect fault status information and link port information.
  • the encapsulation module 32 is configured to encapsulate the fault state information and the link port information into a link switch message.
  • the sending module 33 is configured to send the link switch message to the network processor.
  • FIG. 4 is a schematic block diagram of a data link switching device in accordance with another embodiment of the present invention.
  • the data link switching apparatus 400 may include a receiving module 41, a selecting module 42, and a switching module 43.
  • the receiving module 41 is configured to receive the packet, and if the packet is a link switching packet, extract the link fault status information and the link port information carried in the packet.
  • the selection module 42 selects the target handover port according to the priority in the handover port database.
  • the switching module 43 is configured to switch the forwarding function of the port (ie, the faulty port) to which the link port information is directed to the target switching port selected by the selection module 42.
  • a data link switching device comprising a network processor and a central processing unit.
  • the central processor is configured to collect link fault status information and link port information of the faulty port, and send a link switch message including link fault status information and link port information to the network processor.
  • the network processor is configured to receive the packet sent by the central processing unit, and extract the link fault status information and the link port information carried in the packet when the packet is a link switching message.
  • the target switch port is selected according to the priority in the switch port database, and the forwarding function of the port (ie, the fault port) pointed to by the link port information is switched to The target switches the port.
  • a data link switching method, a data link switching device, and a data link switching device according to an embodiment of the present invention are further described in detail below with reference to the accompanying drawings.
  • a distributed OLT device may include multiple main control boards, uplink cards, passive optical network cards, and subscriber line cards.
  • 5 is a schematic block diagram of a network processor based line card in accordance with an embodiment of the present invention.
  • the network processor-based line card 500 may include a CPU chip 501 and an NP chip 502.
  • the NP chip 502 can include an OAM module Block 5021 and microcode module 5022.
  • the microcode module 5022 can include a microcode link switching information parsing module 5022_1 and a microcode link switching operation execution module 5022_2.
  • the microcode link switching information table (swap_info_table), the link switching function switch table (swap_enable_table), and the link switching port of the line card 500 may be according to specific application scenarios and functional requirements.
  • the conversion table (swap_port_convert_table) is configured.
  • FIG. 6 is a schematic diagram of a link switching port conversion table (swap_port_convert_table) according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of a link switching function switch table (swap_enable_table) according to an embodiment of the present invention
  • FIG. 8 is a diagram according to the present invention.
  • various possible link fault state information may be used (ie, optical network unit signal loss, port not established link, continuity check abnormality, two-way The forwarding detection fails.
  • the combination is used as a table lookup index.
  • the link switching operation to be performed in a specific link switching state environment is specified (the main content is the purpose for which the message should be sent after switching).
  • Port information (swap_port_id)).
  • the link switching function switch table (swap_enable_table) shown in FIG. 7 is used as a table lookup index in the table of the link switching operation type (swap_type) field in the analysis result of the link switching message shown in FIG. 9 to be described below.
  • the entry can record the enabled state of this type of link switch operation. For the most comprehensive OLT devices, all types of link switching operations are enabled.
  • the link switch port conversion table (swap_port_convert_table) shown in FIG. 6 uses the port type (port_type) and the port identity (port_id) as a lookup table index, and the entries in the table record the port and link switching operations before the link switching operation. Correspondence between ports. It should be appreciated that the port described herein may be a physical port of a general subscriber line card or a virtual port (vport) of a passive optical network line card. At the beginning of system power-on, the look-up table result of this table should be consistent with the look-up table index to indicate that there is no link switching process for the data link.
  • the optical network unit When the hardware of the distributed OLT system detects that the port does not establish a link, the optical network unit When the signal is lost, the continuity check is abnormal, or the link failure status information of the two-way forwarding detection fails, the abnormal conditions are reported to the CPU as interrupt information.
  • the CPU can make a link switch decision after receiving the interrupt information. On the one hand, it immediately starts the long-time MAC address synchronization operation, transfers all the MAC addresses learned on the problem port to the new alternate port and notifies all the cards in the distributed system to update the MAC address information.
  • the link failure state information of the port not establishing the link, the optical network unit signal loss, the continuity check abnormality or the bidirectional forwarding detection failure is encapsulated into the link switching information header, and a type is generated to include the Link switch message of the link switch information header.
  • FIG. 9 is a schematic diagram of a link switch message in accordance with an embodiment of the present invention.
  • the message is sent to the microcode module 5022 for processing by calling the packet sending interface of the OAM module 5021 of the NP chip 502.
  • the link switching information header is encapsulated according to the abnormal information, if there are two types of link fault status information belonging to different types (such as the port unchained and the bidirectional forwarding detection failure signal), Then, the two can be encapsulated into the same link switching information header, and a fixed length link switching message is sent to the microcode module 5022 through the OAM module 5021.
  • the OAM module 5021 sends a plurality of fixed-length link switching messages to the microcode module 5022 for parsing and processing.
  • the value of the info_num field in the link switching information header is 0.
  • a plurality of (for example, no more than 5) link failure state information is encapsulated into the same link switching information header, and a variable length message is sent to the microcode through the OAM module 5021.
  • Module 5022 processes that the value of the info_num field in the link switch information header is equal to the number of encapsulated link failure state information.
  • the microcode module 5022 After receiving the packet sent by the OAM module 5021, the microcode module 5022 first checks the func_type field of the packet to check whether the packet is a link switch message. If it is determined to be a link switch message, the link is executed. Switch the parsing program of the message. In the parsing procedure of the link switch packet, check whether the info_num field of the packet is 0. If the info_num field is 0, it is regarded as a fixed-length link switch packet without any packet copy and loopback processing. If the info_num field is not 0 and greater than 1, then follow The value of the info_num field copies multiple fixed-length link switch messages (where each link switch message carries only one link fault status information) and performs loopback processing.
  • Each packet triggers the processing of a fixed-length link switch message.
  • the subsequent processing manner of the microcode module 5022 is the same regardless of the fixed length link switching message sent by the OAM module 5021 or the fixed length link switching message after the loopback processing by the microcode module 5022. .
  • the microcode module 5022 may enter the link switch operation execution phase. At this time, the microcode module 5022 can use the swap_type field in the swap_info_1 as an index to find the link switching function switch table (swap_enable_table) shown in FIG. 7, and confirm whether the link switching operation of this type is enabled in the current system. If not enabled, the link switching procedure is exited directly. On the other hand, after confirming that the link switching operation of the type is enabled, the link fault status information may be used as a lookup table index to find the microcode link switching information table (swap_info_table) shown in FIG. 8 according to the setting.
  • the link fault status information may be used as a lookup table index to find the microcode link switching information table (swap_info_table) shown in FIG. 8 according to the setting.
  • the policy obtains the link switch operation information (swap_info) that should be executed under the current situation, and includes information such as the port type (port_type) and port identity (port_id) of the destination port of the link switch.
  • the microcode module 5022 uses the port type (old_port_type) and the port identity (old_port_id) of the destination port before the switching as the index according to the obtained link switching operation information swap_info, and the port type (new_port_type) of the target switching port after the switching.
  • the link switch port translation table (swap_port_convert_table) shown in FIG. 6 is set, thereby completing the specific execution of the link switching operation.
  • the microcode module 5022 needs to find the link switch function switch table (swap_enable_table) and the link switch information table (swap_info_table) again, and switch the port switch table to the link according to the result of the lookup table (swap_port_convert_table). ) Make the appropriate settings.
  • the data link switching method can implement various abnormal scenarios such as a port not establishing a link, an optical network unit signal loss, a continuity check abnormality, and a bidirectional forwarding detection failure on a general NP.
  • the method of the invention realizes the data link by using the OAM module and the microcode module shared by the general NP Fast switching of the road.
  • the data link switching method has a great advantage in the reliability and efficiency of handover.
  • the time required for the handover of the data link switching method according to the embodiment of the present invention mainly depends on the detection reporting speed of the NP for various abnormal conditions requiring handover processing.
  • the data link switching method according to an embodiment of the present invention can significantly reduce the time required for data link switching, and greatly reduce the handover process.
  • the number of lost packets According to various data link switching methods in various abnormal situations, according to the data link switching method of the embodiment of the present invention, various link switching scenarios can be uniformly processed to avoid logical conflicts and significantly reduce distributed OLT. The average packet loss time and number of lost packets required to perform data link switching in the system.
  • modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein.
  • the steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module.
  • the invention is not limited to any specific combination of hardware and software.
  • the embodiment of the present invention uses the NP widely used in the OLT device to utilize the packet aging and packet sending function of the NP OAM module to link the link fault state information required for the various types of data link received by the NP. And the link port information informs the microcode module in the form of a message.
  • the microcode module searches for a specific microcode table (for example, a link switching information table) according to the link fault state information, obtains the link switching operation information that should be taken, and performs the packet according to the obtained link switching operation information. Forward processing.
  • the embodiment of the invention provides a fast and stable data link switching method applicable to multiple fault scenarios, thereby solving the problem that the device bandwidth utilization is not high, the design is complicated, and the universality is low during the data link switching process. At least one of technical problems such as differentiating processing for multiple fault conditions is not possible.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明提供了一种数据链路切换方法、数据链路切换装置以及数据链路切换设备。所述数据链路切换方法包括：通过中央处理器收集链路故障状态信息和链路端口信息；以及通过所述中央处理器将所述链路故障状态信息和所述链路端口信息封装成链路切换报文并发送到网络处理器。所述链路切换报文用于指示所述网络处理器进行链路切换。摘图1

Description

数据链路切换方法和装置以及数据链路切换设备 技术领域
本发明涉及光通信和计算机领域，特别涉及一种数据链路切换方法和装置。
背景技术
光线路终端(optical line terminal，OLT)设备通常需要具有切换保护功能，当OLT设备中的某个端口由于硬件故障等原因出现链路异常时，需要将本应发往故障端口的报文转而发往用于替代故障端口的某个替代性端口，从而保护OLT设备的正常使用。由于在报文的数据链路切换过程中存在端口转发表的更新，并且在更新过程中无法转发数据，因此会造成部分数据丢失。在分布式交换系统中，由于在OLT设备的各外围交换芯片之间均需要同步更新端口转发表，从而导致端口转发表的更新时间较长，并且需要在更新时使用另外的机制来降低丢包率。在一些情况下，可以采用复制多份报文并且往每个数据链路发往一份报文的方式来降低丢包率,但这会造成OLT设备带宽的浪费，另一方面这种方式在技术处理上比较复杂，从而给设备的维护和可能的调试分析带来困扰。在一些情况下，可以使用特殊设计的硬件来降低丢包率，但这会增加芯片设计成本，并且软件的通用性较差，无法针对多种故障情况有针对性地设计处理方式。
发明内容
本发明提供了一种数据链路切换方法、数据链路切换装置和数据链路切换设备，以解决在数据链路切换过程中设备带宽利用率不高、设计复杂、通用性低、无法针对多种故障情况差异化地进行处理等技术问题中的至少之一。
根据本发明的一方面，提供一种数据链路切换方法，包括步骤：通过中央处理器收集链路故障状态信息和链路端口信息；以及通过所述中央处理器将所述链路故障状态信息和所述链路端口信息封装成 链路切换报文并发送到网络处理器，其中，所述链路切换报文用于指示所述网络处理器进行链路切换。
根据本发明的另一方面，提供一种数据链路切换方法，包括步骤：网络处理器接收报文，在所述报文为链路切换报文的情况下，所述网络处理器提取所述报文中携带的链路故障状态信息和链路端口信息；在所述链路故障状态信息所指示的故障状态为允许切换状态的情况下，所述网络处理器在切换端口数据库中根据优先级选取目标切换端口；以及所述网络处理器将所述链路端口信息所指向的端口的转发功能切换到所述目标切换端口。
根据本发明的另一方面，提供一种数据链路切换装置，包括：收集模块，用于收集链路故障状态信息和链路端口信息；封装模块，用于将所述链路故障状态信息和所述链路端口信息封装成链路切换报文；以及发送模块，用于将所述链路切换报文发送到网络处理器。
根据本发明的另一方面，提供一种数据链路切换装置，包括：接收模块，用于接收报文，并且在所述报文为链路切换报文的情况下，提取所述报文中携带的链路故障状态信息和链路端口信息；选择模块，在所述链路故障状态信息所指示的故障状态为允许切换状态的情况下，所述选择模块在切换端口数据库中根据优先级选取目标切换端口；以及切换模块，用于将所述链路端口信息所指向的端口的转发功能切换到选择模块选取的所述目标切换端口。
根据本发明的另一方面，通过一种数据链路切换设备，包括网络处理器和中央处理器，其中，所述中央处理器构造为，收集故障端口的链路故障状态信息和链路端口信息，并且将包含所述链路故障状态信息和所述链路端口信息的链路切换报文发送给网络处理器，并且所述网络处理器构造为，接收所述中央处理器发送的报文，在所述报文为链路切换报文的情况下，提取所述报文中携带的链路故障状态信息和链路端口信息，在所述链路故障状态信息所指示的故障状态为允许切换状态的情况下，在切换端口数据库中根据优先级选取目标切换端口，并且将所述链路端口信息所指向的端口的转发功能切换到所述目标切换端口。
附图说明
通过参考附图详细描述示例性实施例，本发明构思的上述以及其它特征将变得更加清楚。在附图中，
图1是根据本发明的实施例的数据链路切换方法的流程图；
图2是根据本发明的另一实施例的数据链路切换方法的流程图；
图3是根据本发明的实施例的数据链路切换装置的示意性框图；
图4是根据本发明另一实施例的数据链路切换装置的示意性框图；
图5是根据本发明的实施例的基于网络处理器的线卡的示意性框图；
图6是根据本发明的实施例的链路切换端口转换表的示意图；
图7是根据本发明的实施例的链路切换功能开关表的示意图；
图8是根据本发明的实施例的微码链路切换信息表的示意图；以及
图9是根据本发明的实施例的链路切换报文的示意图。
具体实施方式
下文中将参考附图并结合实施例来详细说明本发明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。
需要说明的是，本发明的说明书和权利要求书及附图中的术语“第一”、“第二”等仅用于对类似的对象进行区分，而不是进行限定也不必用于描述特定的顺序或先后次序。
图1是根据本发明的实施例的数据链路切换方法的流程图。
如图1所示，根据本发明的实施例的数据链路切换方法可以包括步骤：通过中央处理器收集链路故障状态信息和链路端口信息(S101)；以及通过中央处理器将链路故障状态信息和链路端口信息封装成链路切换报文并发送到网络处理器(S102)。链路切换报文可以用于指示网络单元进行链路切换。
通常在分布式系统OLT设备中，同步更新所有板卡的数据状态会导致链路切换所需时间过长、丢包过多，并且在不同的场景下需通过不同的方式来实现数据链路切换会导致逻辑复杂混乱。根据本发明的数据链路切换方法可以有效地避免或减少在进行链路切换时，在板卡的数据状态同步过程中的转发数据的丢包，并且可以解决不能针对具体的故障场景采用差异化解决方法的问题。
根据本发明的实施例，将中央处理器(Central Processing Unit，CPU)用作收集端口的故障信息的搜集单元，并且将网络处理器(Network Processor，NP)用作为转发数据的转发单元。CPU将各类数据链路切换所需的状态信息以报文的形式通知NP，NP依据接收的状态信息查找特定微码表，以获取所应采取的链路切换操作信息，并根据获得的链路切换操作信息对报文进行转发处理。
CPU对需要进行链路切换处理的各种故障场景下的各类状态信息进行检测，并根据状态信息生成报文。报文头部携带所有可能影响到链路切换处理方式的状态信息。
可以通过对各种链路故障的场景进行抽象来生成链路故障状态信息。根据本发明的实施例，链路故障状态信息可以包括下列链路故障之一：光网络单元信号丢失、端口不建立链路、连续性检查异常、双向转发检测失败等。
链路端口信息表示端口的身份，并且可以包括(但不限于)端口类型信息和端口身份信息。
根据本发明的实施例，可以将链路故障状态信息划分为多种类型。在同时存在多种需要处理的链路故障状态信息时，CPU可生成一个或多个报文并发送到NP进行处理。每个报文可以触发一次查找链路切换信息表的操作，从而实现了多个数据链路的切换处理。可以将链路故障状态信息按类型划分串行和并行两个集合，类型可以包括(但不限于)光网络单元信号丢失、端口不建立链路、连续性检查异常、双向转发检测失败等。在串行集合中，如果存在两个或更多类型的链路故障状态信息，则必须相应地生成多个报文。在此情况下，NP收到CPU发来的报文之后，仅查找一次链路切换信息表。在并行集合 中，如果存在两个或更多类型的链路故障状态信息，则可以仅生成一个报文并在报文头部中同时携带各种链路故障状态信息。在此情况下，NP收到CPU发来的报文之后，将多次查找链路切换信息表，并按链路切换信息表的查表结果执行相应的报文转发特殊处理，从而可以提升链路切换的响应速度。
根据本发明的实施例，可使用头部长度可变的报文来封装链路故障状态信息，从而可以只需生成一个携带了多个类型的链路故障状态信息的报文。NP收到报文之后，如果检测到报文头部的是长度可变的头部，则需要从报文头部提取出所有类型(例如，N个类型)的链路故障状态信息，并复制N份以进行环回处理。每份报文根据该份报文的序号取用报文头部的不同偏移位置处的链路故障状态信息，并进行一次查找链路切换信息表的操作，然后依据查表结果执行相应的链路切换操作。
图2是根据本发明另一实施例的数据链路切换方法的流程图。
如图2所示，根据本发明另一实施例的数据链路切换方法可以包括步骤：网络处理器接收报文，在该报文为链路切换报文的情况下，提取该报文中携带的链路故障状态信息和链路端口信息(S201)；以及在链路故障状态信息所指示的故障状态为允许切换状态的情况下，网络处理器在切换端口数据库中根据优先级选取目标切换端口(S202)。
根据本发明的实施例，通过利用NP的操作管理维护(Operation Administration and Maintenance，OAM)模块的构包及发包功能，将NP接收到的各种类型的数据链路切换所需的链路故障状态信息和链路端口信息以报文的形式通知微码模块。微码模块依据链路故障状态信息查找特定的微码表，以获取所应采取的链路切换操作信息，并根据获得的链路切换操作信息对报文进行转发处理。
OAM模块将接收到的报文转发给NP的微码模块。微码模块对报文进行解析，以得到各种用于执行链路切换操作的链路故障状态信息和链路端口信息。微码模块依据链路故障状态信息和链路端口信息查找特定的微码表(例如，链路切换信息表)，以得到用于执行链路切 换操作的操作信息。微码模块根据查找链路切换信息表所得到的操作信息，对需要改变转发路径的报文做相应的处理，从而不再将报文转发给查找转发信息库(Forwarding Information Base，FIB)表所确定的端口，而是转发给查找链路切换信息表所得到的端口。
微码模块对OAM模块发来的报文进行解析。微码模块可对OAM模块发来的多种类型的报文进行处理，例如，MAC老化或快速老化报文、1588报文等等。可以通过自定义的报文头部中的特定字段来区分报文的类型，不同类型的报文可以对应不同的解析方式及处理代码。当微码模块接收到链路切换报文时将执行用于实现链路切换功能的微码代码。
微码模块依据自定义的报文头部中特定的字段信息判定该报文是链路切换报文后，可以进一步判断该报文的头部是固定长度的头部还是长度可变的头部。如果是固定长度的头部，则根据该头部包含的链路故障状态信息类型的数量，确定需执行查找链路切换信息表的操作的次数；如果是长度可变的头部，则将报文复制多份并进行环回处理。
在微码模块根据得到的链路故障状态信息和链路端口信息查找链路切换信息表之前，需要先查找链路切换操作开关表来确定是否允许执行当前的链路故障状态信息所对应的链路切换操作。如果当前的链路故障状态信息所对应的链路切换操作是禁止执行的，则微码模块直接退出相关的处理流程，并且实际上不执行任何链路切换操作。如果当前的链路故障状态信息所对应的链路切换操作是允许执行的，则微码模块继续执行链路切换处理的相关操作。
根据本发明的实施例，在微码模块根据得到的链路故障状态信息查找链路切换信息表时，可能同时得到多种可选的链路切换操作。例如，当端口A由于被动或主动原因而无法正常转发报文时，既可以选择将报文的转发切换至端口B，也可以选择切换至端口C。此时，需要根据各个可选的链路切换操作的优先级来确定实际将要执行的链路切换操作。可以选择具有最高优先级的链路切换操作作为下一步将要实际执行的链路切换操作。
根据本发明的实施例，链路切换操作的优先级可以是通过包括两个因素的函数计算得到的结果。一个因素是链路切换操作的默认优先级，链路切换操作的默认优先级越高，该链路切换操作通过所述函数计算得到的最终的优先级也越高。另一个因素是链路切换操作在一个定的时间间隔(即，单位时间)内被执行的次数，如果执行的次数越多，则通过所述函数计算得到的最终的优先级越低。通过这种优先级控制，可以保证以最优的方式执行链路切换操作，同时在一定程度上还可以避免可能存在的链路切换操作的来回震荡，从而实现稳定、可靠的链路切换操作。
根据本发明的实施例，可以设置数值作为链路切换操作的优先级。设置的数值可以是固定的数值，即，链路切换操作的默认优先级。可替换地，设置的数值可以根据待切换的端口处理能力而(例如，定期地)变化。当待切换的端口处理能力提高时设置的数值可以增大，当待切换的端口处理能力降低时设置的数值可以减小。根据本发明的实施例，可以选择具有最大设置数值的待切换端口为目标切换端口，即，可以选择具有最高优先级的链路切换操作作为下一步将要实际执行的链路切换操作。
根据本发明的实施例，NP根据得到链路故障状态信息和链路端口信息进行切换转发。根据报文中携带的链路故障状态信息和链路端口信息查找FIB表以得到需要被切换的端口(即，故障端口)，将故障端口作为索引查找链路切换信息表，以得到链路切换之后报文实际将要转发到的端口，从而最终实现链路切换操作。通过引入链路切换信息表这种微码表，无需等待分布式系统中各板卡的大量MAC地址同步操作最终完成，就可以将受影响报文的转发的目的端口(即，故障端口)就快速地切换至新的目的端口，从而实现了在不对二层报文的一般处理流程做大幅改变的情况下，对报文转发的目的端口的快速切换。因此，在保证了可靠性的同时，有效地提升了分布式系统时数据链路切换的性能。
图3是根据本发明的实施例的数据链路切换装置的示意性框图。
如图3所示，根据本发明的实施例的数据链路切换装置300可 以包括收集模块31、封装模块32和发送模块33。收集模块31用于收集故障状态信息和链路端口信息。封装模块32用于将故障状态信息和链路端口信息封装成链路切换报文。发送模块33用于将链路切换报文发送到网络处理器。
图4是根据本发明另一实施例的数据链路切换装置的示意性框图。
如图4所示，根据本发明另一实施例的数据链路切换装置400可以包括接收模块41、选择模块42和切换模块43。接收模块41用于接收报文，并且在该报文为链路切换报文的情况下，提取该报文中携带的链路故障状态信息和链路端口信息。在链路故障状态信息所指示的故障状态为允许切换状态的情况下，选择模块42在切换端口数据库中根据优先级选取目标切换端口。切换模块43用于将链路端口信息所指向的端口(即，故障端口)的转发功能切换到选择模块42选取的目标切换端口。
根据本发明的实施例，还提供了一种数据链路切换设备，包括网络处理器和中央处理器。中央处理器构造为，收集故障端口的链路故障状态信息和链路端口信息，并且将包含链路故障状态信息和链路端口信息的链路切换报文发送给网络处理器。网络处理器构造为，接收中央处理器发送的报文，在该报文为链路切换报文的情况下，提取该报文中携带的链路故障状态信息和链路端口信息，在链路故障状态信息所指示的故障状态为允许切换状态的情况下，在切换端口数据库中根据优先级选取目标切换端口，并且将链路端口信息所指向的端口(即，故障端口)的转发功能切换到该目标切换端口。
下面结合附图对根据本发明的实施例的数据链路切换方法、数据链路切换装置和数据链路切换设备作进一步的详细描述。
在一个光网络接入系统中，一个分布式OLT设备可以包括多块主控板、上联卡、无源光网络卡和用户线卡。图5是根据本发明的实施例的基于网络处理器的线卡的示意性框图。
如图5所示，根据本发明的实施例的基于网络处理器的线卡500可以包括CPU芯片501和NP芯片502。NP芯片502可以包括OAM模 块5021和微码模块5022。微码模块5022可以包括微码链路切换信息解析模块5022_1和微码链路切换操作执行模块5022_2。
根据本发明的实施例，在初始化阶段可以根据具体的应用场景及功能要求，对线卡500的微码链路切换信息表(swap_info_table)、链路切换功能开关表(swap_enable_table)及链路切换端口转换表(swap_port_convert_table)进行配置。
图6是根据本发明的实施例的链路切换端口转换表(swap_port_convert_table)的示意图，图7是根据本发明的实施例的链路切换功能开关表(swap_enable_table)的示意图，并且图8是根据本发明的实施例的微码链路切换信息表(swap_info_table)的示意图。
在图8所示的微码链路切换信息表(swap_info_table)中，可以根据各种可能的链路故障状态信息(即，光网络单元信号丢失、端口不建立链路、连续性检查异常、双向转发检测失败)组合作为查表索引，通过适当设置各个条目的内容，来指定特定的链路切换状态环境下所需执行的链路切换操作(其主要内容是切换之后报文应发往的目的端口信息(swap_port_id))。
图7所示的链路切换功能开关表(swap_enable_table)以下面将要描述的图9所示的链路切换报文的解析结果中的链路切换操作类型(swap_type)字段作为查表索引，表中的条目可以记录该种类型的链路切换操作的启用状态。对于功能支持最为全面的OLT设备，所有类型的链路切换操作均为启用状态。
图6所示的链路切换端口转换表(swap_port_convert_table)以端口类型(port_type)和端口身份(port_id)作为查表索引，表中的条目记录了链路切换操作之前的端口与链路切换操作之后的端口之间的对应关系。应当认识到，这里所述的端口可以是一般用户线卡的物理端口，也可以是无源光网络线卡的虚拟端口(vport)。在系统上电之初，此表格的查表结果与查表索引应保持一致，以表示没有数据链路的链路切换处理。
当分布式OLT系统的硬件检测到端口不建立链路、光网络单元 信号丢失、连续性检查异常或者双向转发检测失败的链路故障状态信息时，会将这些异常情况以中断信息的形式上报给CPU。CPU收到中断信息之后可以做出链路切换决策。一方面立即启动耗时较长的MAC地址同步操作，将问题端口上学习到的所有MAC地址转移至新的替代端口上并通知分布式系统内的所有板卡更新MAC地址信息。另一方面则将端口不建立链路、光网络单元信号丢失、连续性检查异常或者双向转发检测失败的链路故障状态信息封装到链路切换信息头部之中，并生成一个类型为包含该链路切换信息头部的链路切换报文。
图9是根据本发明的实施例的链路切换报文的示意图。
参照图5和图9，通过调用NP芯片502的OAM模块5021的发包接口，将报文发送给微码模块5022进行处理。为提升性能和效率，在根据异常信息封装链路切换信息头部时，如果同时存在两种分属不同类型的链路故障状态信息(比如同时存在端口不建链及双向转发检测失败信号)，则可将两者封装到同一个链路切换信息头部之中，并通过OAM模块5021将一份固定长度的链路切换报文发送给微码模块5022。如果同时存在同一类型的多个链路故障状态信息，可以有两种处理方法：一是对每个链路故障状态信息进行遍历式的逐一处理，将各个链路故障状态信息分别封装在多个链路切换信息头部之中，通过OAM模块5021将多份固定长度的链路切换报文发送给微码模块5022进行解析和处理，此时链路切换信息头部中info_num字段的数值为0；二是将多个(例如，不多于5个)链路故障状态信息封装到同一个链路切换信息头部之中，通过OAM模块5021将一份长度可变的报文发送给微码模块5022处理，此时链路切换信息头部中的info_num字段的数值等于封装的链路故障状态信息的数量。
微码模块5022接收到OAM模块5021发来的报文之后，首先检查报文的func_type字段，以检查该报文是否为链路切换报文，如果确定为链路切换报文，则执行链路切换报文的解析程序。在链路切换报文的解析程序中，先检查报文的info_num字段是否为0，如果info_num字段为0则视为固定长度的链路切换报文，而无需做任何报文复制和环回处理；如果info_num字段不为0且大于1，则按照 info_num字段的数值复制多份固定长度的链路切换报文(其中每份链路切换报文只携带一个链路故障状态信息)并进行环回处理。每份报文都触发一次针对固定长度的链路切换报文的处理。无论对于OAM模块5021发出的固定长度的链路切换报文，还是对于经微码模块5022复制环回处理之后的固定长度的链路切换报文，微码模块5022的后续处理方式都是相同的。
微码模块5022对OAM模块5021发来的链路切换报文的链路故障状态信息的解析完成之后，可以进入到链路切换操作执行阶段。此时微码模块5022可以使用swap_info_1中的swap_type字段作为索引，查找图7所示的链路切换功能开关表(swap_enable_table)，确认该类型的链路切换操作在当前系统中是否启用。如果未启用，则直接退出链路切换操作程序。另一方面，在确认该类型的链路切换操作已启用之后，可以按照链路故障状态信息作为查表索引，查找图8所示的微码链路切换信息表(swap_info_table)，以根据设定的策略获得在当前情况下所应执行的链路切换操作信息(swap_info)，其中包含链路切换的目的端口的端口类型(port_type)和端口身份(port_id)等信息。之后，微码模块5022根据所获得链路切换操作信息swap_info，以切换前的目的端口的端口类型(old_port_type)和端口身份(old_port_id)作为索引，以切换后的目标切换端口的端口类型(new_port_type)和端口身份(new_port_id)作为条目内容，对图6所示的链路切换端口转换表(swap_port_convert_table)进行设置，从而完成链路切换操作的具体执行。如果swap_info_2字段中的数据有效，则微码模块5022还需要再次查找链路切换功能开关表(swap_enable_table)和链路切换信息表(swap_info_table)，并根据查表结果对链路切换端口转换表(swap_port_convert_table)进行适当设置。
综上，根据本发明的实施例的数据链路切换方法，可以在一般的NP上实现端口不建立链路、光网络单元信号丢失、连续性检查异常、双向转发检测失败等各种异常场景下的数据链路快速切换的功能。本发明方法利用一般NP所共有的OAM模块和微码模块来实现数据链 路的快速切换。相比针对每种异常情况分别设置一套链路切换机制、需要等待分布式系统中所有板卡MAC地址同步操作全部完成后再进行数据链路切换的常见处理方式，根据本发明的实施例的数据链路切换方法在切换的可靠性及效率上有较大优势。根据本发明的实施例的数据链路切换方法的切换所需时间主要取决于NP对于需要进行切换处理的各种异常情况的检测上报速度。当分布式系统中执行数据链路切换所需要同步的MAC地址的数量较多时，利用根据本发明的实施例数据链路切换方法，可显著减小数据链路切换所需时间，大幅降低切换过程中的丢包数量。综合各种异常情况下数据链路切换的各种情形，根据本发明的实施例的数据链路切换方法，可对各种链路切换场景做统一处理以避免逻辑冲突、显著减小分布式OLT系统中执行数据链路切换所需的平均丢包时间及丢包数量。
显然，本领域的技术人员应该明白，上述的本发明的各模块或各步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，可选地，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，并且在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本发明不限制于任何特定的硬件和软件结合。
本发明的实施例通过使用在OLT设备中得到广泛应用的NP，利用NP的OAM模块的构包及发包功能，将NP接收到的各种类型的数据链路切换所需的链路故障状态信息和链路端口信息以报文的形式通知微码模块。微码模块依据链路故障状态信息查找特定的微码表(例如，链路切换信息表)，以获取所应采取的链路切换操作信息，并根据获得的链路切换操作信息对报文进行转发处理。本发明的实施例提供了一种能适用于多种故障场景的、快速稳定的数据链路切换方法，从而解决了在数据链路切换过程中设备带宽利用率不高、设计复杂、通用性低、无法针对多种故障情况差异化处理等技术问题中的至少之一。
以上该仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (14)

  1. 一种数据链路切换方法，包括步骤：
    通过中央处理器收集链路故障状态信息和链路端口信息；以及
    通过所述中央处理器将所述链路故障状态信息和所述链路端口信息封装成链路切换报文并发送到网络处理器，
    其中，所述链路切换报文用于指示所述网络处理器进行链路切换。
  2. 根据权利要求1所述的方法，其中，所述链路故障状态信息包括下列链路故障之一：光网络单元信号丢失、端口不建立链路、连续性检查异常、双向转发检测失败。
  3. 根据权利要求1所述的方法，其中，所述链路端口信息包括端口类型信息和端口身份信息。
  4. 根据权利要求2所述的方法，其中，
    所述链路切换报文包含同一种链路故障状态下的所有故障端口的链路端口信息。
  5. 根据权利要求1所述的方法，其中，
    所述链路切换报文包含单个端口的链路端口信息和所述链路故障状态信息。
  6. 一种数据链路切换方法，包括步骤：
    网络处理器接收报文，在所述报文为链路切换报文的情况下，所述网络处理器提取所述报文中携带的链路故障状态信息和链路端口信息；
    在所述链路故障状态信息所指示的故障状态为允许切换状态的情况下，所述网络处理器在切换端口数据库中根据优先级选取目标切 换端口；以及
    所述网络处理器将所述链路端口信息所指向的端口的转发功能切换到所述目标切换端口。
  7. 根据权利要求6所述的方法，其中，将所述链路端口信息所指向的端口的转发功能切换到所述目标切换端口的步骤包括：
    通过转发表将需要通过所述链路端口信息所指向的端口转发的数据转发到所述目标切换端口，其中所述转发表包含有所述链路端口信息与所述目标切换端口的信息；以及
    通过所述目标切换端口转发所述数据。
  8. 根据权利要求6所述的方法，其中，在将所述链路端口信息所指向的端口的转发功能切换到所述目标切换端口之前还包括步骤：
    将所述链路端口信息所指向的端口登记到转发表中的源端口字段，并且将所述目标切换端口登记到所述转发表中的目的切换端口字段。
  9. 根据权利要求6所述的方法，其中，在切换端口数据库中根据优先级选取目标切换端口的步骤包括：
    为切换端口设置固定的数值作为该切换端口的优先级；以及
    选择数值大的优先级所对应的切换端口为所述目标切换端口。
  10. 根据权利要求6所述的方法，其中，在切换端口数据库中根据优先级选取目标切换端口的步骤包括：
    为切换端口设置数值作为该切换端口的优先级；
    根据切换端口处理能力定期地改变与该切换端口对应的设置的数值，其中当切换端口的处理能力提高时设置的数值增大，并且当切换端口的处理能力降低时设置的数值减小；以及
    选择数值大的优先级所对应的切换端口为所述目标切换端口。
  11. 根据权利要求6所述的方法，其中，通过包含两个因素的函数来计算得到所述优先级，所述两个因素分别为切换端口的默认优先级以及切换端口在一定的时间间隔内被选为目标切换端口的次数，所述默认优先级与计算得到的所述优先级成正比，并且所述次数与计算得到的所述优先级成反比。
  12. 一种数据链路切换装置，包括：
    收集模块，用于收集链路故障状态信息和链路端口信息；
    封装模块，用于将所述链路故障状态信息和所述链路端口信息封装成链路切换报文；以及
    发送模块，用于将所述链路切换报文发送到网络处理器。
  13. 一种数据链路切换装置，包括：
    接收模块，用于接收报文，并且在所述报文为链路切换报文的情况下，提取所述报文中携带的链路故障状态信息和链路端口信息；
    选择模块，在所述链路故障状态信息所指示的故障状态为允许切换状态的情况下，所述选择模块在切换端口数据库中根据优先级选取目标切换端口；以及
    切换模块，用于将所述链路端口信息所指向的端口的转发功能切换到选择模块选取的所述目标切换端口。
  14. 一种数据链路切换设备，包括网络处理器和中央处理器，其中，
    所述中央处理器构造为，收集故障端口的链路故障状态信息和链路端口信息，并且将包含所述链路故障状态信息和所述链路端口信息的链路切换报文发送给网络处理器，并且
    所述网络处理器构造为，接收所述中央处理器发送的报文，在所述报文为链路切换报文的情况下，提取所述报文中携带的链路故障状态信息和链路端口信息，在所述链路故障状态信息所指示的故障状态为允许切换状态的情况下，在切换端口数据库中根据优先级选取目 标切换端口，并且将所述链路端口信息所指向的端口的转发功能切换到所述目标切换端口。
PCT/CN2017/116457 2016-12-15 2017-12-15 数据链路切换方法和装置以及数据链路切换设备 Ceased WO2018108149A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611160771.2 2016-12-15
CN201611160771.2A CN108234301B (zh) 2016-12-15 2016-12-15 一种数据链路切换方法和装置

Publications (1)

Publication Number Publication Date
WO2018108149A1 true WO2018108149A1 (zh) 2018-06-21

Family

ID=62558038

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/116457 Ceased WO2018108149A1 (zh) 2016-12-15 2017-12-15 数据链路切换方法和装置以及数据链路切换设备

Country Status (2)

Country Link
CN (1) CN108234301B (zh)
WO (1) WO2018108149A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111277440A (zh) * 2020-01-20 2020-06-12 新华三信息安全技术有限公司 网络设备
CN112383471A (zh) * 2020-10-30 2021-02-19 新华三信息技术有限公司 一种刀箱链路管理方法、装置、设备及机器可读存储介质
CN113784411A (zh) * 2021-09-13 2021-12-10 展讯通信（上海）有限公司 链路质量评估方法、链路切换方法、设备和存储介质
CN113852547A (zh) * 2021-09-10 2021-12-28 锐捷网络股份有限公司 一种报文转发方法、装置、线卡及存储介质
CN114205263A (zh) * 2021-12-08 2022-03-18 中国信息通信研究院 用于Ether CAT网络的通信方法、系统和存储介质
CN114430310A (zh) * 2020-10-29 2022-05-03 南京中兴新软件有限责任公司 信息处理方法及其装置、宽带接入服务器、存储介质
CN115515199A (zh) * 2022-09-23 2022-12-23 北京东土军悦科技有限公司 一种网络选择方法及网关

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110138657B (zh) * 2019-05-13 2021-11-09 北京东土军悦科技有限公司 交换机间的聚合链路切换方法、装置、设备及存储介质
CN115473840A (zh) * 2021-06-11 2022-12-13 华为技术有限公司 一种报文发布方法、转发路径处理方法及装置
CN114938346B (zh) * 2022-05-13 2023-06-06 苏州盛科通信股份有限公司 一种芯片和cpu协同实现业务快速切换的方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101860492A (zh) * 2010-06-28 2010-10-13 中兴通讯股份有限公司 快速切换的方法、装置和系统
CN102857419A (zh) * 2012-10-12 2013-01-02 华为技术有限公司 链路聚合端口故障的处理方法和装置
CN105281927A (zh) * 2014-05-28 2016-01-27 中兴通讯股份有限公司 多链路保护倒换的方法及装置
CN105871674A (zh) * 2015-01-23 2016-08-17 华为技术有限公司 环保护链路故障保护方法、设备及系统

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006180162A (ja) * 2004-12-22 2006-07-06 Nec Corp パケット交換装置およびパケット交換方法
CN101126997B (zh) * 2007-09-26 2010-06-16 中兴通讯股份有限公司 一种网络处理器微引擎活性检测的方法
CN101277269A (zh) * 2008-05-06 2008-10-01 杭州华三通信技术有限公司 实现可靠通信的终端、终端切换方法及系统和适用该系统的方法
US8554075B2 (en) * 2009-01-13 2013-10-08 Hitachi, Ltd. Communication system, subscriber accommodating apparatus and communication method
US8654630B2 (en) * 2010-03-19 2014-02-18 Brocade Communications Systems, Inc. Techniques for link redundancy in layer 2 networks
CN102368722B (zh) * 2011-12-01 2013-06-05 盛科网络（苏州）有限公司 以太网环网保护中快速切换的芯片实现方法及系统
CN102420768A (zh) * 2011-12-15 2012-04-18 北京星网锐捷网络技术有限公司 链路切换方法及装置、数据传输方法及装置、网络设备
WO2013155675A1 (en) * 2012-04-18 2013-10-24 Telefonaktiebolaget L M Ericsson (Publ) System and method to send oam packets on redundancy paths
CN103036716B (zh) * 2012-12-12 2018-06-22 中兴通讯股份有限公司 链路切换方法及系统
CN103200048B (zh) * 2013-04-02 2018-07-13 中兴通讯股份有限公司 一种网络处理器异常检测方法、装置及网络处理设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101860492A (zh) * 2010-06-28 2010-10-13 中兴通讯股份有限公司 快速切换的方法、装置和系统
CN102857419A (zh) * 2012-10-12 2013-01-02 华为技术有限公司 链路聚合端口故障的处理方法和装置
CN105281927A (zh) * 2014-05-28 2016-01-27 中兴通讯股份有限公司 多链路保护倒换的方法及装置
CN105871674A (zh) * 2015-01-23 2016-08-17 华为技术有限公司 环保护链路故障保护方法、设备及系统

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111277440A (zh) * 2020-01-20 2020-06-12 新华三信息安全技术有限公司 网络设备
CN111277440B (zh) * 2020-01-20 2022-04-08 新华三信息安全技术有限公司 网络设备
CN114430310A (zh) * 2020-10-29 2022-05-03 南京中兴新软件有限责任公司 信息处理方法及其装置、宽带接入服务器、存储介质
CN112383471A (zh) * 2020-10-30 2021-02-19 新华三信息技术有限公司 一种刀箱链路管理方法、装置、设备及机器可读存储介质
CN113852547A (zh) * 2021-09-10 2021-12-28 锐捷网络股份有限公司 一种报文转发方法、装置、线卡及存储介质
CN113852547B (zh) * 2021-09-10 2023-07-25 锐捷网络股份有限公司 一种报文转发方法、装置、线卡及存储介质
CN113784411A (zh) * 2021-09-13 2021-12-10 展讯通信（上海）有限公司 链路质量评估方法、链路切换方法、设备和存储介质
CN114205263A (zh) * 2021-12-08 2022-03-18 中国信息通信研究院 用于Ether CAT网络的通信方法、系统和存储介质
CN114205263B (zh) * 2021-12-08 2023-10-13 中国信息通信研究院 用于Ether CAT网络的通信方法、系统和存储介质
CN115515199A (zh) * 2022-09-23 2022-12-23 北京东土军悦科技有限公司 一种网络选择方法及网关

Also Published As

Publication number Publication date
CN108234301A (zh) 2018-06-29
CN108234301B (zh) 2022-04-19

Similar Documents

Publication Publication Date Title
WO2018108149A1 (zh) 数据链路切换方法和装置以及数据链路切换设备
US10938710B2 (en) Protection switching method and system, and network device
US10534601B1 (en) In-service software upgrade of virtual router with reduced packet loss
EP2204952B1 (en) Automatic protection switching
US8369211B2 (en) Network distribution prevention when virtual chassis system undergoes splits and merges
US20220345404A1 (en) Packet sending method, routing entry generation method, apparatus, and storage medium
CN110519079B (zh) 一种数据转发方法、装置、网板、网络设备和存储介质
CN105791126B (zh) 一种三态内容可寻址存储器tcam查表方法及装置
CN110417569A (zh) 一种网络链路故障处理方法和隧道端点设备
CN102006189A (zh) 用于双机冗余备份的主用接入服务器确定方法及装置
US20220311702A1 (en) Efficient propagation of fault routing notifications
US20240235992A9 (en) Packet transmission method and apparatus, and device
CN113839862A (zh) Mclag邻居之间同步arp信息的方法、系统、终端及存储介质
US20200136912A1 (en) Method, Device, and System for Implementing MUX Machine
CN108989204B (zh) 一种链路状态确定方法及设备
US20120294307A1 (en) Apparatus and Method for Identifier Management
CN106850268B (zh) 一种线性保护倒换的实现装置及方法
US9866428B2 (en) Fabric switchover for systems with control plane and fabric plane on same board
CN109286564B (zh) 一种报文转发方法及装置
CN112887201A (zh) 基于vrrp的接口更新方法、装置及存储介质
EP2953299B1 (en) Protection switching method, system and node
CN114760245A (zh) 一种堆叠成员设备及其转发广播报文的方法
CN108965126B (zh) 一种报文转发方法及装置
US12425431B2 (en) Fault tolerance methodology to detect and recover network uplink fault automatically
EP3200403B1 (en) Packet forwarding method and device, sdn and system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17880927

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17880927

Country of ref document: EP

Kind code of ref document: A1