Congestion traffic migration method and device, network equipment and computer readable storage medium

Abstract

The invention provides a congestion traffic migration method and device, network equipment and a computer readable storage medium, and relates to the field of communication. A first device regularly counts packet loss statistical information of a port accessed by each access device, and then determines at least one packet loss convergence group and a migration VLAN corresponding to the packet loss convergence group under the condition that a second congestion VLAN group corresponding to the packet loss convergence group on a second device is empty based on the packet loss statistical information. And then, based on the port of the peer-to-peer link and the destination IP address and the destination MAC address of each packet loss message belonging to the migration VLAN, updating the MAC address table and the ARP table, so that the flow needing to reach the destination IP address and the destination MAC address of each packet loss message belonging to the migration VLAN can be subsequently migrated to a second device for forwarding through the peer-to-peer link. Congestion and packet loss of the first device are relieved.

Description

Technical Field

[0001] The present invention relates to the field of communications, and in particular to a congestion traffic migration method, device, network equipment and computer-readable storage medium. Background Art

[0002] In the current network deployment design, especially in the deployment design of data centers and campus networks, MLAG (Multi-chassis Link Aggregation Group) technology has become a mainstream solution. Compared with the traditional active-standby stacking technology, MLAG provides a more flexible network architecture and can improve network availability and redundancy without sacrificing performance.

[0003] However, in the MLAG scenario, since the two devices each maintain an independent control plane, the traffic loads on the two devices are unbalanced, and it is easy for one end to have excess bandwidth while the other end may have experienced congestion and packet loss. Summary of the invention

[0004] The purpose of the present invention is to provide a congestion traffic migration method, device, network equipment and computer-readable storage medium to improve the problems existing in the prior art.

[0005] The embodiments of the present invention can be implemented as follows:

[0006] In a first aspect, the present invention provides a congested traffic migration method, which is applied to a first device of an MLAG system, wherein the MLAG system further includes a second device that establishes a peer link with the first device; at least one access device is dual-homed to the MLAG system; the first device maintains a first congested VLAN group corresponding to each aggregation group, and stores a second congested VLAN group corresponding to each aggregation group on the second device; the method includes:

[0007] Regularly obtain the packet loss statistics of each access device's access port;

[0008] Based on the packet loss statistics of each of the ports and the aggregation group to which it belongs, determine the packet loss data of at least one packet loss aggregation group, wherein the packet loss data includes the destination IP address, the destination MAC address and the VLAN ID of the plurality of packet loss messages;

[0009] For each of the packet loss aggregation groups, when the second congested VLAN group corresponding to the packet loss aggregation group is empty, determining the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group;

[0010] Based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet belonging to the migration VLAN, the MAC address table and the ARP table are updated;

[0011] The migration VLAN is added to the first congested VLAN group corresponding to the packet loss aggregation group, and the first congested VLAN group corresponding to the packet loss aggregation group is sent to the second device.

[0012] Optionally, the step of determining a migration VLAN corresponding to the packet loss aggregation group based on a VLAN identifier of each packet loss message in the packet loss aggregation group includes:

[0013] Based on the VLAN identifier of each packet loss message in the packet loss aggregation group, determining the number of packet losses corresponding to each packet loss VLAN identifier in the packet loss aggregation group;

[0014] Based on the number of packet losses corresponding to each packet loss VLAN identifier, all packet loss VLAN identifiers in the packet loss aggregation group are arranged in descending order to obtain a sorted list;

[0015] The first packet loss VLAN identifier in the sorted list is used as the pending VLAN;

[0016] Determine whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device;

[0017] If yes, then use the next packet loss VLAN identifier of the pending VLAN in the sorted list as the new pending VLAN, and return to the step of determining whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device;

[0018] If not, the pending VLAN is used as the migration VLAN corresponding to the packet loss aggregation group.

[0019] Optionally, the step of updating the MAC address table and the ARP table based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet belonging to the migration VLAN includes:

[0020] Obtaining the destination IP address and destination MAC address of each lost packet belonging to the migration VLAN;

[0021] For each of the lost packet messages, searching the MAC address table based on the destination MAC address and VLAN identifier of the lost packet message, and searching the ARP table based on the destination IP address and destination MAC address of the lost packet message;

[0022] If a matching dynamic MAC table entry can be found in the MAC address table and a matching dynamic ARP table entry can be found in the ARP table, after deleting the found dynamic MAC table entry and the dynamic ARP table entry, a static MAC table entry is created in the MAC address table based on the port of the peer link and the destination MAC address and VLAN identifier of the lost packet message, and a static ARP table entry is created in the ARP table based on the destination IP address and destination MAC address of the lost packet message. Optionally, after the step of creating a static MAC table entry in the MAC address table based on the port of the peer link and the destination MAC address and VLAN identifier of the lost packet message, and creating a static ARP table entry in the ARP table based on the destination IP address and destination MAC address of the lost packet message, the method further includes:

[0023] Using the source MAC address field and the VLAN ID field as matching fields, IPFIX traffic collection starts at each port of the packet loss aggregation group, and outputs a MAC address flow table;

[0024] Performing aging processing on each flow table entry of the MAC address flow table based on a preset flow table entry aging time;

[0025] For each static MAC table entry to be aged including the port of the peer link and the migration VLAN in the MAC address table, at every preset MAC address aging time, searching the MAC address flow table for a MAC address in the static MAC table entry to be aged; wherein the flow entry aging time is less than or equal to the MAC address aging time;

[0026] If not, the static MAC table entry to be aged is deleted from the MAC address table.

[0027] Optionally, after the step of creating a static MAC table entry in the MAC address table based on the port of the peer link and the destination MAC address and VLAN identifier of the lost packet message, and creating a static ARP table entry in the ARP table based on the destination IP address and destination MAC address of the lost packet message, the method further includes:

[0028] For each IP address of the static ARP table entry in the ARP table, sending an ARP message to the IP address at a preset period;

[0029] If no ARP response message replying to the IP address is received within the preset waiting time, the static ARP table entry including the IP address is deleted from the ARP table.

[0030] Optionally, after the step of adding the migration VLAN to the first congested VLAN group corresponding to the packet loss aggregation group, and sending the first congested VLAN group corresponding to the packet loss aggregation group to the second device, the method further includes:

[0031] Based on the migration VLAN corresponding to the packet loss aggregation group, configure a traffic matching rule in the outbound direction of the port of the peer link;

[0032] At every preset statistical interval, the remaining bandwidth of the packet loss aggregation group is obtained, and the traffic sending rate belonging to the migration VLAN and sent from the port of the peer link based on the traffic matching rule is obtained;

[0033] If the remaining bandwidth is greater than the traffic sending rate and the difference between the two is not less than a set threshold, searching for at least one matching static MAC table entry from the MAC address table based on the migration VLAN and deleting the entry;

[0034] Based on the MAC address in each static MAC table entry found, the matching static ARP table entry is found from the ARP table and deleted;

[0035] After deleting the migration VLAN from the first congested VLAN group corresponding to the packet loss aggregation group, the first congested VLAN group corresponding to the packet loss aggregation group is sent to the second device.

[0036] In a second aspect, the present invention provides a congestion traffic migration device, which is applied to a first device of an MLAG system, wherein the MLAG system further includes a second device that establishes a peer link with the first device; at least one access device is dual-homed to the MLAG system; the first device maintains a first congested VLAN group corresponding to each aggregation group, and stores a second congested VLAN group corresponding to each aggregation group on the second device; the device includes:

[0037] The packet loss statistics module is used to periodically obtain the packet loss statistics information of the port to which each access device is connected;

[0038] A data sorting module, configured to determine packet loss data of at least one packet loss aggregation group based on packet loss statistics of each of the ports and the aggregation group to which it belongs, wherein the packet loss data includes a destination IP address, a destination MAC address and a VLAN identifier of a plurality of packet loss messages;

[0039] A migration determination module, for each of the packet loss aggregation groups, when the second congested VLAN group corresponding to the packet loss aggregation group is empty, based on the VLAN identifier of each packet loss message in the packet loss aggregation group, determines the migration VLAN corresponding to the packet loss aggregation group;

[0040] A traffic migration module, configured to update a MAC address table and an ARP table based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet message belonging to the migration VLAN;

[0041] The traffic migration module is further configured to add the migration VLAN to the first congested VLAN group corresponding to the packet loss aggregation group, and send the first congested VLAN group corresponding to the packet loss aggregation group to the second device.

[0042] In an optional implementation manner, the migration determination module is used to determine the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group, and is specifically used to:

[0043] Based on the VLAN ID of each packet loss message in the packet loss aggregation group, determining the number of packet loss for each packet loss VLAN ID in the packet loss aggregation group;

[0044] Based on the number of packet losses of each packet loss VLAN ID, all packet loss VLAN IDs in the packet loss aggregation group are arranged in descending order to obtain a sorted list;

[0045] The first packet loss VLAN identifier in the sorted list is used as the pending VLAN;

[0046] Determine whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device;

[0047] If yes, then use the next packet loss VLAN identifier of the pending VLAN in the sorted list as the new pending VLAN, and return to the step of determining whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device;

[0048] If not, the pending VLAN is used as the migration VLAN corresponding to the packet loss aggregation group.

[0049] In a third aspect, the present invention provides a network device, comprising: a memory and a processor, wherein the memory stores a software program, and when the network device is running, the processor executes the software program to implement the congestion traffic migration method as described in the first aspect above.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the congestion traffic migration method described in the first aspect is implemented.

[0051] Compared with the prior art, the embodiments of the present invention provide a congestion traffic migration method, apparatus, network device and computer-readable storage medium, wherein a first device periodically counts packet loss statistics of a port to which each access device is connected, and then determines the destination IP address, destination MAC address and VLAN identifier of multiple packet loss messages of at least one packet loss aggregation group based on the information; for each packet loss aggregation group, when the corresponding second congested VLAN group of the packet loss aggregation group on the second device is empty, the migration VLAN corresponding to the packet loss aggregation group can be determined, and then the MAC address table and ARP table are updated based on the port of the peer link and the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN, so that the traffic that needs to reach the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN can be subsequently migrated to the second device through the peer link for forwarding, so as to alleviate the congestion and packet loss of the first device. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 The figure is a schematic diagram of an MLAG network.

[0054] Figure 2 An example diagram of an application scenario provided for an embodiment of the present invention.

[0055] Figure 3 One of the flow charts of a congestion traffic migration method provided in an embodiment of the present invention.

[0056] Figure 4 The second flowchart of a congestion traffic migration method provided by an embodiment of the present invention.

[0057] Figure 5 A schematic diagram of the structure of a congestion traffic migration device provided in an embodiment of the present invention.

[0058] Figure 6 A schematic diagram of the structure of a network device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0060] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0062] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.

[0063] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0064] Here, the key words or key terms involved in the present invention are first introduced:

[0065] 1. MLAG: Please combine Figure 1 When MLAG networking is used, different ports on different devices SwitchA and SwitchB are combined into an aggregation group to achieve the same function as a normal LAG (Link Aggregation Group). For access devices, the two MLAG devices are considered to be one device. In addition, a peer-link is configured between the two MLAG devices, so that the two MLAG devices can synchronize protocol packets to achieve the purpose of active-active forwarding.

[0066] Among them, Eth-Trunk (Ethernet link aggregation technology) is a technology that bundles multiple physical Ethernet interfaces into one logical interface, mainly used to increase network bandwidth, enhance network reliability, and achieve load balancing. This technology bundles multiple independent physical interfaces together to form a logical interface with higher bandwidth, thereby improving the overall performance and availability of the network.

[0067] 2. MOD (Mirror On Drop) is a technology specifically used to monitor packet loss during the forwarding process inside the device. Once MOD detects packet loss inside the device, it will immediately collect the time, cause and characteristics of the packet loss message, and report it so that the administrator can understand the packet loss inside the device in a timely manner. MOD will also use the IPFIX resources of the switching chip to identify packet loss information.

[0068] 3. IPFIX: IP Flow Information Export, which is the IP data flow information output. It is a function of the switching chip. This function can be used to collect port traffic to generate seven-tuple table entries. The statistical information does not consume CPU resources.

[0069] Next, the application scenarios of the present invention are introduced.

[0070] In the application scenario of the present invention, the MLAG system includes a first device and a second device, and a peer link is established between the two; at least one access device is dual-homed to the MLAG system.

[0071] The first device and the second device may be network devices such as switches or distributed switching devices, and the access device may be a switching device, a router with a switching function, a server, etc. The first device and the second device may be a master device and a backup device, respectively, or the first device and the second device may be a backup device and a master device, respectively, which is not limited here.

[0072] There may be more than one connection link between each access device and the first device. For example, see Figure 2 , Figure 2 Figure 2 shows two access devices (access device 1 and access device 2) in a dual-homed MLAG system. The distribution of their aggregation groups is as follows:

[0073] The first device includes two aggregation groups, Group 1 and Group 2. Group 1 includes three ports, a1, a2, and a3, and Group 2 includes two ports, a4 and a5.

[0074] The second device also includes two aggregation groups, Group 1 and Group 2. Group 1 includes two ports, b1 and b2, and Group 2 includes two ports, b3 and b4.

[0075] Figure 2 The above is only an example and is not intended to be limiting.

[0076] In the MLAG system, the first device and the second device each include at least one aggregation group, and each maintains the MLAG status information corresponding to each of its own aggregation groups. For the first device, the MLAG status information of an aggregation group A includes: the link status, data status, and state machine information of the local aggregation group A, as well as the link status and data status of the aggregation group A on the second device. For example, taking the congestion of vlan55 in the aggregation group 22 in the first device as an example, the MLAG status information of the aggregation group 22 includes the following content (the text after / / is only for explanation):

[0077] --Link status:DOWN / / Link status

[0078] --Data status:DOWN / / Data status

[0079] --Active mlag vlans:1,55,100 / / Active vlan

[0080] --Redirect FSM state:UNREDIRECT / / State machine information

[0081] --Isolate FSM state:UNISOLATE / / State machine information

[0082] --Block FSM state:BLOCK / / State machine information

[0083] --Remote interface:link-aggregation 22 / / The corresponding aggregation group 22 on the second device

[0084] --Remote link status:DOWN / / Link status of aggregation group 22 on the second device

[0085] --Remote data status:DOWN / / Data status of aggregation group 22 on the second device

[0086] In the present invention, a first congested VLAN group and a second congested VLAN group are added to the MLAG state information corresponding to each aggregation group, which are used to store VLAN identifiers of VLAN traffic migrated due to congestion packet loss on the first device and the second device, respectively.

[0087] That is, the first device maintains a first congestion VLAN group corresponding to each aggregation group (Local congestion vlan in the MLAG status information), and stores a second congestion VLAN group corresponding to each aggregation group on the second device (Remote congestion vlan in the MLAG status information); and the second device stores a first congestion VLAN group corresponding to each aggregation group on the first device (Remote congestion vlan in the MLAG status information), and maintains a second congestion VLAN group corresponding to each aggregation group (Local congestion vlan in the MLAG status information).

[0088] Moreover, on the first device and the second device, each first congested VLAN group and each second congested VLAN group are initially empty. When the first device updates any first congested VLAN group (adds or deletes a certain VLAN), it will be synchronized to the second device, so that the second device stores the updated first congested VLAN group; similarly, when the second device updates any second congested VLAN group (adds or deletes a certain VLAN), it will be synchronized to the second device, so that the first device stores the updated second congested VLAN group.

[0089] The following mainly introduces the congested VLAN traffic migration process and the re-migration process implemented by the first device based on the first congested VLAN group and the second congested VLAN group corresponding to the aggregation group.

[0090] Please refer to Figure 3 , Figure 3 A flow chart of a congestion traffic migration method provided by an embodiment of the present invention is provided. The execution subject of the method is the first device in the above MLAG system. The method includes the following steps S101 to S105:

[0091] S101. Regularly obtain packet loss statistics information of a port to which each access device is connected.

[0092] In this embodiment, during the network deployment process, both devices in the MLAG system can enable the MOD function, so that the first device can count the packet loss statistics of each access device access port within the current statistical period every preset statistical period (for example, 5 minutes or 10 minutes). If a port does not generate packet loss, its packet loss statistics may be empty; if packet loss occurs, the packet loss statistics may include but are not limited to: the discarding time, discarding reason, forwarding port, VLAN ID, source IP address, destination IP address, source MAC address, destination MAC address, etc. of each discarded message.

[0093] S102: Determine packet loss data of at least one packet loss aggregation group based on packet loss statistics information of each port and the aggregation group to which it belongs.

[0094] In this embodiment, the packet loss data includes the destination IP addresses, destination MAC addresses and VLAN identifiers of multiple packet loss messages.

[0095] Optionally, the first device may first group the packet loss statistics of all ports belonging to the same aggregation group into a group, and then for each aggregation group: if there are no discarded packets in each packet loss statistics of the group, delete them directly; if there are discarded packets in any packet loss statistics of the group, extract the destination IP address, destination MAC address and VLAN ID of each packet loss packet to obtain the packet loss data of a packet loss aggregation group.

[0096] S103: For each packet loss aggregation group, when the second congested VLAN group corresponding to the packet loss aggregation group is empty, determine a migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group.

[0097] It is understandable that steps S103 to S105 need to be performed for each packet loss aggregation group. For each packet loss aggregation group, the second congested VLAN group corresponding to the packet loss aggregation group stored by the first device may remain empty in the initial state, or, when the first device receives the second congested VLAN group corresponding to the packet loss aggregation group synchronized by the second device, it will delete the previously stored one and store the one sent by the second device.

[0098] In this embodiment, for a packet loss aggregation group (assuming it is aggregation group A), if the second congested VLAN group corresponding to the aggregation group A is empty, it means that there is surplus bandwidth of the aggregation group A on the second device and there is no congestion. Then the first device can determine the migration VLAN corresponding to the aggregation group A based on the VLAN identifier of each packet loss message of the aggregation group A.

[0099] S104: Update the MAC address table and the ARP table based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet belonging to the migration VLAN.

[0100] In this embodiment, the MAC address table and ARP table are updated based on the port of the peer link and the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN. Then, the first device forwards traffic based on the updated MAC address table and ARP table, so that part of the network traffic belonging to the migration VLAN can be migrated to the second device through the peer link for forwarding, thereby alleviating congestion and packet loss of the first device.

[0101] S105: Add a migration VLAN to the first congested VLAN group corresponding to the packet loss aggregation group, and send the first congested VLAN group corresponding to the packet loss aggregation group to the second device.

[0102] In this embodiment, the first device adds the migrated VLAN to the first congested VLAN group corresponding to the packet loss aggregation group and synchronizes it to the second device. The second device stores the latest first congested VLAN group of the packet loss aggregation group received, and then deletes the first congested VLAN group of the packet loss aggregation group stored previously. This ensures that the second device will not migrate the VLAN traffic of the packet loss aggregation group to the first device due to congestion, thereby aggravating the congestion of the first device.

[0103] The congestion traffic migration method adopted in the embodiment of the present invention is to periodically count the packet loss statistics of the port to which each access device is connected by the first device, and then determine at least one packet loss aggregation group based on this, and the migration VLAN corresponding to the packet loss aggregation group when the second congestion VLAN group corresponding to the packet loss aggregation group on the second device is empty, and then update the MAC address table and ARP table based on the port of the peer link and the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN, so that the traffic that needs to reach the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN can be subsequently migrated to the second device through the peer link for forwarding, so as to alleviate the congestion and packet loss of the first device.

[0104] In an optional implementation, in the above step S103, the migration VLAN may be determined based on the number of packet losses corresponding to each VLAN ID in the packet loss aggregation group and the first congested VLAN group corresponding to the packet loss aggregation group. Figure 3 Based on Figure 4 The process of determining the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group in the above step S103 may include the following sub-steps S1031 to S1036.

[0105] S1031. Determine the number of lost packets corresponding to each packet loss VLAN identifier in the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group.

[0106] S1032: based on the number of packet losses corresponding to each packet loss VLAN identifier, arrange all packet loss VLAN identifiers in the packet loss aggregation group in descending order to obtain a sorted list.

[0107] S1033: Use the first packet loss VLAN identifier in the sorted list as the pending VLAN.

[0108] S1034: Determine whether there is a pending VLAN in the first congested VLAN group corresponding to all aggregation groups on the first device.

[0109] S1035: Use the next packet loss VLAN identifier of the pending VLAN in the sorted list as a new pending VLAN.

[0110] S1036: Use the undetermined VLAN as the migration VLAN corresponding to the packet loss aggregation group.

[0111] In this embodiment, if the pending VLAN exists in the first congested VLAN group corresponding to all the aggregation groups on the first device (assuming that the pending VLAN is already in the first congested VLAN group corresponding to the aggregation group B), it means that the traffic belonging to the pending VLAN that should be loaded through the aggregation group B has been migrated due to congestion and packet loss, and is currently in the migration state, so this time, other VLANs after sorting are selected for migration, and step S1035 is executed and then return to step S1034. If the pending VLAN does not exist in the first congested VLAN group corresponding to any aggregation group of the first device, it means that the traffic of the pending VLAN passing through the packet loss aggregation group has not been congested before, or has been congested and congested traffic has been migrated and then migrated back, and step S1036 is continued to be executed this time.

[0112] In an optional implementation, the process of updating the MAC address table and the ARP table based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet belonging to the migration VLAN in the above step S104 may include the following sub-steps S1041 to S1042.

[0113] S1041. Obtain a destination IP address and a destination MAC address of each lost packet belonging to a migration VLAN.

[0114] S1042: For each packet loss message, search the MAC address table based on the destination MAC address and VLAN ID of the packet loss message, and search the ARP table based on the destination IP address and destination MAC address of the packet loss message.

[0115] S1043. If a matching dynamic MAC table entry can be found in the MAC address table and a matching dynamic ARP table entry can be found in the ARP table, after deleting the found dynamic MAC table entry and dynamic ARP table entry, a static MAC table entry is created in the MAC address table based on the port of the peer link and the destination MAC address and VLAN ID of the lost packet, and a static ARP table entry is created in the ARP table based on the destination IP address and destination MAC address of the lost packet.

[0116] In this embodiment, the created static MAC table entry includes the port of the peer link and the destination MAC address and VLAN ID of the lost packet message, and the created static ARP table entry includes the destination IP address and destination MAC address of the lost packet message.

[0117] Optionally, if the peer link is a physical link, the port in the created static MAC table entry is the physical port; if the peer link is a logical link formed by link aggregation of multiple physical links, then the port in the created static MAC table entry can be a logical port or an aggregation group identifier (MLAG Group ID) corresponding to the peer link.

[0118] Among them, for any packet loss message, if no matching dynamic MAC table entry is found in the MAC address table and no matching dynamic ARP table entry is found in the ARP table, it means that the packet carrying the packet loss belongs to one of broadcast traffic, multicast traffic and unknown unicast traffic. These three types of traffic do not need to be migrated because they have been flooded through the peer link.

[0119] In the ARP table and MAC address table, dynamic entries (i.e., the type field is dynamic) can be automatically aged (deleted or updated), but static entries (i.e., the type field is static) will not be aged under normal circumstances, nor will they be overwritten by dynamic ARP entries. In the present invention, it is necessary to allow the static MAC entry and the static ARP entry in step S1042 to be automatically aged, so that when the host corresponding to the IP address and MAC address in the static ARP entry leaves the access device or goes offline, the static MAC entry and the static ARP entry can be automatically deleted, and the static MAC entry and the static ARP entry of the offline host will not be gradually accumulated.

[0120] Therefore, in order to realize automatic aging of the static MAC entry created in step S1043, after the above step S1043, the method may further include the following steps S1044 to S1046.

[0121] S1044. Using the source MAC address field and the VLAN ID field as matching fields, IPFIX traffic collection begins at each port of the packet loss aggregation group, and a MAC address flow table is output.

[0122] In this embodiment, the source MAC address field and the VLAN ID field are used as matching fields, and IPFIX traffic collection is continuously performed on the inbound traffic of each port in the packet loss aggregation group, and a MAC address flow table is output. The MAC address flow table belongs to the IPFIX flow table. Then, a flow table item of the MAC address flow table includes but is not limited to the source MAC address, VLAN ID and recording time. The source MAC address and VLAN ID between any two flow table items are different.

[0123] S1045: Perform aging processing on each flow table entry of the MAC address flow table based on a preset flow table entry aging time.

[0124] In the optional example, assuming that the aging time of the flow table entry is t1, the first flow table entry in the MAC address flow table includes MAC1, VLAN100 and the recording time, and in the process of continuous IPFIX traffic collection, the process of aging the first flow table entry is: as long as MAC1 and VLAN100 are collected from the message, the recording time of the first flow table entry will be updated once, and if MAC1 and VLAN100 are not collected again within the updated t1 period, it means that the host corresponding to MAC1 is offline, and the first flow table entry will be directly deleted. This example is only an example and is not limited here.

[0125] S1046. For each static MAC table entry to be aged in the MAC address table including the port of the peer link and the migration VLAN, check whether the MAC address in the static MAC table entry to be aged exists in the MAC address flow table every preset MAC address aging time; if not, delete the static MAC table entry to be aged from the MAC address table.

[0126] In this embodiment, in the MAC address table, each static MAC table entry including the port of the peer link and the migration VLAN is a static MAC table entry that needs to be automatically aged (referred to as a static MAC table entry to be aged). For any static MAC table entry to be aged, if the MAC address in the static MAC table entry to be aged does not exist in the MAC address flow table, it means that the host corresponding to the MAC address is offline, and the static MAC table entry to be aged needs to be deleted from the MAC address table.

[0127] Optionally, the flow entry aging time is less than or equal to the MAC address aging time. For example, the flow entry aging time is 20s, while the MAC address aging time is 25s. This example is only an example and is not limited here. When the CPU is busy, setting the flow entry aging time to be less than the MAC address aging time can reserve a certain buffer time for executing step S1046.

[0128] The present invention outputs a MAC address table by utilizing the IPFIX function, thereby utilizing an automatic aging mechanism from the MAC address table to flow table items, thereby realizing automatic aging processing of each static MAC table item in the MAC address table.

[0129] In order to realize automatic aging of the static ARP entry created in step S1043, after step S1043, the method may further include the following steps S1047 to S1048.

[0130] S1047. For each IP address of a static ARP entry in the ARP table, send an ARP message to the IP address at a preset period.

[0131] S1048: If no ARP response message replying the IP address is received within the preset waiting time, the static ARP table entry including the IP address is deleted from the ARP table.

[0132] Optionally, the preset period may be 1 minute, and the preset waiting time may be 3 seconds. This example is only an example and is not limited here.

[0133] The present invention performs ARP detection on the IP address of each static ARP table item in the ARP table, so that each static ARP table item in the ARP table can be automatically aged like a dynamic ARP table item.

[0134] An optional implementation method is that, for each packet loss aggregation group, after part of the network traffic of the migrated VLAN in the packet loss aggregation group is migrated to the second device through a peer link for forwarding, if the remaining bandwidth of the packet loss aggregation group is sufficient, this part of the network traffic needs to be migrated back in time to avoid congestion and packet loss caused by the second device being subjected to additional load of this part of the network traffic for a long time.

[0135] Therefore, in order to perform traffic relocation in a timely manner, after the above step S105, the method may further include steps S106 to S110.

[0136] S106: Based on the migration VLAN corresponding to the packet loss aggregation group, configure a traffic matching rule in the outbound direction of the port of the peer link.

[0137] In this embodiment, the traffic matching rule is an ACL (Access Control List) rule, and the configured ACL rule is used to count the traffic sending rate of the network traffic belonging to the migration VLAN and sent out in the outbound direction of the port of the peer link.

[0138] S107. Obtain the remaining bandwidth of the packet loss aggregation group at every preset statistical interval, and obtain the sending rate of traffic belonging to the migration VLAN and sent from the port of the peer link based on traffic matching rule statistics.

[0139] Optionally, the preset statistical interval may be 5s, 10s, etc., which is not limited here.

[0140] S108: If the remaining bandwidth is greater than the traffic sending rate and the difference between the two is not less than the set threshold, at least one matching static MAC table entry is searched from the MAC address table based on the migration VLAN and then deleted.

[0141] In this embodiment, the remaining bandwidth is greater than the traffic sending rate and the difference between the two is not lower than the set threshold, which means that the first device can carry the migrated network traffic belonging to the migration VLAN without congestion. Optionally, the set threshold can be 10Gbps, 5Gbps, 3Gbps, etc., which is not limited here.

[0142] S109: Find out a matching static ARP entry from the ARP table based on the MAC address in each found static MAC entry and delete it.

[0143] After the MAC address table and the ARP table are updated respectively based on the above steps S108 and S109, the traffic migration is completed.

[0144] S110: After deleting the migration VLAN from the first congested VLAN group corresponding to the packet loss aggregation group, the first congested VLAN group corresponding to the packet loss aggregation group is sent to the second device.

[0145] In this embodiment, after the migration VLAN is deleted from the first congested VLAN group corresponding to the packet loss aggregation group, the first congested VLAN group corresponding to the packet loss aggregation group is sent to the second device, so that the second device only retains and stores the latest first congested VLAN group received. If congestion and packet loss occur in the aggregation group on the second device later, congested traffic can be considered to be migrated to the first device.

[0146] It should be noted that the above steps S103 to S110 and their sub-steps are only for explaining the traffic migration and re-migration of a packet loss aggregation group. And the execution order of each step in the above method embodiment is not limited to the steps shown in the accompanying drawings and the step numbers, and the execution order of each step shall be based on the actual application situation. In addition, the above content introduces the process of congested traffic migration starting from the first device, and the second device also needs to perform congested traffic migration. The implementation principle is similar to the above content and will not be repeated here.

[0147] In order to execute the corresponding steps in the above method embodiment and each possible implementation mode, an implementation mode of a congestion traffic migration device is provided below.

[0148] See also Figure 5 , Figure 5The structural schematic diagram of the congested traffic migration device provided by an embodiment of the present invention is shown. The congested traffic migration device 200 is applied to the first device of the MLAG system, and the MLAG system also includes a second device that establishes a peer link with the first device; at least one access device is dual-homed to the MLAG system; the first device maintains a first congested VLAN group corresponding to each aggregation group, and stores a second congested VLAN group corresponding to each aggregation group on the second device. The congested traffic migration device 200 includes: a packet loss statistics module 210, a data sorting module 220, a migration determination module 230, and a traffic migration module 240.

[0149] The packet loss statistics module 210 is used to periodically obtain the packet loss statistics information of the port to which each access device is connected;

[0150] The data sorting module 220 is used to determine the packet loss data of at least one packet loss aggregation group based on the packet loss statistics of each port and the aggregation group to which it belongs, where the packet loss data includes the destination IP address, the destination MAC address and the VLAN ID of the plurality of packet loss messages;

[0151] The migration determination module 230 is used to determine the migration VLAN corresponding to each packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group when the second congested VLAN group corresponding to the packet loss aggregation group is empty;

[0152] Traffic migration module 240, used to update the MAC address table and ARP table based on the port of the peer link and the destination IP address and destination MAC address of each lost packet belonging to the migration VLAN;

[0153] The traffic migration module 240 is further configured to add a migration VLAN to the first congested VLAN group corresponding to the packet loss aggregation group, and send the first congested VLAN group corresponding to the packet loss aggregation group to the second device.

[0154] Optionally, the migration determination module 230 is used to determine the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group. Specifically, it can be used to: determine the number of packet losses for each packet loss VLAN identifier in the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group; based on the number of packet losses for each packet loss VLAN identifier, arrange all packet loss VLAN identifiers in the packet loss aggregation group in descending order to obtain a sorted list; use the first packet loss VLAN identifier in the sorted list as the pending VLAN; determine whether there is a pending VLAN in the first congested VLAN group corresponding to all aggregation groups on the first device; if not, use the next packet loss VLAN identifier of the pending VLAN in the sorted list as the new pending VLAN, and return to execute the step of determining whether there is a pending VLAN in the first congested VLAN group corresponding to all aggregation groups on the first device; if so, use the pending VLAN as the migration VLAN corresponding to the packet loss aggregation group.

[0155] Optionally, the traffic migration module 240 is used to update the MAC address table and the ARP table based on the port of the peer link and the destination IP address and destination MAC address of each lost packet belonging to the migration VLAN. Specifically, it can be used to: obtain the destination IP address and destination MAC address of each lost packet belonging to the migration VLAN; for each lost packet, search the MAC address table based on the destination MAC address and VLAN identifier of the lost packet, and search the ARP table based on the destination IP address and destination MAC address of the lost packet; if a matching dynamic MAC table entry can be found in the MAC address table and a matching dynamic ARP table entry can be found in the ARP table, then after deleting the found dynamic MAC table entry and dynamic ARP table entry, a static MAC table entry is created in the MAC address table based on the port of the peer link and the destination MAC address and VLAN identifier of the lost packet, and a static ARP table entry is created in the ARP table based on the destination IP address and destination MAC address of the lost packet.

[0156] Optionally, the congested traffic migration device 200 may also include an entry aging module 250, which is used to: start IPFIX traffic collection at each port of the packet loss aggregation group with the source MAC address field and the VLAN identification field as matching fields, and output a MAC address flow table; perform aging processing on each flow table entry of the MAC address flow table based on a preset flow table entry aging time; for each static MAC table entry to be aged in the MAC address table including the port of the peer link and the migration VLAN, search the MAC address flow table for the MAC address address whether there is a MAC address in the static MAC table entry to be aged every preset MAC address aging time; wherein the flow table entry aging time is less than or equal to the MAC address aging time; if not, delete the static MAC table entry to be aged from the MAC address table. Optionally, the entry aging module 250 may also be used to: for the IP address of each static ARP table entry in the ARP table, send an ARP message to the IP address every preset period; if no ARP response message of the IP address reply is received within the preset waiting time, delete the static ARP table entry including the IP address from the ARP table.

[0157] Optionally, the congested traffic migration device 200 may also include a traffic back-migration module 260, which is used to: configure traffic matching rules in the outbound direction of the port of the peer link based on the migration VLAN corresponding to the packet loss aggregation group; obtain the remaining bandwidth of the packet loss aggregation group at every preset statistical interval, and obtain the traffic sending rate belonging to the migration VLAN sent from the port of the peer link based on the traffic matching rule statistics; if the remaining bandwidth is greater than the traffic sending rate and the difference between the two is not less than a set threshold, then find out at least one matching static MAC table entry from the MAC address table based on the migration VLAN and delete it; find out a matching static ARP table entry from the ARP table based on the MAC address in each static MAC table entry found and delete it; after deleting the migration VLAN from the first congested VLAN group corresponding to the packet loss aggregation group, send the first congested VLAN group corresponding to the packet loss aggregation group to the second device.

[0158] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the congestion traffic migration device 200 described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0159] See also Figure 6 , Figure 6 The schematic diagram of the structure of a network device provided by an embodiment of the present invention is as follows: The network device 300 includes a processor 310 , a memory 320 and a bus 330 , and the processor 310 is connected to the memory 320 via the bus 330 .

[0160] The memory 320 may be used to store software programs, for example, software programs corresponding to the congested traffic migration device 200 provided in the embodiment of the present invention. The processor 310 executes various functional applications and data processing to implement the congested traffic migration method provided in the embodiment of the present invention by running the software programs stored in the memory 320.

[0161] Among them, the memory 320 can be but is not limited to: RAM (Random Access Memory), ROM (Read Only Memory), FLASH (Flash Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electric Erasable Programmable Read-Only Memory), etc.

[0162] The processor 310 may be an integrated circuit chip with signal processing capability. The processor 310 may be a general-purpose processor, including: CPU (Central Processing Unit), NP (Network Processor), etc.; it may also be: DSP (Digital Signal Processing), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0163] Understandably, Figure 6 The structure shown is for illustration only. The network device 300 may also include Figure 6 More or fewer components as shown, or with Figure 6 Different configurations are shown. Figure 6 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0164] The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the congestion traffic migration method disclosed in the above embodiment is implemented. The computer-readable storage medium can be, but is not limited to, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a PROM, an EPROM, an EEPROM, a FLASH disk, or an optical disk.

[0165] In summary, the embodiments of the present invention provide a congestion traffic migration method, apparatus, network device and computer-readable storage medium, wherein the first device periodically counts the packet loss statistics of the port to which each access device is connected, and then determines the destination IP address, destination MAC address and VLAN identifier of multiple packet loss messages of at least one packet loss aggregation group based on the information; for each packet loss aggregation group, when the corresponding second congested VLAN group of the packet loss aggregation group on the second device is empty, the migration VLAN corresponding to the packet loss aggregation group can be determined, and then the MAC address table and ARP table are updated based on the port of the peer link and the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN, so that the traffic that needs to reach the destination IP address and destination MAC address of each packet loss message belonging to the migration VLAN can be subsequently migrated to the second device through the peer link for forwarding, so as to alleviate the congestion and packet loss of the first device.

[0166] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A congestion traffic migration method, characterized in that: A first device applied to an MLAG system, the MLAG system further comprising a second device establishing a peer link with the first device; at least one access device dual-homing accesses the MLAG system; the first device maintains a first congested VLAN group corresponding to each aggregation group, and stores a second congested VLAN group corresponding to each aggregation group on the second device; the method comprising: Regularly obtain packet loss statistics of ports connected to each access device; Based on the packet loss statistics of each of the ports and the aggregation group to which it belongs, determine the packet loss data of at least one packet loss aggregation group, wherein the packet loss data includes the destination IP address, the destination MAC address and the VLAN ID of the plurality of packet loss messages; For each of the packet loss aggregation groups, when the second congested VLAN group corresponding to the packet loss aggregation group is empty, determining the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group; Based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet belonging to the migration VLAN, the MAC address table and the ARP table are updated; The migration VLAN is added to the first congested VLAN group corresponding to the packet loss aggregation group, and the first congested VLAN group corresponding to the packet loss aggregation group is sent to the second device.

2. The method according to claim 1, characterized in that The step of determining the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group includes: Based on the VLAN identifier of each packet loss message in the packet loss aggregation group, determining the number of packet losses corresponding to each packet loss VLAN identifier in the packet loss aggregation group; Based on the number of packet losses corresponding to each packet loss VLAN identifier, all packet loss VLAN identifiers in the packet loss aggregation group are arranged in descending order to obtain a sorted list; The first packet loss VLAN identifier in the sorted list is used as the pending VLAN; Determine whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device; If yes, then use the next packet loss VLAN identifier of the pending VLAN in the sorted list as the new pending VLAN, and return to the step of determining whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device; If not, the pending VLAN is used as the migration VLAN corresponding to the packet loss aggregation group.

3. The method according to claim 1, characterized in that The step of updating the MAC address table and the ARP table based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet message belonging to the migration VLAN comprises: Obtaining the destination IP address and destination MAC address of each lost packet belonging to the migration VLAN; For each of the lost packet messages, searching the MAC address table based on the destination MAC address and VLAN identifier of the lost packet message, and searching the ARP table based on the destination IP address and destination MAC address of the lost packet message; If a matching dynamic MAC table entry can be found in the MAC address table and a matching dynamic ARP table entry can be found in the ARP table, after deleting the found dynamic MAC table entry and the dynamic ARP table entry, a static MAC table entry is created in the MAC address table based on the port of the peer link and the destination MAC address and VLAN ID of the lost packet message, and a static ARP table entry is created in the ARP table based on the destination IP address and destination MAC address of the lost packet message.

4. The method according to claim 3, characterized in that After the steps of creating a static MAC table entry in the MAC address table based on the port of the peer link and the destination MAC address and VLAN identifier of the lost packet message, and creating a static ARP table entry in the ARP table based on the destination IP address and the destination MAC address of the lost packet message, the method further includes: Using the source MAC address field and the VLAN ID field as matching fields, IPFIX traffic collection starts at each port of the packet loss aggregation group, and outputs a MAC address flow table; Performing aging processing on each flow table entry of the MAC address flow table based on a preset flow table entry aging time; For each static MAC table entry to be aged including the port of the peer link and the migration VLAN in the MAC address table, at every preset MAC address aging time, searching the MAC address flow table for a MAC address in the static MAC table entry to be aged; wherein the flow entry aging time is less than or equal to the MAC address aging time; If not, the static MAC table entry to be aged is deleted from the MAC address table.

5. The method according to claim 3, characterized in that: After the steps of creating a static MAC table entry in the MAC address table based on the port of the peer link and the destination MAC address and VLAN identifier of the lost packet message, and creating a static ARP table entry in the ARP table based on the destination IP address and the destination MAC address of the lost packet message, the method further includes: For each IP address of the static ARP table entry in the ARP table, sending an ARP message to the IP address at a preset period; If no ARP response message replying to the IP address is received within the preset waiting time, the static ARP table entry including the IP address is deleted from the ARP table.

6. The method according to claim 1, characterized in that After the step of adding the migration VLAN to the first congested VLAN group corresponding to the packet loss aggregation group, and sending the first congested VLAN group corresponding to the packet loss aggregation group to the second device, the method further includes: Based on the migration VLAN corresponding to the packet loss aggregation group, configure a traffic matching rule in the outbound direction of the port of the peer link; At every preset statistical interval, the remaining bandwidth of the packet loss aggregation group is obtained, and the traffic sending rate belonging to the migration VLAN and sent from the port of the peer link based on the traffic matching rule is obtained; If the remaining bandwidth is greater than the traffic sending rate and the difference between the two is not less than a set threshold, searching for at least one matching static MAC table entry from the MAC address table based on the migration VLAN and deleting the entry; Based on the MAC address in each static MAC table entry found, the matching static ARP table entry is found from the ARP table and deleted; After deleting the migration VLAN from the first congested VLAN group corresponding to the packet loss aggregation group, the first congested VLAN group corresponding to the packet loss aggregation group is sent to the second device.

7. A congestion traffic migration device, characterized in that: A first device applied to an MLAG system, the MLAG system further comprising a second device establishing a peer link with the first device; at least one access device dual-homing accesses the MLAG system; the first device maintains a first congested VLAN group corresponding to each aggregation group, and stores a second congested VLAN group corresponding to each aggregation group on the second device; the device comprises: The packet loss statistics module is used to periodically obtain the packet loss statistics information of the port to which each access device is connected; A data sorting module, configured to determine packet loss data of at least one packet loss aggregation group based on packet loss statistics of each of the ports and the aggregation group to which it belongs, wherein the packet loss data includes a destination IP address, a destination MAC address and a VLAN identifier of a plurality of packet loss messages; A migration determination module, for each of the packet loss aggregation groups, when the second congested VLAN group corresponding to the packet loss aggregation group is empty, based on the VLAN identifier of each packet loss message in the packet loss aggregation group, determines the migration VLAN corresponding to the packet loss aggregation group; A traffic migration module, configured to update a MAC address table and an ARP table based on the port of the peer link and the destination IP address and the destination MAC address of each lost packet message belonging to the migration VLAN; The traffic migration module is further configured to add the migration VLAN to the first congested VLAN group corresponding to the packet loss aggregation group, and send the first congested VLAN group corresponding to the packet loss aggregation group to the second device.

8. The device according to claim 7, characterized in that The migration determination module is used to determine the migration VLAN corresponding to the packet loss aggregation group based on the VLAN identifier of each packet loss message in the packet loss aggregation group, and is specifically used to: Based on the VLAN ID of each packet loss message in the packet loss aggregation group, determining the number of packet loss for each packet loss VLAN ID in the packet loss aggregation group; Based on the number of packet losses of each packet loss VLAN ID, all packet loss VLAN IDs in the packet loss aggregation group are arranged in descending order to obtain a sorted list; The first packet loss VLAN identifier in the sorted list is used as the pending VLAN; Determine whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device; If yes, then use the next packet loss VLAN identifier of the pending VLAN in the sorted list as the new pending VLAN, and return to the step of determining whether the pending VLAN exists in the first congested VLAN group corresponding to all aggregation groups on the first device; If not, the pending VLAN is used as the migration VLAN corresponding to the packet loss aggregation group.

9. A network device, characterized in that: include: A memory and a processor, wherein the memory stores a software program, and when the network device is running, the processor executes the software program to implement the congested traffic migration method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the congested traffic migration method according to any one of claims 1 to 6 is implemented.

Images