Physical path fault detection method and device, equipment, medium and program product

Abstract

The invention relates to a physical path fault detection method and device, equipment, a medium and a program product. The physical path fault detection method comprises the following steps: aiming at each target port, taking the target port as a source port to carry out traversal combination with other target devices except the device, and generating a plurality of in-band network telemetering messages of which the protocol type is a remote direct memory access protocol based on the converged Ethernet; forwarding the in-band network telemetering message to a corresponding receiving end based on a target switch corresponding to the equipment and quintuple information of the in-band network telemetering message, so that the receiving end generates a forwarding path based on information of at least one corresponding switch in the received in-band network telemetering message to obtain a plurality of forwarding paths; and performing fault detection on the physical path of the data center network based on the plurality of forwarding paths to obtain a fault detection result, thereby realizing forwarding of the message as a real service flow and fault detection of switch configuration.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data center network fault detection, and in particular, to a physical path fault detection method, device, equipment, medium and program product. Background Art

[0002] The data center network is an important part of modern information technology, which is responsible for connecting and managing various devices and resources within the data center. The accuracy and rationality of the configuration and wiring of the physical devices in the data center network are the basis for ensuring the normal forwarding of service data and the uniform forwarding of traffic. Although the cluster construction personnel can confirm the connection status through the connection information between devices after the network deployment is completed, this method cannot detect problems such as incorrect device configuration or wiring that does not meet the expected design. Such errors may lead to substandard performance of the computing cluster. Therefore, there is an urgent need for a detection means to accept and detect all physical paths of the data center network.

[0003] The existing detection means mainly rely on the Link Layer Discovery Protocol (LLDP). This protocol helps network devices discover directly connected physical devices as expected, and obtains the topology information of the entire network through hierarchical queries, so as to detect whether the pairwise wiring between devices meets the expectations.

[0004] However, as a link layer protocol, the LLDP protocol is quite different from real service traffic packets and cannot perform real tests on actual services. Moreover, the LLDP protocol can only discover the connectivity between two devices, but cannot detect configuration errors or faults in switches. Summary of the Invention

[0005] To solve the above technical problems, the present disclosure provides a physical path fault detection method, device, equipment, medium and program product.

[0006] The first aspect of the embodiments of the present disclosure provides a physical path fault detection method, including:

[0007] For each target port corresponding to each device in the data center network, taking the target port as the source port and traversing and combining it with other target devices except the device to generate a plurality of in-band network telemetry packets, and the protocol type of the in-band network telemetry packets is the Remote Direct Memory Access protocol based on Converged Ethernet;

[0008] For each in-band network telemetry message, forward the in-band network telemetry message to the target switch corresponding to the device, so that the target switch forwards the in-band network telemetry message to the corresponding receiving end based on the five-tuple information of the in-band network telemetry message. After receiving the in-band network telemetry message, the receiving end generates a forwarding path based on at least one switch information corresponding in the received in-band network telemetry message, obtaining multiple forwarding paths. The at least one switch information is the configuration information and forwarding status information respectively corresponding to each switch passed through during the forwarding process of the in-band network telemetry message;

[0009] Perform a fault detection on the physical path of the data center network based on the multiple forwarding paths to obtain a fault detection result.

[0010] The second aspect of the embodiments of the present disclosure provides a physical path fault detection device, including:

[0011] A message generation module, configured to, for each target port corresponding to each device in the data center network, use the target port as the source port to perform a traversal combination with other target devices except the device to generate multiple in-band network telemetry messages. The protocol type of the in-band network telemetry message is a remote direct memory access protocol based on converged Ethernet;

[0012] A message forwarding module, configured to, for each in-band network telemetry message, forward the in-band network telemetry message to the target switch corresponding to the device, so that the target switch forwards the in-band network telemetry message to the corresponding receiving end based on the five-tuple information of the in-band network telemetry message. After receiving the in-band network telemetry message, the receiving end generates a forwarding path based on at least one switch information corresponding in the received in-band network telemetry message, obtaining multiple forwarding paths. The at least one switch information is the configuration information and forwarding status information respectively corresponding to each switch passed through during the forwarding process of the in-band network telemetry message;

[0013] A fault detection module, configured to perform a fault detection on the physical path of the data center network based on the multiple forwarding paths to obtain a fault detection result.

[0014] The third aspect of the embodiments of the present disclosure provides an electronic device, including:

[0015] A processor;

[0016] A memory, configured to store executable instructions;

[0017] Wherein, the processor is configured to read the executable instructions from the memory and execute the executable instructions to implement the physical path fault detection method provided in the first aspect above.

[0018] The fourth aspect of the embodiments of the present disclosure provides a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the physical path fault detection method provided in the first aspect above.

[0019] The fifth aspect of the embodiments of the present disclosure provides a computer program product including a computer program or instruction, which, when executed by a processor, implements the physical path fault detection method as described in the first aspect above.

[0020] The technical solutions provided by the embodiments of the present disclosure have the following advantages compared with the prior art:

[0021] The physical path fault detection method, device, equipment, medium and program product provided by the embodiments of the present disclosure can, for each target port corresponding to each device in the data center network, use the target port as the source port to traverse and combine with other target devices except the device to generate multiple in-band network telemetry messages, and the protocol type of the in-band network telemetry message is the remote direct memory access protocol based on converged Ethernet; after generating multiple in-band network telemetry messages, for each in-band network telemetry message, forward the in-band network telemetry message to the target switch corresponding to the device, so that the target switch forwards the in-band network telemetry message to the corresponding receiving end based on the five-tuple information of the in-band network telemetry message. After receiving the in-band network telemetry message, the receiving end generates a forwarding path based on at least one switch information corresponding in the received in-band network telemetry message to obtain multiple forwarding paths, and at least one switch information is the configuration information and forwarding status information corresponding to each switch passed through during the forwarding process of the in-band network telemetry message; perform fault detection on the physical path of the data center network based on the multiple forwarding paths to obtain a fault detection result. Thus, it is possible to generate corresponding in-band network telemetry messages for each target port to ensure that each physical path can be covered during the forwarding of the in-band network telemetry message, and at the same time, by generating in-band network telemetry messages with the protocol type of the remote direct memory access protocol based on converged Ethernet for message forwarding, it is realized that the message is forwarded as real traffic, and switch information is obtained during the message forwarding process, and then the fault detection of the switch configuration is realized according to the obtained switch information and forwarding path. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure and, together with the specification, are used to explain the principles of the present disclosure.

[0023] To more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0024] Figure 1 is a flowchart of a physical path fault detection method provided by an embodiment of the present disclosure;

[0025] Figure 2 is a schematic structural diagram of a message format provided by an embodiment of the present disclosure;

[0026] Figure 3 is a schematic diagram of a message forwarding process provided by an embodiment of the present disclosure;

[0027] Figure 4 is a schematic structural diagram of a physical path fault detection device provided by an embodiment of the present disclosure;

[0028] Figure 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure. Detailed Embodiments

[0029] In order to better understand the above objects, features, and advantages of the present disclosure, the following will further describe the solutions of the present disclosure. It should be noted that, without conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other.

[0030] Many specific details are set forth in the following description to facilitate a thorough understanding of the present disclosure, but the present disclosure may be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some embodiments of the present disclosure, rather than all embodiments.

[0031] It should be understood that the steps recorded in the method embodiments of the present disclosure can be executed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0032] It should be noted that in this text, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the said element.

[0033] It should be noted that the modifiers "a" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0034] Generally, the existing physical path detection means for data center networks mainly rely on the Link Layer Discovery Protocol (LLDP). This protocol helps network devices discover directly connected physical devices as expected and obtain the topology information of the entire network through step-by-step queries, so as to detect whether the wiring between two devices meets the expectations. However, as a link layer protocol, the LLDP protocol is quite different from real service traffic packets and cannot perform real tests on actual services. Moreover, the LLDP protocol can only discover the connectivity between two devices and cannot detect configuration errors or faults in switches. To address this problem, the embodiments of this disclosure provide a physical path fault detection method, which will be introduced below in combination with specific embodiments.

[0035] Figure 1 It is a flowchart of a physical path fault detection method provided by the embodiments of this disclosure. This method can be executed by a physical path fault detection device, which can be implemented in software and / or hardware, and can be configured in an electronic device, such as a server or a terminal. Among them, the terminal specifically includes a mobile phone, a computer or a tablet computer, etc.

[0036] As Figure 1 shown, the physical path fault detection method provided in this embodiment includes the following steps.

[0037] S110. For each target port corresponding to each device in the data center network, use the target port as the source port to traverse and combine with other target devices except the device to generate multiple in-band network telemetry messages. The protocol type of the in-band network telemetry message is the Remote Direct Memory Access (RDMA) protocol based on Converged Ethernet.

[0038] In the embodiments of the present disclosure, the data center network usually includes a large number of physical devices, which is an important part of modern information technology and is responsible for connecting and managing various devices and resources in the data center.

[0039] Each device in the data center network includes one or more ports.

[0040] The in-band network telemetry message is the In-band Network Telemetry message, abbreviated as the INT message.

[0041] The Remote Direct Memory Access (RDMA) protocol based on Converged Ethernet is the RDMA over Converged Ethernet (RoCE) protocol, abbreviated as the RoCE protocol.

[0042] In the embodiments of the present disclosure, each device in the data center network can be a network card or the like, which is not limited herein. When the device is a network card, the target port can be any port corresponding to the network card.

[0043] Specifically, the electronic device can, in response to a fault detection instruction, obtain information about each device in the data center network. For each target port corresponding to each device, generate an in-band network telemetry message, that is, an INT message, based on a preset message format and message generation rule. The protocol type of the INT message is the Remote Direct Memory Access (RDMA) protocol based on Converged Ethernet.

[0044] In the embodiments of the present disclosure, by traversing and combining each target port corresponding to each device with other target devices except the device to generate multiple in-band network telemetry messages, it is possible to ensure that each physical path is covered by traffic during the forwarding process of the multiple in-band network telemetry messages.

[0045] S120. For each in-band network telemetry message, forward the in-band network telemetry message to the target switch corresponding to the device, so that the target switch forwards the in-band network telemetry message to the corresponding receiving end based on the five-tuple information of the in-band network telemetry message. After receiving the in-band network telemetry message, the receiving end generates a forwarding path based on at least one switch information corresponding to the received in-band network telemetry message to obtain multiple forwarding paths. The at least one switch information is the configuration information and forwarding status information respectively corresponding to each switch passed through during the forwarding process of the in-band network telemetry message.

[0046] In the embodiments of the present disclosure, at least one switch information is the information appended to the in-band network telemetry message after at least one switch forwards the in-band network telemetry message. Specifically, it may include the ID information, the incoming port number, and the outgoing port number corresponding to each switch passed through during the forwarding process of the in-band network telemetry message, as well as information such as the timestamp corresponding to the forwarded message. Among them, the configuration information may include the ID information corresponding to the switch, and the forwarding status information may include information such as the incoming port number, the outgoing port number, and the timestamp corresponding to the in-band network telemetry message being forwarded.

[0047] Specifically, after the electronic device generates multiple in-band network telemetry messages, for each in-band network telemetry message, it determines the device corresponding to the in-band network telemetry message and the target switch corresponding to the device, and forwards the in-band network telemetry message to the target switch. Then, the target switch forwards the in-band network telemetry message based on the five-tuple information of the in-band network telemetry message and the preset forwarding rules until it is forwarded to the receiving end. Among them, the specific implementation manner of the switch forwarding the in-band network telemetry message to the corresponding receiving end based on the five-tuple information of the in-band network telemetry message is similar to the existing implementation manner of forwarding the in-band network telemetry message, and will not be elaborated here.

[0048] Further, the in-band network telemetry message received by the receiving end is an in-band network telemetry message appended with at least one switch information. After the receiving end receives the in-band network telemetry message, the server corresponding to the receiving end can, based on the at least one switch information appended to the in-band network telemetry message, determine the forwarding path corresponding to the in-band network telemetry message, and then obtain the forwarding paths in the network corresponding to each five-tuple combination and save them. Among them, the physical path between any two switches can be uniquely identified by the switch ID, the outgoing port number, and the incoming port number.

[0049] S130. Perform a fault detection on the physical path of the data center network based on multiple forwarding paths to obtain a fault detection result.

[0050] In the embodiments of the present disclosure, the fault detection includes link detection and / or switch configuration detection.

[0051] The fault detection result may include a link detection result for indicating whether there is a link disconnection in the physical path and / or a switch configuration detection result for indicating whether there is hash unevenness in any one of the switches in the physical path.

[0052] Specifically, the electronic device can obtain multiple forwarding paths by sending a forwarding path acquisition request from a preset database or to the server corresponding to the receiving end, and perform link detection of the physical path of the data center network and / or switch configuration detection based on the corresponding switch information, five-tuple information, initial configuration information of the switch, etc. in the multiple forwarding paths.

[0053] In the embodiments of the present disclosure, for each target port corresponding to each device in the data center network, the target port can be used as the source port to be traversed and combined with other target devices except the device to generate multiple in-band network telemetry messages. The protocol type of the in-band network telemetry message is a remote direct memory access protocol based on converged Ethernet. After generating multiple in-band network telemetry messages, for each in-band network telemetry message, the in-band network telemetry message is forwarded to the corresponding receiving end based on the five-tuple information of the in-band network telemetry message, so that the receiving end generates a forwarding path based on at least one switch information corresponding in the received in-band network telemetry message, and obtains multiple forwarding paths. The at least one switch information is the configuration information and forwarding status information corresponding to each switch passed through during the forwarding process of the in-band network telemetry message. The physical path of the data center network is subjected to fault detection based on the multiple forwarding paths to obtain a fault detection result. Thus, corresponding in-band network telemetry messages can be generated for each target port to ensure that each physical path can be covered when the in-band network telemetry message is forwarded. At the same time, by generating in-band network telemetry messages with a protocol type of a remote direct memory access protocol based on converged Ethernet for message forwarding, the message is forwarded as real traffic, and switch information is obtained during the message forwarding process. Furthermore, based on the obtained switch information and forwarding path, fault detection of the switch configuration is realized.

[0054] In the above embodiments of the present disclosure, using the target port as the source port to be traversed and combined with other target devices except the device to generate multiple in-band network telemetry messages may specifically include: obtaining a preset message format; for each target device, modifying the target field in the in-band flow analysis message header corresponding to the preset message format for indicating the next header type to the user datagram protocol number; modifying the source port number field in the user datagram protocol message header corresponding to the preset message format to the port number corresponding to the target port, and modifying the destination port number field to the dedicated port number of the remote direct memory access based on converged Ethernet; adding the remote direct memory access message header based on converged Ethernet to the message payload field in the preset message format; determining the IP address of the device corresponding to the target port as the source IP address and the IP address of the target device as the destination IP address to generate an in-band network telemetry message.

[0055] In the embodiments of the present disclosure, the preset message format may be a preset INT message format, such asFigure 2 As shown, the preset message format can be Figure 2 the message format shown on the left, including Layer2, Layer3, IFA header (i.e., INT message header), UDP Header (i.e., User Datagram Protocol message header), IFA Metadata Header (i.e., message header for collecting information), Termination Function Metadata (i.e., collected information of the tail node), Transit Function hop 0-N Metadata (i.e., collected information of intermediate nodes), Intiating Function Metadata (i.e., collected information of the head node), and Payload (i.e., message payload).

[0056] Specifically, after the electronic device obtains the preset message format, first, it modifies the target field in the in-band flow analysis message header corresponding to the preset message format that indicates the next header type to the User Datagram Protocol number 17 at the RoCE v2 bottom layer; second, it modifies the source port number field in the User Datagram Protocol message header corresponding to the preset message format to the port number corresponding to the target port, that is, it modifies the source port number through the probe program to control the message forwarding path; third, it modifies the destination port number field in the User Datagram Protocol message header to the dedicated port number of Remote Direct Memory Access based on Converged Ethernet, that is, the RoCE v2 dedicated port number 4791. By modifying the destination port number to the RoCE v2 dedicated port number 4791, it is possible to forward the in-band network telemetry message as real traffic; fourth, it adds the RoCE v2 message header to the message payload field in the preset message format to implement message encapsulation. Further, it determines the IP address of the device corresponding to the target port as the source IP address and the IP address of the target device as the destination IP address to generate the in-band network telemetry message.

[0057] In the embodiments of the present disclosure, it is possible to generate an in-band network telemetry message by modifying the fields in the preset message format, so that the message protocol type of the generated in-band network telemetry message is the Remote Direct Memory Access protocol based on Converged Ethernet. At the same time, by modifying the source port number, it is possible to control the message forwarding path to ensure that all paths are covered by traffic.

[0058] Figure 3 is a schematic diagram of a message forwarding process provided by the embodiments of the present disclosure, as Figure 3As shown in the figure, the process of forwarding an in-band network telemetry message to the corresponding receiver based on the five-tuple information of the in-band network telemetry message is as follows: After generating the in-band network telemetry message, determine that the sender is NIC1 and the receiver is NIC7 based on the five-tuple information of the in-band network telemetry message. Determine that the first node of the in-band network telemetry message is the top-of-rack switch 1, i.e., TOR switch 1. At this time, send the in-band network telemetry message to the top-of-rack switch 1. The top-of-rack switch 1 forwards the in-band network telemetry message to the aggregation switch 1, i.e., AGG switch 1, and appends switch information 1 to the message. After receiving the in-band network telemetry message, the aggregation switch 1 forwards the message to the top-of-rack switch 3 and appends switch information 2 to the message at the same time. Similarly, after receiving the in-band network telemetry message, the top-of-rack switch 3 forwards the message to NIC7 and appends switch information 3 to the message. Furthermore, the server corresponding to NIC7 receives the in-band network telemetry message appended with switch information 1, switch information 2, and switch information 3, generates a forwarding path based on switch information 1, switch information 2, switch information 3, and the corresponding five-tuple information, and saves it. Thus, the forwarding path corresponding to each five-tuple information is obtained.

[0059] In an embodiment of the present disclosure, when the fault detection result is a link detection result used to characterize whether there is a link disconnection in the physical path, the physical path of the data center network is fault-detected based on multiple forwarding paths to obtain the fault detection result. Specifically, it may include: For each switch, obtain multiple target forwarding paths including the switch, and determine the target input port number and the target output port number of the switch corresponding to the multiple target forwarding paths; determine the number of entry ports and the number of exit ports through which the traffic flowing through the switch enters the switch based on the target input port number and the target output port number; obtain the port information of the switch, and / or, the number of target entry ports and the number of target exit ports of other switches in the same layer as the switch; compare the number of entry ports and the number of exit ports with the port information of the switch, and / or, compare the number of entry ports and the number of exit ports with the number of target entry ports and the number of target exit ports, and determine the fault detection result based on the comparison result.

[0060] In an embodiment of the present disclosure, the port information of the switch can be understood as the port configuration information of the switch, such as the number of ports.

[0061] In some embodiments of the present disclosure, the electronic device can compare the number of entry ports and the number of exit ports through which the traffic flowing through the switch enters the switch with the number of ports in the port information of the switch to obtain a first comparison result. When the first comparison result is that the number of entry ports and the number of exit ports match the number of ports, it is determined that there is no link disconnection; otherwise, vice versa.

[0062] Exemplarily, for switch A, the port information in the initial configuration information of switch A is that the total number of ports is 3, including port 1, port 2, and port 3. After obtaining multiple target forwarding paths including switch A based on the ID information of switch A, if the target ingress port numbers and target egress port numbers of switch A in the multiple target forwarding paths are only port 1 and port 2 and do not include port 3, at this time, it indicates that there is a link disconnection situation for port 3. If the target ingress port numbers and target egress port numbers of switch A in the multiple target forwarding paths include port 1, port 2, and port 3, at this time, it indicates that there is no link disconnection situation for this switch. When there is no link disconnection situation for all switches, it is determined that there is no link disconnection in the physical path. When there is a link disconnection situation for at least one switch, it is determined that there is a link disconnection in the physical path.

[0063] In some other embodiments of the present disclosure, the electronic device can compare the number of ingress ports and the number of egress ports with the target number of ingress ports and the target number of egress ports of other switches at the same layer to obtain a second comparison result. When the second comparison result is that the number of ingress ports and the number of egress ports are the same as the target number of ingress ports and the target number of egress ports of other switches at the same layer, that is, there is no difference, it is determined at this time that no link disconnection has occurred; otherwise, vice versa. Thus, it is possible to determine whether a path failure has occurred based on the target number of ingress ports and the target number of egress ports of other switches at the same layer, improving the accuracy and comprehensiveness of path failure detection.

[0064] Exemplarily, for switch A, the number of ingress ports is 7 and the number of egress ports is 8. For other switch B at the same layer as switch A, the number of ingress ports is 8 and the number of egress ports is 8. For other switch C, the number of ingress ports is 8 and the number of egress ports is 8. At this time, switch A has only 7 ingress ports, while the ingress ports of other switches at the same layer are all 8. At this time, it is determined that switch A has a path failure.

[0065] In some other embodiments of the present disclosure, the electronic device can compare the number of ingress ports and the number of egress ports through which the traffic flowing through the switch enters the switch with the number of ports in the port information of the switch to obtain a first comparison result. At the same time, compare the number of ingress ports and the number of egress ports with the target number of ingress ports and the target number of egress ports of other switches at the same layer to obtain a second comparison result. When the first comparison result is that the number of ingress ports and the number of egress ports match the number of ports, and the second comparison result is that the number of ingress ports and the number of egress ports are the same as the target number of ingress ports and the target number of egress ports of other switches at the same layer, it is determined that no link disconnection has occurred, that is, there is no path failure; otherwise, vice versa.

[0066] In the embodiments of the present disclosure, it is possible to perform link detection on the physical path of the data center network based on the actual service traffic forwarding, improving the accuracy of link detection.

[0067] In an embodiment of the present disclosure, when the fault detection result is a switch configuration detection result used to characterize whether any switch in the physical path has hash imbalance, the physical path of the data center network is fault-detected based on multiple forwarding paths to obtain the fault detection result, which may specifically include: for each switch, determining the number of packets flowing through each port of the switch in the multiple forwarding paths; comparing the number of packets flowing through with the target number of packets flowing through the other ports of the switch, and when the comparison result shows that the difference in the number of packets flowing through is greater than a preset threshold, determining that the switch has hash imbalance.

[0068] Specifically, after obtaining multiple forwarding paths, the electronic device, for each switch, determines the number of packets flowing through each port of the switch respectively based on the five-tuple information corresponding to the multiple forwarding paths, compares the number of packets flowing through with the target number of packets flowing through the other ports of the switch. When the difference in the number of packets flowing through between multiple ports of the switch is greater than the preset threshold, it indicates that there is an ECMP hash imbalance phenomenon in the switch, and thus it is determined that the switch has a configuration fault; when the difference in the number of packets flowing through between multiple ports of the switch is less than or equal to the preset threshold, it indicates that there is no ECMP hash imbalance phenomenon in the switch, and thus it is determined that the switch has no configuration fault.

[0069] In an embodiment of the present disclosure, switch configuration detection can be performed through the generated multiple forwarding paths and the number of packets flowing through each port of the switch, improving the accuracy of configuration detection.

[0070] Figure 4 It is a schematic structural diagram of a physical path fault detection device provided by an embodiment of the present disclosure.

[0071] In an embodiment of the present disclosure, the physical path fault detection device can be disposed in an electronic device and is understood as some functional modules in the above-mentioned electronic device. Specifically, the electronic device can be a server or a terminal, where the terminal specifically includes a mobile phone, a computer, a tablet computer, etc., which is not limited herein.

[0072] As Figure 4 shown, the physical path fault detection device 400 may include a packet generation module 410, a packet forwarding module 420, and a fault detection module 430.

[0073] The packet generation module 410 can be used to, for each target port corresponding to each device in the data center network, use the target port as the source port to traverse and combine with other target devices except the device to generate multiple in-band network telemetry packets, and the protocol type of the in-band network telemetry packets is the remote direct memory access protocol based on converged Ethernet.

[0074] The packet forwarding module 420 can be used to forward each in-band network telemetry packet to the target switch corresponding to the device, so that the target switch forwards the in-band network telemetry packet to the corresponding receiving end based on the five-tuple information of the in-band network telemetry packet. After receiving the in-band network telemetry packet, the receiving end generates a forwarding path based on at least one switch information corresponding in the received in-band network telemetry packet, obtaining multiple forwarding paths. The at least one switch information is the configuration information and forwarding status information respectively corresponding to each switch passed through during the forwarding process of the in-band network telemetry packet.

[0075] The fault detection module 430 can be used to perform fault detection on the physical path of the data center network based on multiple forwarding paths, obtaining a fault detection result.

[0076] In the embodiments of the present disclosure, for each target port corresponding to each device in the data center network, the target port can be used as the source port to perform traversal combination with other target devices except the device, generating multiple in-band network telemetry packets. The protocol type of the in-band network telemetry packet is the remote direct memory access protocol based on converged Ethernet; after generating multiple in-band network telemetry packets, for each in-band network telemetry packet, the in-band network telemetry packet is forwarded to the target switch corresponding to the device, so that the target switch forwards the in-band network telemetry packet to the corresponding receiving end based on the five-tuple information of the in-band network telemetry packet. After receiving the in-band network telemetry packet, the receiving end generates a forwarding path based on at least one switch information corresponding in the received in-band network telemetry packet, obtaining multiple forwarding paths. The at least one switch information is the configuration information and forwarding status information respectively corresponding to each switch passed through during the forwarding process of the in-band network telemetry packet; perform fault detection on the physical path of the data center network based on multiple forwarding paths, obtaining a fault detection result. Thus, corresponding in-band network telemetry packets can be generated for each target port to ensure that each physical path can be covered when forwarding the in-band network telemetry packet. At the same time, by generating in-band network telemetry packets with the protocol type of the remote direct memory access protocol based on converged Ethernet for packet forwarding, it realizes forwarding the packets as real service traffic, obtaining switch information during the packet forwarding process, and then realizing fault detection of the switch configuration according to the obtained switch information and forwarding paths.

[0077] In some embodiments of the present disclosure, the packet generation module 410 can be specifically used to obtain a preset packet format;

[0078] For each target device, modify the target field in the in-band flow analysis packet header corresponding to the preset packet format for indicating the next header type to the user datagram protocol number;

[0079] Modify the source port number field in the User Datagram Protocol (UDP) header corresponding to the preset message format to the port number corresponding to the destination port, and modify the destination port number field to the dedicated port number for Remote Direct Memory Access (RDMA) over Converged Ethernet (RoCE);

[0080] Add the header for RDMA over Converged Ethernet to the message payload field in the preset message format;

[0081] Determine the IP address of the device corresponding to the destination port as the source IP address, and determine the IP address of the destination device as the destination IP address to generate an in-band network telemetry message.

[0082] In some embodiments of the present disclosure, at least one switch information is information appended to the in-band network telemetry message after at least one switch forwards the in-band network telemetry message, including the ID information, incoming port number, and outgoing port number corresponding to each switch passed through during the forwarding process of the in-band network telemetry message.

[0083] In some embodiments of the present disclosure, the fault detection module 430 may specifically be used when the fault detection result is a link detection result used to characterize whether there is a link disconnection in the physical path. For each switch, obtain multiple target forwarding paths including the switch, and determine the target incoming port number and target outgoing port number of the switch corresponding to the multiple target forwarding paths;

[0084] Determine the number of incoming ports and the number of outgoing ports through which the traffic flowing through the switch enters the switch based on the target incoming port number and target outgoing port number;

[0085] Obtain the port information of the switch, and / or the target number of incoming ports and target number of outgoing ports of other switches at the same layer as the switch;

[0086] Compare the number of incoming ports and the number of outgoing ports with the port information of the switch, and / or compare the number of incoming ports and the number of outgoing ports with the target number of incoming ports and target number of outgoing ports, and determine the fault detection result based on the comparison result.

[0087] In some embodiments of the present disclosure, the fault detection module 430 may also specifically be used when the fault detection result is a switch configuration detection result used to characterize whether there is hash imbalance in any switch in the physical path. For each switch, determine the number of packets flowing through each port of the switch corresponding to the multiple forwarding paths;

[0088] Compare the number of packets flowing through with the target number of packets flowing through the other ports of the switch. When the comparison result shows that the difference in the number of packets flowing through is greater than a preset threshold, determine that there is hash imbalance in the switch.

[0089] It should be noted that Figure 4The physical path fault detection device 400 shown can execute each step in the above method embodiments, and implement each process and effect in the above method embodiments, which will not be elaborated here.

[0090] Figure 5 It is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.

[0091] In the embodiments of the present disclosure, Figure 5 The electronic device shown can be a server or a terminal. Among them, the terminal specifically includes a mobile phone, a computer, a tablet computer, etc., which are not limited here.

[0092] Such as Figure 5 As shown, the electronic device may include a processor 510 and a memory 520 storing computer program instructions.

[0093] Specifically, the above-mentioned processor 510 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present disclosure.

[0094] The memory 520 may include a mass storage for information or instructions. By way of example and not limitation, the memory 520 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In a suitable case, the memory 520 may include removable or non-removable (or fixed) media. In a suitable case, the memory 520 may be inside or outside the integrated gateway device. In a specific embodiment, the memory 520 is a non-volatile solid-state memory. In a specific embodiment, the memory 520 includes a read-only memory (ROM). In a suitable case, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0095] The processor 510 reads and executes the computer program instructions stored in the memory 520 to perform the steps of the physical path fault detection method provided by the embodiments of the present disclosure.

[0096] In one example, the electronic device may further include a transceiver 530 and a bus 540. Among them, as Figure 5 shown, the processor 510, the memory 520, and the transceiver 530 are connected through the bus 540 and complete communication with each other.

[0097] The bus 540 includes hardware, software, or both. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side BUS (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. In a suitable case, the bus 540 may include one or more buses.

[0098] The embodiments of the present disclosure also provide a computer-readable storage medium that may store a computer program. When the computer program is executed by a processor, the processor implements the physical path fault detection method provided by the embodiments of the present disclosure.

[0099] The above storage medium may include, for example, a memory 520 storing computer program instructions, and the above instructions may be executed by a processor 510 of an electronic device to implement the physical path fault detection method provided by the embodiments of the present disclosure. Optionally, the storage medium may be a non-transitory computer-readable storage medium. For example, the non-transitory computer-readable storage medium may be ROM, random access memory (RAM), compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0100] The embodiments of the present disclosure further provide a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a processor, the physical path fault detection method provided by the embodiments of the present disclosure is implemented, and the various processes and effects in the above embodiments of the present disclosure can be achieved. Details are not described herein again.

[0101] The above description is only the specific implementation manners of the present disclosure, which enable those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A physical path failure detection method, characterized in that: include: For each target port corresponding to each device in the data center network, the target port is used as a source port to traverse and combine with other target devices except the device to generate multiple in-band network telemetry messages, and the protocol type of the in-band network telemetry message is a remote direct memory access protocol based on converged Ethernet; For each in-band network telemetry message, forward the in-band network telemetry message to a target switch corresponding to the device, so that the target switch forwards the in-band network telemetry message to a corresponding receiving end based on the five-tuple information of the in-band network telemetry message, and after receiving the in-band network telemetry message, the receiving end generates a forwarding path based on at least one switch information corresponding to the received in-band network telemetry message to obtain multiple forwarding paths, wherein the at least one switch information is configuration information and forwarding state information corresponding to each switch passed through during the forwarding process of the in-band network telemetry message; Fault detection is performed on the physical paths of the data center network based on the multiple forwarding paths to obtain a fault detection result.

2. The method according to claim 1, characterized in that The step of using the target port as a source port and traversing and combining with other target devices other than the device to generate multiple in-band network telemetry messages includes: Get the preset message format; For each target device, modifying the target field for indicating the next header type in the in-band flow analysis message header corresponding to the preset message format to a user datagram protocol number; Modify the source port number field in the user datagram protocol message header corresponding to the preset message format to the port number corresponding to the target port, and modify the destination port number field to the dedicated port number of remote direct memory access based on converged Ethernet; Adding a message header of remote direct memory access based on converged Ethernet to a message payload field in the preset message format; The IP address of the device corresponding to the target port is determined as the source IP address, and the IP address corresponding to the target device is determined as the destination IP address to generate an in-band network telemetry message.

3. The method according to claim 1, characterized in that The at least one switch information is information added to the in-band network telemetry message by the at least one switch after forwarding the in-band network telemetry message, including ID information, ingress port number and egress port number corresponding to each switch passed through during the forwarding process of the in-band network telemetry message.

4. The method according to claim 1, characterized in that When the fault detection result is a link detection result for characterizing whether a physical path has a link disconnection, the performing fault detection on the physical path of the data center network based on the multiple forwarding paths to obtain the fault detection result includes: For each switch, obtain multiple target forwarding paths including the switch, and determine the target inbound port number and the target outbound port number of the switch corresponding to the multiple target forwarding paths; Determine the number of ingress ports and the number of egress ports of the traffic flowing through the switch based on the target ingress port number and the target egress port number; Obtaining port information of the switch, and / or target ingress numbers and target egress numbers of other switches at the same layer as the switch; The number of inlets and the number of outlets are compared with the port information of the switch, and / or the number of inlets and the number of outlets are compared with the target number of inlets and the target number of outlets, and the fault detection result is determined based on the comparison result.

5. The method according to claim 1, characterized in that When the fault detection result is a switch configuration detection result for characterizing whether any switch in the physical path has a hash unevenness, the performing fault detection on the physical path of the data center network based on the multiple forwarding paths to obtain the fault detection result includes: For each switch, determine the number of packets flowing through each port of the switch in the multiple forwarding paths; The number of packets passing through is compared with target packet data passing through corresponding to other ports of the switch, and when the comparison result shows that the difference in the number of packets passing through is greater than a preset threshold, it is determined that the switch has a hash unevenness.

6. A physical path fault detection device, characterized in that: include: A message generation module is used for, for each target port corresponding to each device in the data center network, taking the target port as a source port and traversing and combining it with other target devices except the device, to generate multiple in-band network telemetry messages, wherein the protocol type of the in-band network telemetry message is a remote direct memory access protocol based on a converged Ethernet; a message forwarding module, configured to forward each in-band network telemetry message to a target switch corresponding to the device, so that the target switch forwards the in-band network telemetry message to a corresponding receiving end based on the five-tuple information of the in-band network telemetry message, and after receiving the in-band network telemetry message, the receiving end generates a forwarding path based on at least one switch information corresponding to the received in-band network telemetry message, and obtains a plurality of forwarding paths, wherein the at least one switch information is configuration information and forwarding state information respectively corresponding to each switch passed through during the forwarding process of the in-band network telemetry message; A fault detection module is used to perform fault detection on the physical paths of the data center network based on the multiple forwarding paths to obtain a fault detection result.

7. The device according to claim 6, characterized in that The message generation module is specifically used to obtain a preset message format; For each target device, modifying the target field for indicating the next header type in the in-band flow analysis message header corresponding to the preset message format to a user datagram protocol number; Modify the source port number field in the user datagram protocol message header corresponding to the preset message format to the port number corresponding to the target port, and modify the destination port number field to the dedicated port number of remote direct memory access based on converged Ethernet; Adding a message header of remote direct memory access based on converged Ethernet to a message payload field in the preset message format; The IP address of the device corresponding to the target port is determined as the source IP address, and the IP address corresponding to the target device is determined as the destination IP address to generate an in-band network telemetry message.

8. An electronic device, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the physical path failure detection method according to any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the physical path failure detection method according to any one of claims 1 to 5.

10. A computer program product, comprising a computer program or instructions, characterized in that: When the computer program or instruction is executed by a processor, the physical path failure detection method according to any one of claims 1 to 5 is implemented.

Images