Network telemetry node configuration method and apparatus, switch, and storage medium

CN117834425BActive Publication Date: 2026-09-11INSPUR SUZHOU INTELLIGENT TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202311541680.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-17
Publication Date
2026-09-11
Estimated Expiration
2043-11-17

AI Technical Summary

Technical Problem

[0003]本发明提供一种网络遥测节点配置方法、装置、交换机及存储介质,用以解决现有网络网络遥测节点配置方法会造成控制器负担重,并且,配置调整具有滞后性,不利于网络稳定的缺陷

Benefits of technology

[0039]本发明提供的网络遥测节点配置方法、装置、交换机及存储介质,通过响应于遥测节点端口状态开启事件,通过该端口向遥测节点的所有相邻节点发送第一探测报文;监测相邻节点对所述第一探测报文的响应信息,根据相邻节点对所述第一探测报文的响应信息配置遥测节点的角色;接收遥测节点的相邻节点发送的第二探测报文,解析每个相邻节点的第二探测报文中携带的遥测信息内容,根据解析结果配置遥测节点的遥测信息内容,通过各个节点之间交换探测报文,可以使节点自适应生成遥测功能配置,可以大幅减少控制器负担,尤其是网络拓扑结构发生变动时,无需控制器介入即可自适应更改交换机节点上遥测功能配置,减少配置延时,提高网络稳定性。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117834425B_ABST
    Figure CN117834425B_ABST
Patent Text Reader

Abstract

The application provides a network telemetry node configuration method and device, a switch and a storage medium. In response to a telemetry node port state opening event, a first probe packet is sent to all adjacent nodes of the telemetry node through the port; response information of the adjacent nodes to the first probe packet is monitored, and a role of the telemetry node is configured according to the response information of the adjacent nodes to the first probe packet; a second probe packet sent by an adjacent node of the telemetry node is received, telemetry information content carried in the second probe packet of each adjacent node is analyzed, and telemetry information content of the telemetry node is configured according to the analysis result. Through the exchange of probe packets between nodes, the nodes can adaptively generate telemetry function configuration, the burden on the controller can be greatly reduced, and especially when the network topology structure changes, the telemetry function configuration on the switch node can be adaptively changed without the intervention of the controller, the configuration delay is reduced, and the network stability is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of switch technology, and in particular to a network telemetry node configuration method, apparatus, switch, and storage medium. Background Technology

[0002] In telemetry network applications, the authorization code mode involves complex operations. In this mode, switches are divided into endpoint switches and transit switches. Endpoint switches handle raw packets entering the telemetry network, adding an INT (Inband Network Telemetry) header to these packets before passing them to transit switches. For packets leaving the telemetry network, endpoint switches strip the telemetry data and report them to the controller. In traditional network telemetry, node configuration is done by the controller for each switch, such as configuring telemetry nodes as endpoints or transit nodes, and specifying the telemetry information to be carried in the telemetry packets. However, the controller needs a deep understanding of the network's operational status and performance requirements to make appropriate configuration decisions. Furthermore, since network topology and requirements may change, continuous monitoring and adjustment of telemetry node configurations are necessary, placing a heavy burden on the controller. The controller also needs to monitor network structure changes first and then adjust the telemetry configuration accordingly, resulting in a lag that is detrimental to network stability. Summary of the Invention

[0003] This invention provides a network telemetry node configuration method, device, switch, and storage medium to solve the defects of existing network telemetry node configuration methods, which cause heavy controller burden and have lag in configuration adjustment, thus hindering network stability.

[0004] This invention provides a method for configuring network telemetry nodes, comprising:

[0005] In response to a telemetry node port status open event, a first probe message is sent to the neighboring nodes of the telemetry node through the telemetry node port;

[0006] Monitor the response information of the neighboring nodes to the first probe message, and configure the role of the telemetry node according to the response information of the neighboring nodes to the first probe message;

[0007] The telemetry node receives second probe messages sent by neighboring nodes, parses the telemetry information content carried in the second probe messages of each neighboring node, and configures the telemetry information content of the telemetry node according to the parsing results.

[0008] According to a network telemetry node configuration method provided by the present invention, configuring the role of the telemetry node based on the response information of the neighboring node to the first probe message includes:

[0009] If all ports of the telemetry node receive the response information of the first probe message fed back by the neighboring node, then the telemetry node is configured as a relay node.

[0010] If some ports of the telemetry node receive response information of the first probe message fed back by the neighboring node, or if no port of the telemetry node receives response information of the first probe message fed back by the neighboring node, then the role of the telemetry node is configured as an endpoint node.

[0011] According to a network telemetry node configuration method provided by the present invention, after configuring the telemetry node role as an endpoint node, the method further includes:

[0012] Configure the port in the endpoint node that responds to probe messages sent by other nodes as a termination port, and add the termination port to the termination port list.

[0013] According to a network telemetry node configuration method provided by the present invention, the first probe message or the second probe message includes:

[0014] Ether type value, source code, object code, and payload;

[0015] The Ethernet type value is different from the Ethernet type value of the protocol used for communication between nodes;

[0016] The source code is used to identify the nodes that send probe messages;

[0017] The target code is used to identify the node receiving the probe message;

[0018] The payload includes type, length, and parameter values. The type includes whether it supports data plane telemetry function, standard interface protocol, and private interface protocol.

[0019] The standard interface protocol includes a standard protocol type code, a standard protocol type description, and standard protocol fields. The standard protocol type code is used to identify the protocol interface; the standard protocol type description is used to describe the specific protocol name corresponding to each standard protocol type code; the standard protocol fields include at least one of the following: switch identification code, inbound / outbound switch port, inbound port timestamp, outbound port timestamp, queue information, forwarding delay, outbound port utilization, and queue buffer utilization.

[0020] The private interface protocol includes a private protocol type code, a private protocol type description, and private protocol fields. The private protocol type code is used to privately identify the protocol interface; the private protocol type description is used to describe the specific protocol name corresponding to each private protocol type code; the private protocol fields include support for obtaining current queue packet loss statistics and / or support for obtaining current outgoing port packet transmission rate.

[0021] The private protocol field is defined according to the requirements of the node device manufacturer. It uses strings to describe the parameter names in the protocol, and multiple parameters are separated by commas.

[0022] The telemetry information in the payload is uploaded to the central processing unit (CPU) via an access control list, so that the CPU can parse and obtain the telemetry information carried by each node.

[0023] According to a network telemetry node configuration method provided by the present invention, configuring the telemetry information content of the telemetry node based on the parsing result includes:

[0024] Parse the telemetry information sent by each neighboring node and obtain the protocol fields in the standard interface protocol and the protocol fields in the private interface protocol of each neighboring node;

[0025] The protocol fields in the standard interface protocol shared by each adjacent node are used as the standard interface protocol content in the telemetry information content of the telemetry node, and the protocol fields in the private interface protocol shared by each adjacent node are used as the private interface protocol content in the telemetry information content of the telemetry node.

[0026] According to a network telemetry node configuration method provided by the present invention, if the neighboring nodes of the telemetry node are a first neighboring node, a second neighboring node, and a third neighboring node, configuring the telemetry information content of the telemetry node according to the parsing result includes:

[0027] Parse the telemetry information sent by the first adjacent node, and obtain the standard interface protocol fields in the first adjacent node as switch identification code, ingress port timestamp, egress port timestamp and forwarding delay, and obtain the private interface protocol fields in the first adjacent node as supporting the acquisition of current queue packet loss statistics and supporting the acquisition of current egress port packet transmission rate;

[0028] Parse the telemetry information sent by the second adjacent node, and obtain the standard interface protocol fields in the second adjacent node as switch identification code, ingress port timestamp, queue information and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of current queue packet loss statistics;

[0029] Parse the telemetry information sent by the third adjacent node, and obtain the standard interface protocol fields in the third adjacent node as switch identification code, outgoing port timestamp and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of the current outgoing port packet rate;

[0030] The standard interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node are intersected to obtain the standard interface protocol fields of the telemetry node. The standard interface protocol fields of the telemetry node include the switch identification code and the forwarding delay.

[0031] Take the intersection of the private interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node. If the intersection result is empty, configure the private interface protocol field of the telemetry node to be empty.

[0032] According to a network telemetry node configuration method provided by the present invention, the method further includes: adding the network telemetry node configuration result to a telemetry configuration database, wherein the telemetry configuration database includes switch role data, a list of termination ports, and fields required to be carried by telemetry information.

[0033] The present invention also provides a network telemetry node configuration device, comprising:

[0034] The sending module is used to send a first probe message to all adjacent nodes of the telemetry node through the telemetry node port in response to the telemetry node port status open event;

[0035] The first configuration module is used to monitor the response information of the neighboring nodes to the first probe message, and configure the role of the telemetry node according to the response information of the neighboring nodes to the first probe message.

[0036] The second configuration module is used to receive the second probe messages sent by the neighboring nodes of the telemetry node, parse the telemetry information content carried in the second probe messages of each neighboring node, and configure the telemetry information content of the telemetry node according to the parsing results.

[0037] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the network telemetry node configuration method described in any of the preceding claims.

[0038] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the network telemetry node configuration method described in any of the preceding claims.

[0039] The network telemetry node configuration method, apparatus, switch, and storage medium provided by this invention, in response to a telemetry node port status open event, send a first probe message to all neighboring nodes of the telemetry node through the port; monitor the response information of neighboring nodes to the first probe message, configure the role of the telemetry node according to the response information of neighboring nodes to the first probe message; receive second probe messages sent by neighboring nodes of the telemetry node, parse the telemetry information content carried in the second probe message of each neighboring node, and configure the telemetry information content of the telemetry node according to the parsing result. By exchanging probe messages between nodes, the nodes can adaptively generate telemetry function configurations, which can significantly reduce the controller burden. Especially when the network topology changes, the telemetry function configuration on the switch node can be adaptively changed without controller intervention, reducing configuration latency and improving network stability. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is one of the flowcharts illustrating the network telemetry node configuration method provided by the present invention;

[0042] Figure 2 This is a schematic diagram of the telemetry network provided by the present invention;

[0043] Figure 3 This is a schematic diagram of the role of a telemetry network-level switch provided by the present invention;

[0044] Figure 4 This is the second flowchart illustrating the network telemetry node configuration method provided by the present invention;

[0045] Figure 5 This is a functional structure diagram of the network telemetry node configuration device provided by the present invention;

[0046] Figure 6 This is a functional structure diagram of the switch provided by the present invention;

[0047] Figure 7 This is a schematic diagram of the switch network architecture provided by the present invention;

[0048] Figure 8 This is the timing diagram of the switch network architecture interaction provided by the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0050] Figure 1 The flowchart of the network telemetry node configuration method provided in the embodiments of the present invention is as follows: Figure 1 As shown, the network telemetry node configuration method provided in this embodiment of the invention includes:

[0051] Step 101: In response to the telemetry node port being enabled, send the first probe message to the neighboring nodes of the telemetry node through the telemetry node port;

[0052] In this embodiment of the invention, the network where the telemetry node is located is an INT (Inband Network Telemetry) network, which is a novel network monitoring technology. It achieves comprehensive monitoring and analysis of data flows by adding special metadata to network data packets. This metadata can provide rich network performance information, such as network congestion, latency, and packet loss, and can also be used to identify network attacks and abnormal behavior.

[0053] Step 102: Monitor the response information of neighboring nodes to the first probe message, and configure the role of the telemetry node according to the response information of neighboring nodes to the first probe message;

[0054] In this embodiment of the invention, telemetry nodes are such as switches, and the roles of switches include endpoint switches and trunk switches. This role configuration affects the specific operating mode of the switch and its position in the network topology. Switch roles can be defined according to the specific needs and configuration of the network, and different types of switches have different functions in the network.

[0055] Step 103: Receive the second probe message sent by the neighboring nodes of the telemetry node, parse the telemetry information content carried in the second probe message of each neighboring node, and configure the telemetry information content of the telemetry node according to the parsing result.

[0056] In this embodiment of the invention, the telemetry information that needs to be carried in the telemetry message may vary depending on the performance of the switch chip in the environment. For example, if a switch that does not support outgoing port timestamps is added to the network, the timestamp function on this telemetry path will also be unusable because the information of this hop is missing, and the time will be inaccurate.

[0057] In traditional network telemetry, node configuration is done by the controller for each switch. This includes configuring telemetry nodes as endpoints or relays, and specifying the telemetry information to be carried in the telemetry messages. Figure 2 As shown, the controller occupies a central position in the network, controlling the collection, storage, and transmission of data packets, and generating INT reports to provide network performance information. The controller can interact with the data plane, sending configuration information and control commands through the data plane's API, as well as collecting INT reports and network traffic information. The controller can also manage the collection and storage of INT data, providing data for analysis and monitoring. However, when configuring telemetry nodes, the controller needs a deep understanding of the network's operational status and performance requirements to make appropriate configuration decisions. Furthermore, because network topology and requirements may change, continuous monitoring and adjustment of telemetry node configurations are necessary, placing a heavy burden on the controller. Moreover, the controller needs to monitor network structure changes first, and then adjust the telemetry configuration accordingly, resulting in a lag that is detrimental to network stability.

[0058] The network telemetry node configuration method provided in this embodiment of the invention sends a first probe message to all neighboring nodes of the telemetry node in response to the opening of the telemetry node port; monitors the response information of neighboring nodes to the first probe message, configures the role of the telemetry node according to the response information of neighboring nodes to the first probe message; receives a second probe message sent by the neighboring nodes of the telemetry node, parses the telemetry information content carried in the second probe message of each neighboring node, and configures the telemetry information content of the telemetry node according to the parsing result. By exchanging probe messages between nodes, the nodes can adaptively generate telemetry function configurations, which can significantly reduce the controller burden. Especially when the network topology changes, the telemetry function configuration on the switch node can be adaptively changed without controller intervention, reducing configuration latency and improving network stability.

[0059] In this embodiment of the invention, telemetry function configuration refers to configuring the standard interface DTEL (DataPlane Telemetry) for implementing telemetry functions. To ensure seamless communication between upper-layer software and hardware, the Open Compute Project (OCP) has introduced a standard interface called DTEL. The introduction of this interface provides a universal calling method for all switch chips; as long as the chip supports this interface, it can support standardized telemetry functions.

[0060] Based on any of the above embodiments, configuring the role of the telemetry node according to the response information of neighboring nodes to the first probe message includes:

[0061] If all ports of the telemetry node receive the response information of the first probe message from the neighboring node, then the telemetry node role is configured as a relay node.

[0062] If some ports of the telemetry node receive the response information of the first probe message fed back by the neighboring node, or if no port of the telemetry node receives the response information of the first probe message fed back by the neighboring node, then the telemetry node role is configured as an endpoint node.

[0063] Since all ports on a trunk switch connect to other switches that support telemetry, while some ports on an endpoint switch connect to machines that do not support telemetry (such as servers), a proprietary protocol resolution method can be used to automatically identify these connections. When a switch port is enabled, it sends a proprietary protocol probe message. If a neighboring switch responds, it can be determined that the other party is a switch and also supports DTEL functionality. If there is no response, it indicates that the other party may be a server or a switch that does not support telemetry. Based on the response to the probe message, if all ports on a switch respond, that switch can be configured as a trunk switch; otherwise, it can be configured as an endpoint switch.

[0064] In this embodiment of the invention, after configuring the telemetry node role as an endpoint node, the method further includes:

[0065] Configure the port in the endpoint node that responds to probe messages sent by other nodes as a sink port. These ports are usually ports facing the telemetry network range. Add the sink ports to the sink port list.

[0066] like Figure 3 As shown, the upper port of the first node on the left side of the leaf layer is the sink port, and the lower port, facing the server, is the source port. When it is necessary to measure the network status between ends, the system enters Passport mode. This mode can be configured on the switches through which the packets pass. These switch ports include the source and sink ports. When a data packet arrives at the source port, the switch adds an INT header and metadata to it according to the flow to be telemetryd. The relay switch then inserts metadata after the packet based on the INT header, and so on, until the last hop of the entire telemetry system reaches the endpoint switch. The endpoint switch removes all telemetry information before reporting it to the controller.

[0067] Endpoint switches are responsible for processing raw packets entering the telemetry network. They need to add an INT header to these packets before passing them to trunk switches; this step is called the "source action." For packets leaving the telemetry network, the endpoint switches need to strip the telemetry data and report them to the controller; this process is called the "sink action." On the endpoint switches, it's necessary to determine which port should perform the source action and which should perform the sink action. Therefore, a termination port list is designed. Ports of the endpoint switches not added to the termination port list can be considered as starting ports, thus determining the data flow direction on each endpoint switch.

[0068] Based on any of the above embodiments, the first detection message or the second detection message improved in the embodiments of the present invention includes:

[0069] The Ether type value, source code, destination code, and payload are specified. The Ether type value differs from the Ether type value used in the protocol used for communication between nodes, for example, Ethertype = 9156. The source code identifies the node sending the probe message, for example, Sourcemac = switch port mac. The destination code identifies the node receiving the probe message, for example, Destinationmac = FF:FF:FF:FF:FF:FF. The payload includes type, length, and parameter values, Payload = TLV(Type, length, value). The type includes whether data plane telemetry is supported, standard interface protocol, and private interface protocol.

[0070] The standard interface protocols and private interface protocols are shown in Table 1. The standard interface protocol includes a standard protocol type code, a standard protocol type description, and standard protocol fields. The standard protocol type code identifies the protocol interface; the standard protocol type description describes the specific protocol name corresponding to each standard protocol type code; the standard protocol fields include at least one of the following: switch identification code, inbound / outbound switch port, inbound port timestamp, outbound port timestamp, queue information, forwarding delay, outbound port utilization rate, and queue buffer utilization rate. The private interface protocol includes a private protocol type code, a private protocol type description, and private protocol fields. The private protocol type code is used to privately identify the protocol interface; the private protocol type description describes the specific protocol name corresponding to each private protocol type code; the private protocol fields include support for obtaining current queue packet loss statistics and / or support for obtaining current outbound port packet transmission rate.

[0071] Table 1. Contents of Standard Interface Protocols and Private Interface Protocols

[0072]

[0073]

[0074] The private protocol field is defined according to the requirements of the node device manufacturer. It uses strings to describe the parameter names in the protocol, with multiple parameters separated by commas.

[0075] The telemetry information in the payload is uploaded to the central processing unit (CPU) via an access control list, so that the CPU can parse and obtain the telemetry information carried by each node.

[0076] In this embodiment of the invention, the switch automatically identifies this special packet through the Access Control List (ACL) and passes it to the CPU for further analysis, thereby realizing the processing and identification of probe packets. Since the content of the sent probe packet includes the DTEL support status, the receiving machine can configure telemetry information content according to the DTEL support status of surrounding machines.

[0077] Based on any of the above embodiments, the telemetry information content of the telemetry node is configured according to the parsing result, including:

[0078] Parse the telemetry information sent by each neighboring node and obtain the protocol fields in the standard interface protocol and the protocol fields in the private interface protocol of each neighboring node;

[0079] The protocol fields in the standard interface protocol shared by each adjacent node are used as the standard interface protocol content in the telemetry information content of the telemetry node, and the protocol fields in the private interface protocol shared by each adjacent node are used as the private interface protocol content in the telemetry information content of the telemetry node.

[0080] For example, if the neighboring nodes of a telemetry node are the first neighboring node, the second neighboring node, and the third neighboring node, the telemetry information content of the telemetry node is configured according to the parsing result, including:

[0081] Parse the telemetry information sent by the first adjacent node, and obtain the standard interface protocol fields in the first adjacent node as switch identification code, ingress port timestamp, egress port timestamp and forwarding delay, and obtain the private interface protocol fields in the first adjacent node as support for obtaining current queue packet loss statistics and support for obtaining current egress port packet transmission rate;

[0082] Parse the telemetry information sent by the second adjacent node, and obtain the standard interface protocol fields in the second adjacent node as switch identification code, ingress port timestamp, queue information and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of current queue packet loss statistics;

[0083] Parse the telemetry information sent by the third neighboring node, and obtain the standard interface protocol fields in the third neighboring node as switch identification code, outgoing port timestamp and forwarding delay, and obtain the private interface protocol field in the first neighboring node as supporting the acquisition of the current outgoing port packet rate;

[0084] The standard interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node are intersected to obtain the standard interface protocol fields of the telemetry node. The standard interface protocol fields of the telemetry node include the switch identification code and the forwarding delay.

[0085] Take the intersection of the private interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node. If the intersection result is empty, configure the private interface protocol field of the telemetry node to be empty.

[0086] like Figure 4 As shown, the network telemetry node configuration method provided in this embodiment of the invention specifically includes: receiving a port start event, the port sending a probe message to a neighboring node, waiting for a probe message response from the neighboring node, if the neighboring node receives the message, adding the port to the termination port list, extracting the telemetry information carried fields from the probe message, determining whether all neighbors have received the probe message, if not, configuring the switch role as an endpoint switch and configuring the termination port list, if so, configuring the switch role as a trunk switch, taking the intersection of the telemetry information carried fields of all neighboring nodes, and configuring the fields required to be carried by the telemetry node.

[0087] Based on any of the above embodiments, the network telemetry node configuration method further includes: adding the network telemetry node configuration result to the telemetry configuration database, wherein the telemetry configuration database includes switch role data, a list of termination ports, and fields required to be carried by the telemetry information.

[0088] The network telemetry node configuration method provided in this invention adaptively generates the required configuration in the DTEL interface by carrying DTEL capacity information in the probe messages exchanged between switches. This includes the switch role, sink_port_list, and the information carried in the telemetry messages. This can significantly reduce the controller burden, especially when the topology changes. The DTEL configuration on the switch can be adaptively changed without controller intervention. This enables the INT telemetry scheme of the switch to achieve telemetry configuration without the controller's deep involvement in the switch's telemetry configuration, thus meeting user needs.

[0089] The network telemetry node configuration device provided by the present invention is described below. The network telemetry node configuration device described below and the network telemetry node configuration method described above can be referred to in correspondence.

[0090] Figure 5This is a functional structure diagram of the network telemetry node configuration device provided in an embodiment of the present invention, as shown below. Figure 5 As shown, the network telemetry node configuration device provided in this embodiment of the invention includes:

[0091] The sending module 501 is used to send a first probe message to all adjacent nodes of the telemetry node through the telemetry node port in response to the telemetry node port status open event;

[0092] The first configuration module 502 is used to monitor the response information of the neighboring nodes to the first probe message, and configure the role of the telemetry node according to the response information of the neighboring nodes to the first probe message.

[0093] The second configuration module 503 is used to receive the second probe message sent by the neighboring nodes of the telemetry node, parse the telemetry information content carried in the second probe message of each neighboring node, and configure the telemetry information content of the telemetry node according to the parsing result.

[0094] In this embodiment of the invention, the first configuration module 502 is configured as follows:

[0095] If all ports of the telemetry node receive the response information of the first probe message fed back by the neighboring node, then the telemetry node is configured as a relay node.

[0096] If some ports of the telemetry node receive response information of the first probe message fed back by the neighboring node, or if no port of the telemetry node receives response information of the first probe message fed back by the neighboring node, then the role of the telemetry node is configured as an endpoint node.

[0097] In this embodiment of the invention, the first probe message or the second probe message includes:

[0098] Ether type value, source code, object code, and payload;

[0099] The Ethernet type value is different from the Ethernet type value of the protocol used for communication between nodes;

[0100] The source code is used to identify the nodes that send probe messages;

[0101] The target code is used to identify the node receiving the probe message;

[0102] The payload includes type, length, and parameter values. The type includes whether it supports data plane telemetry function, standard interface protocol, and private interface protocol.

[0103] The standard interface protocol includes a standard protocol type code, a standard protocol type description, and standard protocol fields. The standard protocol type code is used to identify the protocol interface; the standard protocol type description is used to describe the specific protocol name corresponding to each standard protocol type code; the standard protocol fields include at least one of the following: switch identification code, inbound / outbound switch port, inbound port timestamp, outbound port timestamp, queue information, forwarding delay, outbound port utilization, and queue buffer utilization.

[0104] The private interface protocol includes a private protocol type code, a private protocol type description, and private protocol fields. The private protocol type code is used to privately identify the protocol interface; the private protocol type description is used to describe the specific protocol name corresponding to each private protocol type code; the private protocol fields include support for obtaining current queue packet loss statistics and / or support for obtaining current outgoing port packet transmission rate.

[0105] The private protocol field is defined according to the requirements of the node device manufacturer. It uses strings to describe the parameter names in the protocol, and multiple parameters are separated by commas.

[0106] The telemetry information in the payload is uploaded to the central processing unit (CPU) via an access control list, so that the CPU can parse and obtain the telemetry information carried by each node.

[0107] In this embodiment of the invention, the second configuration module 503 is configured as follows:

[0108] Parse the telemetry information sent by each neighboring node and obtain the protocol fields in the standard interface protocol and the protocol fields in the private interface protocol of each neighboring node;

[0109] The protocol fields in the standard interface protocol shared by each adjacent node are used as the standard interface protocol content in the telemetry information content of the telemetry node, and the protocol fields in the private interface protocol shared by each adjacent node are used as the private interface protocol content in the telemetry information content of the telemetry node.

[0110] If the neighboring nodes of the telemetry node are the first neighboring node, the second neighboring node, and the third neighboring node, the second configuration module 503 is further configured as follows:

[0111] Parse the telemetry information sent by the first adjacent node, and obtain the standard interface protocol fields in the first adjacent node as switch identification code, ingress port timestamp, egress port timestamp and forwarding delay, and obtain the private interface protocol fields in the first adjacent node as supporting the acquisition of current queue packet loss statistics and supporting the acquisition of current egress port packet transmission rate;

[0112] Parse the telemetry information sent by the second adjacent node, and obtain the standard interface protocol fields in the second adjacent node as switch identification code, ingress port timestamp, queue information and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of current queue packet loss statistics;

[0113] Parse the telemetry information sent by the third adjacent node, and obtain the standard interface protocol fields in the third adjacent node as switch identification code, outgoing port timestamp and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of the current outgoing port packet rate;

[0114] The standard interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node are intersected to obtain the standard interface protocol fields in the telemetry information content of the telemetry node. The standard interface protocol fields in the telemetry information content of the telemetry node include the switch identification code and the forwarding delay.

[0115] The intersection of the standard interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node is taken, and the intersection result is empty. The private interface protocol field of the telemetry information content of the telemetry node is configured to be empty.

[0116] According to a network telemetry node configuration method provided by the present invention, the method further includes: adding the network telemetry node configuration result to a telemetry configuration database, wherein the telemetry configuration database includes switch role data, a list of termination ports, and fields required to be carried by telemetry information.

[0117] The network telemetry node configuration device provided in this embodiment of the invention sends a first probe message to all neighboring nodes of the telemetry node in response to the opening of the telemetry node port; monitors the response information of neighboring nodes to the first probe message, configures the role of the telemetry node according to the response information of neighboring nodes to the first probe message; receives a second probe message sent by the neighboring nodes of the telemetry node, parses the telemetry information content carried in the second probe message of each neighboring node, and configures the telemetry information content of the telemetry node according to the parsing result. By exchanging probe messages between nodes, the nodes can adaptively generate telemetry function configurations, which can significantly reduce the controller burden. Especially when the network topology changes, the telemetry function configuration on the switch node can be adaptively changed without controller intervention, reducing configuration latency and improving network stability.

[0118] Figure 6 An example is a schematic diagram of the physical structure of a switch, such as... Figure 6As shown, the electronic device may include a processor 610, a communications interface 620, a memory 630, and a communication bus 640. The processor 610, communications interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute a network telemetry node configuration method. This method includes: in response to the telemetry node port being open, sending a first probe message to all neighboring nodes of the telemetry node; monitoring the response information of neighboring nodes to the first probe message, configuring the role of the telemetry node based on the response information of neighboring nodes to the first probe message; receiving second probe messages sent by neighboring nodes of the telemetry node, parsing the telemetry information content carried in the second probe message of each neighboring node, and configuring the telemetry information content of the telemetry node based on the parsing result.

[0119] In embodiments of the present invention, such as Figure 7 As shown, the network architecture of this switch includes: a telemetry configuration adaptive module, a telemetry configuration database, a port driver, and a switch chip;

[0120] The telemetry configuration adaptive module includes a probe packet sending program, a packet receiving analysis program, and an ACL configuration initialization program. The probe packet sending program monitors port up events; once a port is detected as up, it sends probe packets to neighboring machines. The packet receiving analysis program receives probe packets, analyzes them, determines the local telemetry configuration, and updates it to the telemetry configuration database. The ACL configuration initialization program is used to insert ACLs when the adaptive module is enabled, allowing the switch chip to send probe packets to the CPU; these ACLs are removed when the adaptive module is deenabled. The telemetry configuration database contains telemetry-related configuration data. The most important configurations include the switch role (endpoint switch or trunk switch) and, when acting as an endpoint switch, the list of ports that must perform sink actions (sink_port_list). The telemetry information must include fields such as inbound / outbound timestamps, forwarding latency, outbound port utilization, and queue buffer utilization.

[0121] The interaction sequence diagrams of each module are as follows: Figure 8 As shown, the ACL configuration initialization program distributes the ACL configuration, reports the Ethertype=9156 packet to the CPU of the switch chip, the switch chip port starts, the port driver triggers the probe packet sending program, the probe packet sending program sends probe packets to the port driver, the port driver sends probe packets to neighboring machines, the packet receiving and analysis program receives the Ethertype=9156 packet, and writes the telemetry configuration information into the telemetry configuration database based on the analysis results.

[0122] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a network telemetry node configuration method provided by the above methods. The method includes: in response to the telemetry node port being open, sending a first probe message to all neighboring nodes of the telemetry node; monitoring the response information of neighboring nodes to the first probe message, configuring the role of the telemetry node according to the response information of neighboring nodes to the first probe message; receiving a second probe message sent by the neighboring nodes of the telemetry node, parsing the telemetry information content carried in the second probe message of each neighboring node, and configuring the telemetry information content of the telemetry node according to the parsing result.

[0123] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0124] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0125] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for configuring network telemetry nodes, characterized in that, include: In response to a telemetry node port status open event, a first probe message is sent to the neighboring nodes of the telemetry node through the telemetry node port; Monitor the response information of the neighboring nodes to the first probe message, and configure the role of the telemetry node according to the response information of the neighboring nodes to the first probe message; The telemetry node receives second probe messages sent by neighboring nodes, parses the telemetry information content carried in the second probe messages of each neighboring node, and configures the telemetry information content of the telemetry node according to the parsing results. After configuring the telemetry node role as an endpoint node, the following is also included: Configure the port in the endpoint node that responds to probe messages sent by other nodes as a termination port, and add the termination port to the termination port list.

2. The network telemetry node configuration method according to claim 1, characterized in that, The step of configuring the role of the telemetry node based on the response information of the neighboring nodes to the first probe message includes: If all ports of the telemetry node receive the response information of the first probe message fed back by the neighboring node, then the telemetry node is configured as a relay node. If some ports of the telemetry node receive response information of the first probe message fed back by the neighboring node, or if no port of the telemetry node receives response information of the first probe message fed back by the neighboring node, then the role of the telemetry node is configured as an endpoint node.

3. The network telemetry node configuration method according to claim 1, characterized in that, The first probe message or the second probe message includes: Ether type value, source code, object code, and payload; The Ethernet type value is different from the Ethernet type value of the protocol used for communication between nodes; The source code is used to identify the nodes that send probe messages; The target code is used to identify the node receiving the probe message; The payload includes type, length, and parameter values. The type includes whether it supports data plane telemetry function, standard interface protocol, and private interface protocol. The standard interface protocol includes a standard protocol type code, a standard protocol type description, and standard protocol fields. The standard protocol type code is used to identify the protocol interface; the standard protocol type description is used to describe the specific protocol name corresponding to each standard protocol type code; the standard protocol fields include at least one of the following: switch identification code, inbound / outbound switch port, inbound port timestamp, outbound port timestamp, queue information, forwarding delay, outbound port utilization, and queue buffer utilization. The private interface protocol includes a private protocol type code, a private protocol type description, and private protocol fields. The private protocol type code is used to identify the private protocol interface; the private protocol type description is used to describe the specific protocol name corresponding to each private protocol type code; the private protocol fields include support for obtaining current queue packet loss statistics and support for obtaining current outgoing port packet transmission rate. The private protocol fields are customized according to the requirements of the node device manufacturers. The field names are described using strings, and multiple fields are separated by commas. The telemetry information in the payload is uploaded to the central processing unit (CPU) via an access control list, so that the CPU can parse and obtain the telemetry information carried by each node.

4. The network telemetry node configuration method according to claim 3, characterized in that, The step of configuring the telemetry information content of the telemetry node according to the parsing results includes: Parse the telemetry information sent by each neighboring node and obtain the protocol fields in the standard interface protocol and the protocol fields in the private interface protocol of each neighboring node; The protocol fields in the standard interface protocol shared by each adjacent node are used as the standard interface protocol fields in the telemetry information content of the telemetry node, and the protocol fields in the private interface protocol shared by each adjacent node are used as the private interface protocol fields in the telemetry information content of the telemetry node.

5. The network telemetry node configuration method according to claim 4, characterized in that, If the neighboring nodes of the telemetry node are a first neighboring node, a second neighboring node, and a third neighboring node, configuring the telemetry information content of the telemetry node according to the parsing result includes: Parse the telemetry information sent by the first adjacent node, and obtain the standard interface protocol fields in the first adjacent node as switch identification code, ingress port timestamp, egress port timestamp and forwarding delay, and obtain the private interface protocol fields in the first adjacent node as supporting the acquisition of current queue packet loss statistics and supporting the acquisition of current egress port packet transmission rate; Parse the telemetry information sent by the second adjacent node, and obtain the standard interface protocol fields in the second adjacent node as switch identification code, ingress port timestamp, queue information and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of current queue packet loss statistics; Parse the telemetry information sent by the third adjacent node, and obtain the standard interface protocol fields in the third adjacent node as switch identification code, outgoing port timestamp and forwarding delay, and obtain the private interface protocol field in the first adjacent node as supporting the acquisition of the current outgoing port packet rate; The standard interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node are intersected to obtain the standard interface protocol fields of the telemetry node. The standard interface protocol fields of the telemetry node include the switch identification code and the forwarding delay. Take the intersection of the private interface protocol fields of the first adjacent node, the second adjacent node, and the third adjacent node. If the intersection result is empty, configure the private interface protocol field of the telemetry node to be empty.

6. The network telemetry node configuration method according to claim 1, characterized in that, Also includes: The network telemetry node configuration results are added to the telemetry configuration database, which includes switch role data, a list of termination ports, and fields required for telemetry information.

7. A network telemetry node configuration device, applicable to the network telemetry node configuration method according to any one of claims 1 to 6, characterized in that, include: The sending module is used to send a first probe message to the neighboring nodes of the telemetry node through the telemetry node port in response to the telemetry node port status open event. The first configuration module is used to monitor the response information of the neighboring nodes to the first probe message, and configure the role of the telemetry node according to the response information of the neighboring nodes to the first probe message. The second configuration module is used to receive the second probe messages sent by the neighboring nodes of the telemetry node, parse the telemetry information content carried in the second probe messages of each neighboring node, and configure the telemetry information content of the telemetry node according to the parsing results.

8. A switch, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the network telemetry node configuration method as described in any one of claims 1 to 6.

9. A non-transitory readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the network telemetry node configuration method as described in any one of claims 1 to 6.

Citation Information

Patent Citations

  • Automatic clock phase synchronization in OTN multi-chassis system with fail over mechanism

    CN111630794A

  • Intention-based telemetry collection service

    CN114185637A