Detection system and method for network devices

Abstract

A detection system and method of a network device includes a PXE host, a switch and a plurality of server clusters. The PXE host has static data and an OS image file, the OS image file has a device reading program; the server clusters are connected to the switch, each server cluster has a plurality of servers, the servers have a BMC unit and a network unit, the network unit of each server is connected to a port of the switch; the servers acquire the OS image file from the PXE host, the servers execute the device reading program, the device reading program acquires device information of the network unit and port information of the switch, the device reading program encapsulates the device information and the port information as dynamic data, the device reading program sends the dynamic data to the PXE host, and the PXE host generates an exception detection report according to the static data and the dynamic data.

Description

Technical Field

[0001] A testing system and method for an electronic device, and more particularly a testing system and method for a network device. Background Technology

[0002] With the rapid development of the Internet, data centers have been established to transmit massive amounts of information. Data centers require numerous servers and related network devices, such as switches and network interface cards. Since the wiring of network devices requires manual verification of the connections at both ends, manually testing the wiring between servers and switches is a heavy burden, given the presence of multiple servers and network devices in a data center. Incorrect wiring configurations can also lead to reduced transmission speeds. Summary of the Invention

[0003] In view of this, in one embodiment, a network device detection system is provided. The server device detection system includes a PXE host, a switch, and multiple server clusters. The PXE host has static data and an OS image file, the OS image file containing a device reading program; the switch is connected to the PXE host and has multiple ports, through which the switch transmits the device reading program; the server clusters are connected to the switch, each server cluster has multiple servers, each server has a BMC unit and multiple network units, and each server's network unit is connected to a corresponding port; wherein, the server obtains the OS image file from the PXE host, the server executes the device reading program, the device reading program obtains the device information of the network units and the port information of the switch, the device reading program encapsulates the device information and port information into dynamic data, the device reading program sends the dynamic data to the PXE host, and the PXE host generates an anomaly detection report based on the static data and dynamic data.

[0004] In one embodiment, static data records the preset positions of each server. The PXE host compares each preset position with the corresponding dynamic data to see if they are consistent. If any preset position is inconsistent with the corresponding dynamic data, the PXE host generates an anomaly detection report.

[0005] In one embodiment, the preset locations include the MAC address of the LAN management function of the baseboard management controller, the MAC address of the PXE host, the MAC address of the switch, the computer room number, the cabinet number, the cabinet layer number, the network card name, the port number of the switch, and the BlueField interface card number or the ConnectX interface card number.

[0006] In one embodiment, after the PXE host acquires dynamic data, it performs data cleaning processing on the dynamic data to align the format of the dynamic data with that of the static data.

[0007] In one embodiment, the server is powered on, and the server obtains the OS image file from the PXE host and executes the device reading program.

[0008] In one embodiment, a network device detection system includes an external host, a PXE host, a switch, and multiple server clusters. The external host has static data and a detection program; the PXE host is connected to the external host and has an OS image file containing a device reading program; the switch is connected to the PXE host and has multiple ports, through which the device reading program is transmitted; the server clusters are connected to the switch, each server cluster having multiple servers, each server having a BMC unit and multiple network units; wherein, any server obtains the OS image file from the PXE host, the server executes the device reading program, the device reading program obtains device information of the network units and port information of the switch, the device reading program encapsulates the device information and port information into dynamic data, the device reading program sends the dynamic data to the PXE host, the PXE host forwards the dynamic data to the external host, and the detection program generates an anomaly detection report based on the dynamic data and static data.

[0009] In one embodiment, a method for detecting a server device, which detects the network connection status of any server in a server cluster, includes the following steps: the server obtains an OS image file from a PXE host, the OS image file containing a device reading program; when the server loads the OS image file, it executes the device reading program; the device reading program obtains at least one device information of the server and connection port information of the switch; the device reading program encapsulates the device information and connection port information into dynamic data; the server sends the dynamic data to the PXE host; and the PXE host generates an anomaly detection report based on the stored static data and dynamic data.

[0010] In one embodiment, the step of the server obtaining the OS image file from the PXE host includes the server being powered on and obtaining the OS image file from the PXE host before the server obtains the OS image file from the PXE host.

[0011] In one embodiment, after the step of the server obtaining the OS image file from the PXE host, the PXE host obtains dynamic data and performs data cleaning processing on the dynamic data to align the format of the dynamic data with the format of the static data.

[0012] In one embodiment, the step of the PXE host generating an anomaly detection report based on the stored static and dynamic data includes: recording the preset positions of each server in the static data; comparing each preset position with the corresponding dynamic data to see if they are consistent; if any preset position is inconsistent with the corresponding dynamic data, the PXE host generates an anomaly detection report; if the preset position is consistent with the corresponding dynamic data, the PXE host generates a normal detection report.

[0013] The network equipment testing system and methods enable maintenance personnel to quickly check whether each server is set up in the corresponding computer room, rack, and bay according to the predetermined plan, and can also check whether the connection wiring between servers and switches is incorrect. Maintenance personnel can determine whether servers are misplaced by comparing dynamic and static data, without having to cross-check each server or switch individually. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of a detection system for a servo device according to an embodiment.

[0015] Figure 2 This is a schematic diagram of a server and a PXE host according to one embodiment.

[0016] Figure 3 This is a schematic diagram of the detection process of a servo device according to one embodiment.

[0017] Figure 4 This is a schematic diagram of static data before cleaning in one embodiment.

[0018] Figure 5 This is a schematic diagram of static data after cleaning, as shown in one embodiment.

[0019] Figure 6 This is a schematic diagram illustrating the comparison of dynamic and static data in one embodiment.

[0020] Figure 7 This is a schematic diagram of a detection system for a servo device according to another embodiment.

[0021] The reference numerals in the attached figures are explained as follows:

[0022] 10: Detection System

[0023] 100: Servo Cluster

[0024] 110: Server

[0025] 111: First Network Unit

[0026] 112: First storage unit

[0027] 113: BMC Unit

[0028] 114: First Processing Unit

[0029] 115: Download Requirements

[0030] 200: PXE Host

[0031] 210: Second Network Unit

[0032] 220: Second storage unit

[0033] 230: Second Processing Unit

[0034] 400: Switch

[0035] 410: Connection Port

[0036] 510: OS Image File

[0037] 520: Device Reading Program

[0038] 521: Static Data

[0039] 522: Dynamic Data

[0040] 530: Anomaly Detection Report

[0041] 540: Operating System

[0042] 700: External Host

[0043] 710: Detection Procedure

[0044] S310, S320, S330, S340, S350, S360: Steps Detailed Implementation

[0045] Please refer to Figure 1 and Figure 2 The diagrams shown illustrate a server device detection system 10 according to one embodiment and a server 110 and PXE host 200 according to another embodiment. The server device detection system 10 (hereinafter referred to as detection system 10) is applied in a data center or other cabling environment with multiple computer devices. Detection system 10 includes a switch 400, a preboot execution environment (PXE host 200), and at least one server cluster 100. The switch 400 has multiple ports 410, with different ports 410 connected to either the PXE host 200 or the server cluster 100.

[0046] The server cluster 100 consists of multiple servers 110. Each server 110 has a first network unit 111, a first storage unit 112, a Baseboard Management Controller (BMC unit 113), and a first processing unit 114. The first processing unit 114 is connected to the first network unit 111 and the first storage unit 112. The first network unit 111 is connected to the BMC unit 113 and the switch 400. In other words, each first network unit 111 is connected to a designated port 410. The first storage unit 112 stores the operating system 540 and dynamic data 522 (the acquisition method will be detailed later). The servers 110 of the same server cluster 100 are interconnected by signals. Therefore, different servers 110 of the same server cluster 100 can reside in different server rooms or racks.

[0047] In addition to monitoring the operational status of the server 110 internally, the BMC unit 113 can also provide remote start, stop, or restart services for the server 110. Each BMC controller can perform remote management functions such as event log management, firmware or system updates through its associated BMC management network function (Baseboard Management Controller Management LAN, abbreviated as mLAN) (unlabeled).

[0048] The PXE host 200 can be, but is not limited to, the server 110, or a personal computer, an embedded system, or other electronic device with network communication capabilities. The PXE host 200 has at least one second network unit 210, a second storage unit 220, and a second processing unit 230. The second processing unit 230 is connected to the second network unit 210 and the second storage unit 220. The second storage unit 220 contains static data 521 and an operating system image file (OS image file 510). The OS image file 510 further encapsulates a device reader 520. The device reader 520 is used to read device information of specific hardware in the server 110 and port information of the switch 400.

[0049] Static data 521 records various preset information for each server 110. The preset information includes the Media Access Control Address (MAC) for the LAN management function, the MAC address of the PXE host 200, the MAC address of the switch 400, the server room number, the Stock Keeping Unit (SKU), the cabinet number, the number of cabinet layers, the network card name, the network card port number, the port number of the switch 400, and the port number of the BlueField interface card or the ConnectX interface card.

[0050] The second network unit 210 can transmit the OS image file 510, or receive download requests 115, dynamic data 522, and other data. After receiving the download request 115, the second processing unit 230 sends the OS image file 510 to the corresponding server 110 through the second network unit 210. For further explanation of the operation flow of the detection system 10, please refer to [link to relevant documentation]. Figure 3 The detection method for servo devices includes the following steps:

[0051] Step S310: The server obtains the OS image file from the PXE host, which contains a device reading program;

[0052] Step S320: When the server loads the OS image file, the device reading program is executed;

[0053] Step S330: The device reading program obtains at least one device information of the server and the connection port information of the switch;

[0054] Step S340: The device reading program encapsulates the device information and connection port information into dynamic data;

[0055] Step S350: The server sends dynamic data to the PXE host; and

[0056] Step S360: The PXE host generates an anomaly detection report based on the stored static and dynamic data.

[0057] First, any server 110 in server cluster 100 boots up. During the boot process, the LAN management function to which server 110 belongs sends a download request 115 to PXE host 200 (corresponding to step S310) to obtain the OS image file 510 so that server 110 can load the corresponding operating system 540. After obtaining the OS image file 510, server 110 can extract the operating system 540 and device reader 520 from the OS image file 510 (corresponding to step S320). Taking a Uniplexed Information and Computing Service (UNIX) operating system as an example, during the boot process, the UNIX operating system can insert the execution command of device reader 520 into the bootloader (e.g., GRUB configuration file). Alternatively, the UNIX operating system can execute the script file of device reader 520 through the bash command processor.

[0058] Next, the device reading program 520 begins to acquire at least one device information of the server 110 and the connection port information of the switch 400 (corresponding to step S330). The device information includes the media access control address for the LAN management function and the BlueField interface card number or ConnectX interface card number. The connection port information includes the MAC address of the switch 400 and the connection port number of the switch 400. The device reading program 520 encapsulates the device information and connection port information into dynamic data 522 (corresponding to step S340).

[0059] Server 110 sends dynamic data 522 to PXE host 200 (corresponding to step S350). PXE host 200 determines whether the content of dynamic data 522 is the same as that of static data 521. If the content of dynamic data 522 is the same as that of static data 521, it indicates that the network cabling of server 110 and the storage location of the server rack are correct. If the content of dynamic data 522 is different from that of static data 521, PXE host 200 generates an anomaly detection report 530 based on the static data 521 and dynamic data 522 (corresponding to step S360). In some embodiments, PXE host 200 can list the differences between static data 521 and dynamic data 522 and generate anomaly detection report 530.

[0060] In some embodiments, after acquiring dynamic data 522, the PXE host 200 performs data cleaning processing on the dynamic data 522 to align the format of the dynamic data 522 with that of the static data 521. The static data 521 is preset information of the server 110, therefore its format remains unchanged. The PXE host 200 may be affected by program reading specifications or the storage format of specified devices, causing the read device information and port information (corresponding to dynamic data 522) to differ in format from the static data 521. Furthermore, the PXE host 200 may select at least one set of data from the various sets of static data 521. The PXE host 200 performs data cleaning processing on the dynamic data 522 based on the selected static data 521.

[0061] Please refer to Figure 4 , Figure 5 and Figure 6 These are schematic diagrams of static data 521 before cleaning in one embodiment, static data 521 after cleaning in one embodiment, and a schematic diagram comparing dynamic data 522 with static data 521 in one embodiment. Figure 4 and Figure 5 The static data 521 is for illustrative purposes only and is not limited to the data or format shown. Furthermore, both static data 521 and dynamic data 522 can undergo the same cleaning process to obtain data content in the same format. Figure 4 For example, static data 521 lists partial information about any server 110 and its connected switch 400. Figure 5 A lookup table showing the actual location of each server 110 and the corresponding port 410 of the switch 400.

[0062] First, the PXE host 200 will store the static data 521 corresponding to the servers 110 of each server cluster 100, such as... Figure 4 As shown. Static data 521 includes the location of each server 110 (corresponding to the computer room or cabinet), product model, machine serial number, MAC address of BMC unit 113, MAC address corresponding to the port number of BlueField interface card or MAC address corresponding to the port number of ConnectX interface card.

[0063] After any server 110 acquires the OS image file 510, the server 110 will retrieve the device reader 520 from the OS image file 510. The server 110 executes the device reader 520. The device reader 520 begins acquiring at least one device information from the server 110 and the connection port information of the switch 400. Assuming the device reader 520 acquires... Figure 5The device information and port information of the switch 400 are retrieved. The device reading program 520 will send the acquired unwashed static data 521 to the PXE host 200.

[0064] The PXE host 200 can clean up uncleaned static data 521. This cleaning process may include removing symbols or text (numbers) from specific locations. For example, cleaning the MAC address of the connection port 410 between the server 110 and the switch 400 will be performed. Figure 5 The left side shows the static data 521 before cleaning, and the right side shows the static data 521 after cleaning. Taking the data with item number "0" as an example, item number "0" includes values ​​such as "A1-A-01-03, IB2, A1-A-02-41, 01-1". The PXE host 200 can read the text at a specific location using a mask and write the read text sequentially into the corresponding format. Figure 5 After cleaning, the term "A1-A-01-03,IB2" with the left-hand side number "0" will yield... Figure 5 The right side shows "A, 1, 3, IB2" for server 110. After the aforementioned processing, the corresponding data for switch 400, "A, 2, 41, 1, 1", can be obtained. In short, Figure 5 The various device information can be matched with Figure 4 Any column.

[0065] After PXE host 200 obtains the MAC address "aa:bb:cc:dd:ee:ff", it can remove specific characters such as ":" or the symbol "0x" using a regular expression. For other devices, specific characters in their corresponding information can also be removed in this way. For example, the encoded string starting with "swp" for the connection port of switch 400, or the encoded string starting with "0x" for the globally unique identifier (GUID) of switch 400, can also be processed using regular expressions. The resulting MAC address after removal is "aabbccddeeff". At this point, PXE host 200 stores the cleaned data in the second storage unit 220, and the stored data is static data 521.

[0066] Next, the PXE host 200 acquires dynamic data 522 from each server 110. The PXE host 200 performs the aforementioned cleaning process on the dynamic data 522. Finally, it obtains... Figure 6The static data 521 is compared with the cleaned dynamic data 522. The PXE host 200 compares the static data 521 with the dynamic data 522. The PXE host 200 compares the data based on the MAC address of the network card or the MAC address of the switch 400. Figure 6 In the original data, the network card address "74563cb1627a" was connected to port number "2" of switch 400 (corresponding to static data 521). However, the dynamic data 522 records that the network card address "74563cb1627a" was connected to port number "3" of switch 400. Figure 6 The differences detected are indicated by gray blocks, but adjustments can be made according to the format of the detection report. Therefore, the PXE host 200 will determine that the network card is installed on the wrong port 410 of the switch 400. The PXE host 200 generates an anomaly detection report 530. If the dynamic data 522 matches the static data 521, the PXE host 200 generates a normal detection report (without a label).

[0067] In some embodiments, the detection system 10 includes an external host 700, a PXE host 200, a switch 400, and multiple server clusters 100. Please refer to [reference needed]. Figure 7 Switch 400 is network-connected to external host 700, PXE host 200, and server cluster 100. Server 110 has a first network unit 111, a first storage unit 112, and a first processing unit 114. The first storage unit 112 has an operating system 540 and dynamic data 522. PXE host 200 has a second network unit 210, a second storage unit 220, and a second processing unit 230. The second processing unit 230 has an OS image file 510. External host 700 has static data 521 and a detection program 710.

[0068] The acquisition of dynamic information by the server 110 and device reader 520 can be referred to the foregoing operational description. In this embodiment, after receiving dynamic data 522 sent by the server 110, the PXE host 200 can forward the dynamic data 522 to the external host 700. Alternatively, the PXE host 200 can perform data cleaning on the dynamic data 522 before sending it to the external host 700. The detection program 710 of the external host 700 compares the dynamic data 522 with the static data 521. The comparison process of dynamic data 522 and static data 521 can be referred to... Figure 4 , Figure 5 , Figure 6 Related instructions. The detection program 710 sends the comparison results (also known as the anomaly detection report 530 or the normal detection report) to the PXE host 200 or other designated computer device.

[0069] The network equipment detection system 10 and method provide maintenance personnel with the ability to quickly detect whether each server 110 is set up in the corresponding computer room, cabinet, and rack according to the predetermined plan, and can also detect whether the connection wiring between the server 110 and the switch 400 is incorrect. Maintenance personnel can determine whether the server 110 is misplaced by comparing the results of dynamic data 522 and static data 521, without the need for cross-checking of each server 110 or switch 400.

Claims

1. A detection system for a servo device, characterized in that, include: A PXE host has a static data and an OS image file, which has a device read program. A switch, connected to the PXE host, has multiple ports through which the device read program is transmitted; as well as Multiple server clusters are connected to the switch. Each server cluster has multiple servers. Each server has a BMC unit and multiple network units. The network unit of each server is connected to a corresponding port. Specifically, the server obtains the OS image file from the PXE host, executes the device reading program, obtains device information of the plurality of network units and connection port information of the switch, encapsulates the device information and connection port information into dynamic data, and sends the dynamic data to the PXE host. The PXE host generates an anomaly detection report based on the static data and the dynamic data.

2. The servo device detection system as described in claim 1, characterized in that, The static data records a preset position for each server. The PXE host compares each preset position with the corresponding dynamic data to see if they are consistent. If any preset position is inconsistent with the corresponding dynamic data, the PXE host generates an anomaly detection report.

3. The servo device detection system as described in claim 1, characterized in that, The preset location includes the MAC address of the LAN management function of the baseboard management controller, the MAC address of the PXE host, the MAC address of the switch, the room number, the rack number, the rack layer number, the network card name, the port number of the switch, and the BlueField interface card number or the ConnectX interface card number.

4. The servo device detection system as described in claim 1, characterized in that, After the PXE host obtains the dynamic data, it performs data cleaning processing on the dynamic data to align the format of the dynamic data with that of the static data.

5. The servo device detection system as described in claim 1, characterized in that, When the server is powered on, it obtains the OS image file from the PXE host and executes the device read program.

6. A network device detection system, characterized in that, include: An external host has static data and a detection program; A PXE host is connected to the external host. The PXE host has an OS image file, and the OS image file has a device reader. A switch, connected to the PXE host, has multiple ports through which the device read program is transmitted; as well as Multiple server clusters are connected to the switch, each server cluster has multiple servers, and each server has a BMC unit and multiple network units. In this process, any one of the servers obtains the OS image file from the PXE host, the server executes the device reading program, the device reading program obtains device information of the plurality of network units and connection port information of the switch, the device reading program encapsulates the device information and the connection port information into dynamic data, the device reading program sends the dynamic data to the PXE host, the PXE host forwards the dynamic data to the external host, and the detection program generates an anomaly detection report based on the dynamic data and the static data.

7. A method for detecting a servo device, characterized in that, Detect the network connectivity status of any server in a server cluster, including: The server obtains an OS image file from a PXE host, which contains a device reader. When the server loads the OS image file, it executes the device read procedure; The device read program obtains at least one device information of the server and a port information of a switch; The device reading program encapsulates the device information and the port information into dynamic data; The server sends the dynamic data to the PXE host; and The PXE host generates an anomaly detection report based on stored static data and dynamic data.

8. The method for detecting a servo device as described in claim 7, characterized in that, The step preceding "the server obtains the OS image file from the PXE host" includes: When the server is powered on, it obtains the OS image file from the PXE host.

9. The method for detecting a servo device as described in claim 7, characterized in that, The step following "The server obtains the OS image file from the PXE host" includes: After the PXE host obtains the dynamic data, it performs data cleaning processing on the dynamic data to align the format of the dynamic data with that of the static data.

10. The method for detecting a servo device as described in claim 7, characterized in that, The steps in "the PXE host generates the anomaly detection report based on the stored static data and dynamic data" include: This static data record represents a preset position for each server. The PXE host compares each preset location with the corresponding dynamic data to see if they are consistent. If any of the preset locations is inconsistent with the corresponding dynamic data, the PXE host generates the anomaly detection report; and If the preset location matches the corresponding dynamic data, the PXE host generates a normal detection report.

Images