KR20190118064A - Server Validation Automation and Management System - Google Patents

Abstract

The present invention relates to a system for automating and managing server verification, which is capable of conducting a test by using a test package according to constructing device by target server, and determining a defect from the test results. To be more specifically, the system for automating and managing server verification comprises: a data storage for grouping test programs classified according to the purpose of test by constructing device, storing the same as a test package and, in particular, storing the same through classification by constructing device information; a plurality of target servers for transmitting the constructing device information at booting, conducting a test when receiving the test package, and transmitting the test results; and a management server for extracting the test package to be conducted by the relevant target server from schedule information including a test schedule when receiving the constructing device information, creating a work flow by extracting the test package matched with the received constructing device information, and controlling the plurality of target servers to conduct a test according the created work flow.

Description

Server Validation Automation and Management System

The present invention relates to a server verification automation and management system that can test using a test package according to the configuration information of each target server and determine whether there is a defect.

More specifically, the test program classified according to the test purpose grouped by the configuration device and stored as a test package, the data storage to classify and save by the configuration device information, transmits the configuration device information during booting, the test when receiving the test package When receiving a plurality of target server and the configuration information to transmit the test results, extracting the test package to be performed by the target server from the schedule information including the test schedule, the test package matching the received configuration information The present invention relates to a server verification automation and management system comprising a management server that extracts and generates a workflow and controls the plurality of target servers to perform a test according to the generated workflow.

With the introduction of cloud services, the demand for large-scale cloud data centers is increasing, and the demand for high-density server platforms and high-energy efficient servers based on easily expandable x86 boards is increasing.

However, the localization rate of the computer industry is 28.3% per year, lower than half of the 57.4% of 2000. The technology level of the server sector is also very low at 72.6% compared to the US. The input is much higher.

Recently, the x86 server market has exploded with the adoption of standard architecture, which is optimal for distributed and scale-out based cloud infrastructures and software defined platforms, and the top 5 companies (HPE, Dell, IBM, Lenovo, Cisco) It occupies 92.2%. Thus, it is necessary to secure domestic market competitiveness.

To this end, research and development on servers for data centers that are used in the cloud, big data, artificial intelligence IoT, etc. are also active in Korea. On the other hand, when the server is mass-produced, it is determined whether the product is passed through the performance test of the produced server.

In this case, performance tests for each component included in a plurality of servers (hereinafter, referred to as target servers) are required, and how to efficiently perform and manage the test becomes an important issue. Accordingly, the present applicant generates a workflow matching the test schedule for the target server when the target server is booted to perform the test using the test package matched to the target device's constituent device and determine whether there is a defect. Developed a server verification automation and management system.

1. Korean Patent Publication No. 10-2004-0054904

An object of the present invention, by generating a workflow matching the test schedule for the target server at the boot of the target server to perform the test using the test package matched to the configuration device of the target server and determine whether there is a defect as a result To provide a server automation and management system that can be done.

Server verification automation and management system according to an embodiment of the present invention for achieving the above object is to group the test program classified according to the test purpose by the configuration device to store as a test package, classified by configuration device information Data storage to store, transmits the configuration device information at boot, and when receiving a test package receives a plurality of target server and the configuration device information to perform the test and transmit the test results, the target server from the schedule information including the test schedule Extracts a test package to be performed, extracts a test package matching the received device information, generates a workflow, and controls the management server to control the plurality of target servers to perform a test according to the generated workflow. It may include.

In addition, when a boot request is received through the PreBoot Execution Environment (PXE) of the target server, the Dynamic Host Configuration Protocol (DHCP) expansion server which allocates an IP to the target server and transmits an address value of an OS image for booting, and the The server may further include a TFTP (Trivial File Transfer Protocol) extension server that transmits the OS image to the target server through a protocol supported by PXE, and transmits the test package to the target server booted using the OS image. .

The management server transmits a location value and a test setting value of a test package to be performed according to the workflow to the DHCP expansion server, and the target server transmits the test package and the test setting value according to the workflow through the TFTP expansion server. Can be received.

In addition, the configuration device information is the manufacturer and model name of the configuration device configured by the target server, the test schedule may include a test transmission time to be performed for each target server, whether or not re-execution in the event of a test error and the number of test execution times.

The management server may control the DHCP expansion server and the TFTP expansion server to reinstall the test package in which the error occurred in the target server according to the re-execution if it is determined that the received test result is an error.

In addition, when the management server receives a boot notification from the target server, the management server may control to install a tool for interworking with an Intelligent Platform Management Interface (IPMI) on the target server, and generate an IPMI account by assigning an IP for IPMI. have.

The target server may include a plurality of sensors for measuring a state of a configuration device, and the management server may be configured to display the state information measured by the plurality of sensors through an IPMI interworking environment at the request of a verification manager or at a preset interval. When the received state information is out of the preset range, the controller can notify the user.

As described above, according to the present invention, the server verification automation and management system of the present invention, by generating a workflow matching the test schedule for the target server at the time of booting the target server, configuration information of the target server ( Using the test package matched with the manufacturer, model name, etc.), the target server (x86 server mainboard) can perform the test and receive the result from the management server to determine the normal or defect.

In addition, by monitoring the status information of the target server through the IPMI linkage, it is possible to detect the abnormality of the target server.

In addition, it is possible to perform a customized test according to the configuration device for each target server according to the workflow, and by determining whether to re-execute the test through the re-execution included in the schedule information, it is possible to implement test automation.

In addition, a plurality of target servers can be tested at the same time, and OS can be installed by providing an OS image necessary for the target server in the normal determination.

In addition, it is possible to monitor and manage the setting and test history / results of a plurality of test target servers in the management server, and provide statistics according to the defects of each test package, so that the verification manager can decide whether to pass or not and adjust the test schedule. Can be.

1 is a view showing a schematic configuration of a server verification automation and management system according to an embodiment of the present invention.
2 is a block diagram showing the configuration of the management server of FIG.
3 is a view for explaining an operation of the state information monitoring unit of FIG.
4 is a flowchart illustrating a server verification and management method according to an embodiment of the present invention.

The terms or words used in this specification and claims are not to be construed as limiting in their usual or dictionary meanings, and the inventors may appropriately define the concept of terms in order to best describe their invention. It should be interpreted as meaning and concept corresponding to the technical idea of the present invention based on the principle that the present invention.

Therefore, the embodiments described in the specification and the drawings shown in the drawings are only one of the most preferred embodiments of the present invention and do not represent all of the technical idea of the present invention, various equivalents that may be substituted for them at the time of the present application It should be understood that there may be water and variations.

Before describing the present invention with reference to the drawings, it is not shown or specifically described for the matters that are not necessary to reveal the gist of the present invention, that is, those skilled in the art can obviously add. Make a note.

1 is a view showing a schematic configuration of a server verification automation and management system according to an embodiment of the present invention. 1, the server verification automation and management system 1000 according to an embodiment of the present invention is a plurality of target servers (100a to 100n), management server 200, DHCP expansion server 300, TFTP expansion server 400, the network measurement server 500, and the data storage 10.

Server verification automation and management system 1000 according to an embodiment of the present invention is a system capable of performing a remote test, test result monitoring and management of a plurality of target server (100a to 100n) in the production site of the target server Can be.

The plurality of target servers 100a to 100n may be composed of devices such as a main device, an auxiliary device, a network device, and other devices. The server for a data center used in the cloud, big data, artificial intelligence IoT, etc. Can be.

Here, as an example of each component device, the main device may be a CPU, and the auxiliary device may be a memory. In addition, the network device may be a communication device for data communication, and other devices may be devices such as power and fan.

The target server 100 of the present invention has its own unique identifier, and may be initially set to request an IP allocation for booting into a specific DHCP extension server 300 at power-on. Meanwhile, a description of the function and operation of the target server 100 will be described with reference to FIG. 4.

The management server 200 manages a plurality of target servers 100a to 100n: 100, the DHCP expansion server 300 and the TFTP expansion server 400 which are verification targets, and various information for performing the test of the target server 100. (Test package, test set value, OS image, etc.) can be managed, and defect analysis and statistical information can be provided according to test results.

At this time, the information set and managed by the management server 200 may be stored in the data storage 10.

The data storage 10 is a test package for each component of the target server 100, a test setting value for each test package, a threshold setting for determining whether there is a defect, a test schedule, a test workflow, status information for each target server, and an OS for booting. Various information can be saved according to image and test result.

In this case, the test package may be generated for each device information, and the test package for the same device may be classified and stored in a package group. At this time, the component information may be information such as the manufacturer, model name of the component.

For example, referring to FIG. 1, the first test package group (package # 1) is a grouping of a plurality of test packages (package # 1 to package # N) and classified and stored, and the first test package group is a CPU. It may be a test package group associated with, and the plurality of test packages (package # 1 to package # N) may be a test package according to the model name of each manufacturer of the CPU. For example, package # 1 may be a B-model CPU of A company.

In this case, the test package may be classified into a main device quality measurement package, an auxiliary device quality measurement package, a network quality measurement package, and other device quality measurement packages. Each test package may include at least one test suite that classifies a plurality of tests (test cases) according to a test purpose. For example, the main device quality measurement package may be described with reference to Table 1 below, and may be one of the test packages (package # 1 to package # N) of FIG. 1.

Test package Test suit Test case

Main Unit Quality Measurement Package


CPU test suite


stress test pi calculator prime number calculating Floating Point operation test CPU Extended Instruction Test Suite
SSE instruction test SSE2 instruction test

Here, the test case may be a test program. In addition, the test suite is classified by the test program according to the test purpose, the test program for performing a test for the CPU expansion instruction may be classified into a single suit, the test suite is a component of the target server 100 They can be categorized and grouped into packages.

Test cases produced by test package makers can be classified and managed by test administrators into test packages, test suites, and test cases.

Meanwhile, the test package may be automatically installed when downloaded to the target server 100 to perform a test for each test case, and transmit the test result of each test case to the management server 200.

In addition, the test set value for each test package may be set or changed by the verification manager each time the test is performed to values (per test case) that are given to perform the test. In addition, a threshold for determining whether there is a defect may be set for each test case, and may be used when determining whether there is a defect according to a test result. In addition, the OS image is produced by the test image creator and can be used at boot time.

In addition, the test schedule may be information for setting which test for which device in the plurality of grouped target servers 100 to perform in a test execution order input by the verification manager.

In addition, the test workflow may be information generated by matching a test package for executing a specific test set in the test schedule with an execution order of the test schedule. A description of the test workflow generation will be described with reference to FIG. 2.

2 is a block diagram showing the configuration of the management server of FIG. 2, the management server 200 according to an embodiment of the present invention is the setting unit 210, management unit 220, control unit 230, defect determination unit 240 and state information monitoring unit 250 It may include.

The setting unit 210 is the initial setting of the target server 100, the REST API and configuration settings of the DHCP expansion server 300 and TFTP expansion server 400, test setting values, threshold values for each test package, test schedule, etc. from the verification manager Can be set by receiving input.

In this case, the initial setting of the target server 100 may be a setting for transmitting an IP allocation request and configuration device information to the DHCP expansion server 300 which is preset at power on.

Here, the target server 100 by the setting to transmit the configuration device information, after the boot notification to the management server 200, the configuration device information (such as the manufacturer and model name for each configuration device) that is configured by the management server 200 ) Can be sent.

In addition, the DHCP expansion server 300 is configured to transmit a predetermined OS image address value to the target server 100 requesting IP allocation, and extends the TFTP upon receiving the test package location value and the test setting value from the management server 200. The test package and the test set value may be set to be transmitted to the target server 100 through the server 400.

In addition, the test schedule may include a test transmission time to be performed for each target server 100, whether or not to re-execute the corresponding test when an error occurs during the test, the number of times performed per test, and the like. In this case, the test schedule may be set by the verification manager or automatically set based on defect status statistics and verification history of the test package of each target server.

The management unit 220 may classify, store and manage the information set by the setting unit 210 in the data storage 10, and provide a test schedule to the control unit 230. In addition, the management unit 220 may provide the test history and test defect status statistics according to the request of the verification manager through the management UI by date and time, and the test package and real time defect result for each target server being performed in real time. You can also provide

The controller 230 may control the target server 100, the DHCP expansion server 300, and the TFTP expansion server 400 according to the setting of the setting unit 210. The controller 230 may control the respective servers 100, 300, 400, and 500 so that the test may be performed by turning on the target server 100 according to the test schedule.

In addition, the controller 230 may generate a test workflow by matching the test package to execute a specific test set in the test schedule with the execution order of the test schedule when receiving the device information from the booted target server 100. have.

That is, the test workflow may be a task list scheduled in the order of execution of the test schedule, the test package matching the performance test included in the test schedule and the component information received from the target server 100. For example, if the performance test is a cpu extension command test, and the cpu is a B model of the A company, the main device quality measurement package corresponding to the B model of the A company is extracted from the test package stored in the data storage 10 to generate a workflow. Can be.

The control unit 230 controls the extracted test package through the TFTP expansion server 400 by transmitting the position value and the test setting value of the test package to be executed to the DHCP extension server 300 according to the generated test workflow. By doing so, it is possible to perform a test by downloading a test package and a test setting value to be executed on the target server 100.

The defect determination unit 240 may determine whether there is a defect when comparing the test result from the target server 100 with a preset threshold. To understand the description, the network quality test can be described as an example.

In this case, the target server 100 downloads the network quality measurement test package and test setting values through the TFTP server 400, and performs the network quality measurement test through the network quality measurement server 500 for the network quality measurement test. can do.

At this time, the target server 100 may be provided with a plurality of network adapters, the network quality measurement server 500 is matched to perform a network quality measurement test for each of the plurality of network adapters of the target server 100 in parallel A plurality of network adapters can be provided.

In this case, the test set value may be test traffic for network quality measurement. The test traffic may be set by the verification manager and may be packet size, packet type, packet amount, and the like. In this case, the test may be a time measurement and upload bandwidth for each network adapter when the data according to the test traffic is uploaded / downloaded from the target server 100 to the network quality measurement server 500. In addition, the threshold may be a reference time range for uploading and downloading.

In the meantime, the network quality measurement test package has been described as an example, and a test set value and a threshold value may be set according to a test purpose for each test package of each component.

In addition, the defect determination unit 240, when receiving a test result from the target server 100, if it is determined that an error has occurred, the test can be re-run according to the re-execution information for the test included in the test schedule. The request to the re-execution to the control unit 230 may be so.

The controller 230 may receive the test package position value and the test setting value to the DHCP extension server 300 so that the test package may be reinstalled by the target server.

In this case, when the test re-execution of the corresponding test package is completed and the test result is received, the controller 230 may control the target server to perform the test package of the next order according to the workflow.

The status information monitoring unit 250 collects status information from the target server 100 through IPMI (Intelligent Platform Management Interface) interworking to monitor whether there is an abnormality and provide a notification when an abnormality occurs.

Here, the status information is information measured from a sensor provided in the target server 100, the target server 100 such as temperature (board internal temperature, memory temperature, network adapter temperature, etc.), fan speed (CPU Fan, System Fan, etc.). Information of the state of the device.

To this end, the status information monitoring unit 250 may set an environment for IPMI interworking upon receiving the boot notification of the target server 100. The status information monitoring unit 250 may transmit and install a library and a tool for interworking with IPMI to the target server 100 notified of the booting.

In this case, the target server 100 may create an IPMI account for monitoring status information through IP allocation for IPMI when installing a library and a tool. When the IPMI account is created, as shown in FIG. 3, status information may be collected and monitored.

3 is a view for explaining the operation of the state information monitoring unit of FIG. 3 shows that the verification manager collects and provides status information when requesting status information, the management server 200 may collect the status information at a preset cycle according to the verification manager's setting and provide the verification information to the verification manager's terminal. .

Referring to FIG. 3, when the boot notification reception is confirmed, a corresponding connection may be checked from the target server 100 by requesting IPMI access information (IPMI account). When the connection is confirmed, the IPMI status information may be requested to receive the status information of the target server.

At this time, the status information monitoring unit 250 of the management server may provide the IPMI status information along with providing a notification informing the danger when the IPMI status information transmitted in step 7 is out of the preset range (step 8).

4 is a flowchart illustrating a server verification and management method according to an embodiment of the present invention. This may be described with reference to FIGS. 1 to 3.

In order to verify the target server 100 in the management server 200, information for operations of the plurality of target servers 100, the DHCP expansion server 300, and the TFTP expansion server 400 may be set (S300). In this case, the settings may be stored in the corresponding server (100, 300, 400) and the data storage (10).

Next, the management server 200 may remotely start or manually turn on the power through a wake-on-lan signal to the target server 100 to be verified according to a predetermined test schedule. (S305).

Next, the target server 100 may request an IP assignment for booting into a predetermined DHCP expansion server 300 at power-on (S310). DHCP expansion server 300 received the IP allocation request from the target server 100, through the unique identifier of the target server, extracts the address value of the OS image set in the target server 100 from the data storage 10, IP The extracted OS image address value may be transmitted along with the assignment (S315).

Next, the target server 100 accesses the OS image address value through the assigned IP (S320). The OS image stored in the data storage 10 as an OS image address value may be extracted through the TFTP 400 (S325) and downloaded (S330). Next, the target server 100 is booted using the downloaded OS image (S335), and can transmit a boot notification to the management server 200 at boot (S340).

In addition, the booted target server 200 may transmit the configuration device information to the management server 200 according to the initial setting (S341).

In addition, the booted target server 100 may perform an environment setting for IPMI interworking according to an initial setting (S3431), and may monitor and provide status information through the set IPMI account (S3432). At this time, the status information monitoring may be performed at the request of the verification manager or at a preset cycle. Therefore, it may be performed at any stage after the boot notification is received.

The management server 200 may receive the configuration device information of the target server 100 received the boot notification (S342) and may generate a workflow through this (S344). The test workflow may be a task list scheduled according to the execution schedule of the test schedule and a test package matching the performance test included in the test schedule and the configuration device information received from the target server 100.

Next, the management server 200 transmits the location value of the test package to be performed (address value indicating the location stored in the data store 10) and the test setting value to the DHCP extension server 300 according to the generated workflow. It may be (S345). In this case, the test setting value is input by the verification manager and stored in the data storage 10, and the management server 200 may transmit only the storage location value of the test setting value.

Next, the DHCP expansion server 300 accesses the test package stored in the data store 10 through the TFTP expansion server 400 using the location value received from the management server 200 (S350), and the corresponding test package. It may be extracted (S355) and pushed to be downloaded to the target server 100 with the test set value (S360).

Next, the target server 100 may install the downloaded test package (S365) and perform a test for each test case according to the test setting value (S370). In addition, the test result according to the test can be transmitted to the management server (200) (S375).

Next, the management server 200 may receive a test result for each test case for each target server 100 in real time, and determine whether there is a defect by comparing the received test result with a predetermined threshold value (S380).

Here, the threshold may be a specific target value or threshold range. When the threshold value is set to a specific target value, it may be determined that there is no defect when the received test result value is the same as the target value. Alternatively, when the threshold is set to the threshold range, it may be determined that there is no defect when the received test result value is included in the threshold range which is a predetermined specific range.

On the other hand, the management server 200, when receiving a test result, if it is determined that an error occurs, may perform the test again according to the re-execution information for the test included in the test schedule. If there is re-execution information for the test, steps S345 to S380 may be performed again, or steps S370 to S380 may be performed again.

Meanwhile, steps S305 to S380 may be repeatedly performed during the test execution frequency or the test execution period of the test schedule. For example, if the main device test package is scheduled for two days, steps S305 to S380 are repeatedly performed for two days, and the management server 200 performs a test result according to the repetition and statistics on defects according to the test result. It can be calculated and provided in the form of a graph (S385).

Through this, the test defect degree (number of times, etc.) during the two days can be determined whether or not the pass per test package of the target server 100 by numerical verification. This may be determined by the verification manager or may be determined by setting acceptance criteria.

In addition, the management server 200 may adjust the test schedule of the target server 100 according to the test execution history and whether the defect of the test package for each target server 100 (S390), failing the target server 100 Extraction can provide a reason for rejection.

Through this, the verification manager can check whether the target server 100 failed by verifying. In addition, the verification manager may perform re-validation according to the adjusted test schedule after the defect is eliminated through the check on the reject target server 100.

In addition, what was described using FIG. 1 thru | or FIG. 4 above only described the main matter of this invention, and this invention is limited to the structure of FIG. 1 thru | or FIG. 4 as many designs are possible within the technical scope. It is not clear.

1000: Server verification automation and management system
100: target server (100a to 100n) 200: management server
210: setting unit 220: management unit
230: control unit 240: defect determination unit
300: DHCP expansion server 400: TFTP expansion server
500: network quality measurement server 10: data storage

Claims (5)

A data storage grouping the test programs classified according to the test purpose by grouping the devices and storing the test programs as test packages, and classifying the test programs by the device information;
A plurality of target servers transmitting configuration device information during booting and performing a test upon reception of a test package; And
When the configuration device information is received, extract a test package to be performed by the target server from schedule information including a test schedule, generate a workflow by extracting a test package matching the received configuration device information, and And a management server for controlling the plurality of target servers to perform a test according to a workflow.
The method of claim 1,
A DHCP (Dynamic Host Configuration Protocol) expansion server for allocating an IP to the target server and transmitting an address value of an OS image for booting when a boot request is received through the target server's PreBoot Execution Environment (PXE); And
Further comprising a TFTP (Trivial File Transfer Protocol) expansion server for transmitting the OS image to the target server through a protocol supported by the PXE, and transmits the test package to the target server booted using the OS image,
The management server transmits a location value and a test setting value of a test package to be performed according to the workflow to the DHCP expansion server, and the target server transmits the test package and the test setting value according to the workflow through the TFTP expansion server. Server verification automation and management system, characterized in that for receiving.
The method of claim 2,
The configuration device information is a manufacturer and model name of the configuration device configured by the target server, the test schedule includes a test transmission time to be performed for each target server, whether or not to re-execute when a test error occurs, and the number of times of test execution,
The management server, if it is determined that the received test result is an error, the server characterized in that for controlling the DHCP expansion server and TFTP expansion server to reinstall the failed test package in the target server according to the re-execution Verification automation and management system.
The method of claim 1,
When the management server receives a boot notification from the target server, the management server controls to install a tool for interworking the Intelligent Platform Management Interface (IPMI) on the target server, and generates an IPMI account by assigning an IP for IPMI. Server verification automation and management system.
The method of claim 1,
The target server is provided with a plurality of sensors for measuring the state of the configuration device,
The management server, at the request of a verification manager or at a predetermined cycle, receives the status information measured by the plurality of sensors through the IPMI interworking environment, and the server characterized in that notified when the received status information is out of the preset range Verification automation and management system.

Images