CN102479142A - Rack server system and monitoring method thereof - Google Patents

Abstract

The invention provides a rack server system and a monitoring method thereof, wherein the rack server system comprises a plurality of racks and a management unit. The racks respectively comprise a plurality of servers and a control unit. The control unit is connected with the servers and acquires state information of each server, and the control unit updates the state information along with time. The management unit is used for being in communication connection with the control units on the racks, the management unit obtains the state information of the server from the control units to generate a state information table, and meanwhile the management unit establishes an automatic program. And the automatic program triggers the control unit on each rack and updates the state information to the management unit every other preset period. Therefore, repeated login authentication procedures and waiting time can be saved, and a user or a system manager can more conveniently master the overall state of the rack server system.

Description

A rack server system and monitoring method thereof

technical field

The present invention relates to a monitoring mechanism, and in particular to a monitoring system and method of a rack server system. the

Background technique

Due to the increasing progress of the computer industry and the increasing dependence of enterprises on industrial computer systems, industrial computers usually refer to non-personal computers or non-consumer electronic systems. For example, industrial computers include core control equipment used in factory automation , Web server, enterprise data backup server, etc. the

With the development of electronic technology and related applications of industrial computers, the industry's requirements for server systems have naturally increased relatively. In order to cope with the growth of enterprises, many enterprises usually integrate a large number of server units into computer cases, and can use multiple sets of computer cases to form industrial computer units to meet the needs of a large amount of data or network traffic. It can meet the needs of expansion and upgrade of the server system in the future. the

Generally speaking, a set of industrial computer units may include multiple system racks, and each system rack further includes multiple computer servers (such as blade computers). the

In addition, general industrial computers need to be in a running state for a long time. Once any hardware failure occurs, enterprises often require the server system to be repaired or replaced without stopping. Therefore, system functions such as the Hot Swap requirement of the motherboard, power supply distribution, and mutual support of cooling fans are also created accordingly. In addition, in order to improve the stability of power supply in industrial computer equipment, many protection mechanisms such as system power status monitoring, power module error correction, and voltage and current control have been developed. the

In practical applications, users or computer system administrators need to monitor the operation status of each server on each rack. On existing computer sets, users can connect to the baseboard management controllers on each rack through a specific network communication protocol (such as SSH encrypted network protocol) through the browser program or the control interface of other connection software. to view status information for a specific server on a specific rack. However, a computer group may have several or even dozens of racks, and each rack may contain multiple servers. If monitoring is required, it is necessary to log in to the monitoring interface on each rack one by one, and Waiting for password verification is very inconvenient to use. the

In addition, the existing monitoring interface cannot refresh the current state information, that is, when the user needs to reconnect to the control unit (such as the rack baseboard management controller) on each rack to obtain the latest state information, It also caused troubles in monitoring and management. the

Contents of the invention

In order to solve the above problems, the present invention proposes a rack server system and monitoring method, through which the management unit communicates with the control units on each rack, and the management unit can read the status of the rack specified by the user through the corresponding control unit Information, and its content can be integrated and displayed on a display, which is convenient for users to review. In addition, when the management unit obtains the status information, an automatic program is established at the same time, and every predetermined period, the automatic program can trigger the status information to be actively sent back to the management unit for updating. In this way, repeated login authentication procedures and waiting time can be saved, allowing users or system administrators to more easily grasp the overall status of the rack server system. the

Therefore, one of the objectives of the present invention is to provide a rack server system, which includes a plurality of racks and a management unit. Wherein, each rack contains multiple servers and a control unit. The control unit is connected with the server and obtains a state information of each server, and the control unit updates the state information with time. The management unit is used for communicating with the control unit on each rack. The management unit obtains the status information of the server from the control unit to generate a status information table, and at the same time, the management unit establishes an automatic program. Wherein, every predetermined period, the automatic program triggers the control unit on each rack and updates the status information to the management unit. the

According to an embodiment of the present invention, each server includes a server baseboard management controller to receive a temperature value of each server. the

According to an embodiment of the present invention, the control unit of each rack is a rack baseboard management controller. the

According to an embodiment of the present invention, each rack further includes a plurality of fans, and the control unit on each rack receives a plurality of rotational speed values of the fans, and controls the rotational speed of the fans. the

According to an embodiment of the present invention, the state information table includes multiple network addresses of racks, multiple temperature values of servers, and multiple rotational speed values of fans. the

According to an embodiment of the present invention, wherein the management unit selects a part of the racks according to user operations, the management unit obtains the status information from the corresponding control unit for the selected racks and establishes the automatic program. the

According to an embodiment of the present invention, the automatic program is established and executed on an arithmetic processor or the control unit of the racks. the

According to an embodiment of the present invention, the management unit displays the content of the state information table on a display. the

Another object of the present invention is to provide a monitoring method for a rack server system, the rack server system includes a plurality of racks and a management unit, the racks respectively have a plurality of servers and a control unit, wherein The monitoring method includes the following steps:

(1) In the rack, obtain a state information of each server through the control unit, and the control unit updates the state information with time;

(2) The management unit communicates with the control unit to obtain status information; and

(3) Establishing an automatic program, so as to trigger the control unit on each rack to send status information to the management unit every predetermined period. the

According to an embodiment of the present invention, the monitoring method further includes the following steps: select a part of the system units from the system units according to the user operation, and in the above step (2) and step (3), the management unit targets the selected The rack obtains the status information and establishes the automatic program by the corresponding control unit. the

In this way, repeated login authentication procedures and waiting time can be saved, allowing users or system administrators to more easily grasp the overall status of the rack server system. the

Description of drawings

In order to make the above and other objects, features, advantages and embodiments of the present invention more obvious and understandable, the detailed description of the accompanying drawings is as follows:

FIG. 1 is a schematic diagram of a rack server system according to an embodiment of the present invention. the

FIG. 2 is a flow chart of a monitoring method according to an embodiment of the present invention. the

FIG. 3 is a schematic flowchart of an operation example of the monitoring method in FIG. 2 . the

【Description of main component symbols】

100: rack server system 120: rack

122: Server 124: Control Unit

126: Fan 140: Snap-in unit

160: Display S100: Steps

S102: Step S104: Step

S200: Step S202: Step

S204: Step S206: Step

S208: Step S210: Step

Detailed ways

Please refer to FIG. 1 , which is a schematic diagram of a rack server system 100 according to an embodiment of the present invention. As shown in FIG. 1 , the rack server system 100 includes multiple racks 120 , a management unit 140 and a display 160 . In this embodiment, the management unit 140 can be used to correspondingly manage multiple groups of racks 120. Only three groups of racks 120 are shown in FIG. 1 as an example, but the present invention is not limited to three groups. In this embodiment, the rack server system 100 as a central data server (data center) may include a total of dozens of racks 120 or more. the

Wherein, each group of racks 120 includes a plurality of servers (servers) 122 and a control unit 124 . In FIG. 1 , each rack includes two groups of servers 122 as an example, but the present invention is not limited thereto. In this embodiment, each server 122 may include a server baseboard management controller (BMC), and the server baseboard management controller may be used to receive the temperature value of each server 122 during operation. the

In this embodiment, the control unit 124 is a rack baseboard management controller (BMC) on the rack 120, the control unit 124 can be connected with the server baseboard management controllers on the respective servers 122 and obtain a state of each server 122 information, the control unit 124 updates the status information over time. the

Wherein each frame 120 can further comprise a plurality of fans 126, and the control unit 124 (frame BMC) on each frame 120 receives the rotating speed value of fan 126, and controls the rotating speed of fan 126, comprises with each frame 120 among Fig. 1 Two sets of fans 126 are taken as an example, but the present invention is not limited thereto. the

The control unit 124 can obtain status information of each server 122, and can obtain and control the rotation speed of the fans on the rack 120, but the present invention is not limited thereto. In practical applications, the items that can be monitored by the control unit 124 may further include temperature, humidity, power supply voltage, fan speed, communication parameters, operating system functions, and other operating parameters on the rack. In addition, in this embodiment, the control unit 124 can update the status information over time to maintain the immediacy of the status information. the

The management unit 140 is used to communicate with the control unit 124. Therefore, the management unit 140 obtains the status information of the server 122 and various information on the rack 120 from the control unit 124, and the management unit 140 then generates a status information table accordingly, wherein the status The content of the information table may include the network address (such as IP address) of the rack 120, the temperature value of the server 122 and the rotational speed value of the fan 126. In this embodiment, the management unit 140 communicates with the control unit 124 through a network protocol. In order to enhance the security of the transmission and perform user identity authentication, the network protocol here can adopt the Secure Shell (SSH) encrypted network protocol in practical applications. When a user logs in to the control unit 124 through the SSH encrypted network protocol, a user name and a password need to be input, and a certain authentication procedure and authentication time are required. the

It should be noted that the management unit 140 obtains status information and simultaneously establishes an automatic program. This automatic procedure is used to trigger the control unit 124 on the rack 120 to transmit status information to the management unit 140 every predetermined period. In this embodiment, the automatic program may be a software script (software script) established on the control unit 124, but the present invention is not limited thereto. In another embodiment, the rack 120 may further include a computing processor (not shown), and the automatic program may also be a software script created and executed on the computing processor. the

In this way, every predetermined period, the automatic program can trigger the rack 120 to actively send back the status information to the management unit 140 for updating. In this way, repeated authentication procedures and waiting time for logging in to the control unit 124 can be saved, and the management unit 140 can display the contents of the generated status information table on the display 160, so that the user or system administrator can more easily grasp the rack The overall status of the server system 100. the

In addition, it should be added that, for the convenience of use or the consideration of processing efficiency, the management unit 140 is not limited to reading the status information of all racks 120 at the same time. In practical applications, the rack server system 100 can provide users with an operation Interface (such as keyboard, mouse and other input devices, switch buttons, touch interface, etc.), the management unit 140 can select a part of the racks from the racks 120 according to the user operation, for example, select the first group and the second group of machines shelf. The management unit 140 obtains status information from the corresponding control unit 124 for the selected rack 120 and establishes an automatic program. the

Please refer to FIG. 2 , which shows a flow chart of a monitoring method according to an embodiment of the present invention. In this embodiment, the monitoring method can be used in a rack server system, the rack server system includes a plurality of racks and a management unit, and the racks respectively have a plurality of servers and a control unit. The internal hardware structure of the rack server system here may be substantially similar to the rack server system 100 in the previous embodiment, please refer to the previous embodiment of the present invention and the rack server system 100 in FIG. 1 . As shown in FIG. 2 , the monitoring method of the present invention first executes step S100 , and acquires a state information of each server through the control unit in the rack, wherein the control unit updates the state information with time. the

Next, step S102 is executed, the management unit communicates with the control unit, and the management unit acquires status information from the control unit. the

Finally, step S104 is executed to establish an automatic program, so as to trigger the rack to transmit the status information to the management unit every predetermined period, and the management unit can further establish a status information table. In this way, the automatic program can actively return the status information to the management unit periodically, so that the management unit can refresh the contents of the status information table. In this way, the procedure and time of repeated communication connection formation can be saved, and the user or system administrator can monitor the overall status of the rack server system more conveniently. the

In addition, the monitoring method of the present invention is not limited to reading the state information tables of all racks at the same time. In another embodiment, the monitoring method of the present invention can further execute a step to select a part of racks from the racks according to the user operation, and correspondingly in the above step S102 and step S104, the management unit for The selected rack obtains status information from the corresponding control unit and establishes an automatic program, thereby improving data processing efficiency. the

In addition, please also refer to FIG. 3 , which shows a schematic flowchart of an operation example of the monitoring method in FIG. 2 . In practical application, the control unit in the rack can be preset to automatically generate status information (ie step S100 in FIG. 2 ). As shown in FIG. 3 , in an operation example of a system administrator or a user, step S200 may be executed first, and a corresponding connection page on a browser is opened on the rack. the

Next, step S202 is executed to allow the user to select a single rack or a rack group to be displayed on the connection page. the

Next, the management unit executes step S204 to determine whether the status information of the selected rack exists on the management unit. the

At this time, if the above status information table does not exist, it means that the management unit has not yet been connected to the selected rack. At this time, steps S206-S208 (see steps S102-S104 in the preceding paragraph) can be executed to obtain Status information and build automated programs. Then step S210 can be executed to create a state information table according to the obtained state information, and display its content on the corresponding connection page of the browser, and the automatic program will automatically update the content of the state information table. the

If in step S204, it is judged that the above status information already exists, it means that the management unit has completed the connection with the selected rack, and the status information has been obtained and an automatic program has been established. Displayed on the corresponding connection page of the browser. the

It can be seen that, only when the management unit is connected to a specific rack for the first time, it needs to obtain the state information table from the rack through a network protocol (such as the SSH protocol) and establish an automatic program (that is, steps S102-S104 in FIG. 2 or Steps S206-S208 in 3), after that, the rack will actively refresh the contents of the state information table on the management unit. the

In this way, repeated login authentication procedures and waiting time can be saved, allowing users or system administrators to more easily grasp the overall status of the rack server system. the

Although the present invention has been disclosed as above in terms of implementation, it is not intended to limit the present invention. Any person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, this The scope of protection of the invention shall be defined by the scope of the appended patent application. the

Claims (10)

1. A rack server system, characterized in that the system comprises: a plurality of racks, each rack containing: multiple servers; and a control unit, the control unit is connected to the server and obtains a state information of each of the servers, and the control unit updates the state information over time; and A management unit is used to communicate with the control unit on each of the racks, the management unit obtains the status information of the server from the control unit to generate a status information table, and establishes an automatic program; Wherein, every predetermined period, the automatic program triggers the control unit on each rack and updates the status information to the management unit. 2 . The rack server system according to claim 1 , wherein each of the servers comprises a server baseboard management controller to receive a temperature value of each of the servers. 3 . 3. The rack server system according to claim 1, wherein the control unit of each rack is a rack baseboard management controller. 4. The rack server system according to claim 1, wherein each of the racks further comprises a plurality of fans, and the control unit on each of the racks receives a plurality of rotational speed values of the fans, And control the speed of the fan. the 5. The rack server system according to claim 1, wherein the state information table includes multiple network addresses of the rack, multiple temperature values of the server, and multiple rotational speeds of the fan value. 6. The rack server system according to claim 1, wherein the management unit selects a part of the racks from the racks according to the user operation, and the management unit controls the selected racks by The corresponding control unit obtains the state information and establishes the automatic program. 7. The rack server system according to claim 1, wherein the automatic program is established and executed on an arithmetic processor of the rack or the control unit. 8. The rack server system according to claim 1, wherein the management unit displays the content of the state information table on a display. 9. A monitoring method for a rack server system, the rack server system comprising a plurality of racks and a management unit, the racks respectively have a plurality of servers and a control unit, characterized in that the Said monitoring method comprises the following steps: In the rack, obtain a piece of status information of each of the servers through the control unit, and the control unit updates the status information with time; The management unit is communicatively connected to the control unit to obtain the status information; and An automatic program is established to trigger the control unit on each of the racks to transmit the status information to the management unit every predetermined period. 10. The monitoring method according to claim 9, further comprising the following steps: A part of the racks is selected according to the user's operation, and the management unit obtains the state information and establishes the automatic program from the corresponding control unit for the selected racks. the

Images