JP5531487B2 - Server system and server system management method - Google Patents

Description

  The present invention relates to a server system including a plurality of servers represented by a blade server and the like, and a management method thereof.

  Conventionally, in a system where computers are connected to each other via a local area network, if the board is replaced due to a failure of the communication control board, it is necessary to notify all computers connected to the local area network of the changed MAC address. there were.

  On the other hand, Patent Document 1 discloses a technique that allows both the MAC address before replacement and the MAC address after replacement to be used when the MAC address is changed by replacing the communication control board. In such a MAC address management system, it is not necessary for another computer to be aware of the MAC address change.

JP-A-10-65702

  By the way, also in a server system, when a certain server fails, the problem similar to the above arises. That is, for example, when the server system is operated in a redundant configuration, the failed server is replaced after being switched to the standby server. However, the hardware specific information of the standby server is different from the hardware specific information of the failed server. For this reason, application software that depends on this information and external management software may need to be reinstalled or reconfigured before starting operation on the standby server. Here, the hardware specific information indicates a serial number, UUID (Universally Unique Identifier), MAC address (Media Access Control address), WWPN (World Wide Port Name), and the like.

  Further, even in the case of a non-redundant system, the failed server has hardware-specific information different from that of the new server, so the OS (Operating System) and system management environment before the failure cannot be used as they are. There is a case. In other words, when a server fails, if the failed server is replaced, various hardware-specific information changes, so that the operations of the management software and application software that operate by referring to them are not affected. You need to change these software settings or reinstall the software. It is desired to simplify such complicated processes such as application software when replacing a failed server, reinstallation of external management software, and resetting.

  The present invention has been made to solve such problems, and an object of the present invention is to provide a server system and a management method thereof that can facilitate replacement of a failed server.

  The server system according to the present invention includes a plurality of servers and a management module that manages the plurality of servers. The management module has a function of managing virtual HW information indicating virtual hardware specific information for all servers and association information for assigning the virtual HW information to each server. The server has a virtual HW information storage unit that stores the virtual HW information allocated by the management module, and generates system management BIOS (Basic Input / Output System) information based on the virtual HW information. Is provided.

  The server system management method according to the present invention is a server system management method having a plurality of servers. First, virtual HW information indicating virtual hardware specific information and association information for assigning the virtual HW information to each server are generated for all servers. Next, the virtual HW information is transferred to all servers according to the association information. Each server generates system management BIOS (Basic Input / Output System) information based on the transferred virtual HW information.

  ADVANTAGE OF THE INVENTION According to this invention, the server system which can make easy replacement | exchange of a failure server, and its management method can be provided.

It is a figure which shows schematic structure of the server system concerning embodiment of this invention. It is a figure which shows the blade server system concerning Embodiment 1 of this invention. In the server system concerning an embodiment of the invention, it is a sequence figure showing a procedure until virtual HW information is generated and it reflects in each blade server. It is a sequence diagram which shows the operation | movement at the time of blade server replacement | exchange of the server system concerning embodiment of this invention. It is a sequence diagram explaining operation | movement at the time of the failure generation of the server system concerning embodiment of this invention. It is a figure which shows the blade server system concerning Embodiment 2 of this invention.

  Hereinafter, specific embodiments to which the present invention is applied will be described in detail with reference to the drawings. In this embodiment, the present invention is applied to a blade server system. However, the present invention is not limited to a blade server system, and the same applies to other server systems such as a rack mount server and a tower server. Can be applied to.

  First, the outline of the present invention will be described. FIG. 1 is a diagram showing a schematic configuration of a server system according to an embodiment of the present invention. As illustrated in FIG. 1, the blade server system 100 includes a plurality of blade servers 300 and a management module 200 that manages the plurality of blade servers 300.

  The management module 200 has a function of managing virtual HW information indicating virtual hardware specific information for all blade servers and linking information for assigning the virtual HW information to each blade server. The hardware specific information is a serial number, UUID, MAC address, WWPN (World Wide Port Name), and the like, and the virtual HW information is virtual information that replaces the hardware specific information. The association information includes position information of the slot in which each blade server is stored, and using this association information, the management module 200 transfers the virtual HW information to each blade server.

  The blade server 300 includes a virtual HW information storage unit 304 that stores virtual HW information allocated by the management module 200. A function of generating system management BIOS (Basic Input / Output System) information (hereinafter referred to as SMBIOS (System Management BIOS) information) is provided based on the allocated virtual HW information.

  That is, in this embodiment, the hardware specific information of each blade server 300 is virtualized. The management module 200 generates virtual HW information instead of the hardware specific information of the blade server 300 and manages it collectively. By setting arbitrary virtual HW information in each blade server 300, hardware can be virtualized for application software, external management software, and OS.

  Normally, the OS on the blade server 300 and the application software running on the OS acquire hardware specific information (such as a serial number and UUID) from the SMBIOS information. This SMBIOS information is generated by the BIOS of each blade server 300. In the present embodiment, the management module 200 has a function of generating and managing virtual HW information for all servers in the system. The virtual HW information can be exchanged directly with the BIOS of each blade server 300 or via another module. Further, the BIOS of each blade server 300 has a function of generating SMBIOS information based on the received virtual HW information. The BIOS of each blade server 300 generates the SMBIOS based on the virtual HW information, not the normal HW information. Therefore, even if the server at the time of the server failure is replaced, the virtual HW information can be used as it is. .

  In this way, by applying the virtual HW information to all blade servers 300 in the system, and operating the standby blade in the event of a failure, the same virtual HW information is also applied to the standby blade. The blade server 300 can be easily replaced at the time of failure. That is, by using the same virtual HW information even if the server is replaced, the application software and the external management software do not need to be reinstalled or reset before starting operation on the standby server.

Embodiment 1 of the Invention
FIG. 2 is a diagram illustrating the blade server system according to the first embodiment of the present invention. The blade server system 100 includes a management module 200 that performs overall system control, fault monitoring, and the like, and a plurality of blade servers 300. In this embodiment, the management server 400 that performs fault monitoring and operation management by the management software 401 can communicate with the management module 200 and the blade server 300 via a network.

  The management module 200 includes virtual HW information 201 and association information 202 for assigning the virtual HW information 201 to each blade server 300.

  The blade server 300 includes a virtual HW information storage part 304 that stores assigned virtual HW information, an HW information storage unit 305 that stores original hardware-specific information, and SMBIOS information in which SMBIOS information generation results described later are stored. A generation result storage unit 306. In addition, a management controller (hereinafter referred to as a BMC (Baseboard Management Controller)) 301 that performs failure monitoring and power control inside the server is included. Furthermore, a program group (hereinafter referred to as BIOS) 302 for controlling peripheral devices such as a disk drive, a keyboard, or a video card connected to the blade server 300 is provided. The BIOS 302 generates SMBIOS information. Furthermore, it has OS308 which operate | moves on the blade server 300. FIG. The OS 308 has an interface API (application programming interface) that operates application software.

  The management module 200 can communicate with the BMC 301 on the blade server 300 through a network or a serial connection. The management module 200 can access the virtual HW information storage unit 304 via the BMC 301. The management module 200 has a function of generating and managing the association information 202 so that the virtual HW information assigned to the blade server 300 does not overlap the virtual HW information assigned to the other blade server 300.

  The BMC 301 of the blade server 300 is connected to the virtual HW information storage unit 304. The BMC 301 can communicate with the management module 200 and the BIOS 302, whereby the management module 200 and the BIOS 302 can access the virtual HW information storage unit 304.

  The management software 401 on the management server 400 manages each blade server 300 with reference to virtual HW information handled on the OS 308. Note that the communication interface between the management server 400 and the management module 200 and between the management module 200 and the BMC 301 is not particularly defined.

  Next, a method for setting the virtual HW information in each blade server 300 will be described. FIG. 3 is a sequence diagram showing a procedure for generating virtual HW information and reflecting it on each blade server 300.

  The user creates virtual HW information 201 and stores it in the management module 200. The management module 200 has a user interface for creating the virtual HW information 201.

  The user assigns the stored virtual HW information 201 to the blade server 300 using the user interface of the management module 200 (step S101). The management module 200 confirms that the virtual HW information 201 specified by the user does not overlap with the virtual HW information 201 of other blade servers 300 (step S102).

  If there is no overlap in the virtual HW information 201, the management module 200 generates the association information 202 according to the user instruction (step S103). Then, the management module 200 transfers the virtual HW information 201 to the BMC 301 of each blade server 300 according to the association information 202 (Step S104). The BMC 301 stores the received virtual HW information 201 in the virtual HW information storage unit 304 (step S105). The BMC 301 transmits a reception completion notification notifying the management module 200 that reception of the virtual HW information 201 has been completed (step S106).

  Next, the user uses the interface of the management module 200 to instruct the BMC 301 to turn on the blade server 300 (step S107). The management module 200 instructs the BMC 301 to power on the instructed blade server 300 (step S108). As a result, the BMC 301 turns on the blade server 300 (step S109).

  When the power is turned on, the BIOS 302 requests the virtual HW information 201 from the BMC 301 (step S110). The BMC 301 refers to the virtual HW information storage unit 304 and transfers the virtual HW information 201 to the BIOS 302 when the virtual HW information 201 exists. If the virtual HW information 201 does not exist, the HW information storage unit 305 is referred to and the original HW information is transferred to the BIOS 302 (step S111).

  The BIOS 302 generates SMBIOS information 303 based on the received virtual HW information 201 (step S112). The BIOS 302 transfers the result of whether or not the SMBIOS information 303 has been normally generated to the BMC 301 (step S113). The BMC 301 stores the SMBIOS information generation result 306 in the SMBIOS information generation result storage unit 307 (step S114). Finally, the BIOS 302 notifies the management module 200 that the SMBIOS information generation result 306 has been stored in the information generation result storage unit 307 (step S115).

  Next, an operation when replacing the failed blade server 300 will be described. FIG. 4 is a sequence diagram showing an operation at the time of blade server replacement. First, the management module 200 detects that the replacement blade server 310 is installed (step S201). The management module 200 has RAS (Reliability, Availability, Serviceability) functions such as mounting detection and failure detection of each blade. Then, the management module 200 confirms whether or not the association information 202 for the exchanged slot exists (step S202). If the link information 202 of the slot in which the replacement blade server 310 is mounted exists, the virtual HW information 201 is transferred to the BMC 301 of the mounted replacement blade server 310 according to the link information 202. Subsequent operations are the same as those after S105 described above.

  Next, the operation for switching the operation to the standby blade when a failure occurs will be described. FIG. 5 is a sequence diagram for explaining the operation when a failure occurs. The redundant system is equipped with a spare blade server (standby blade server) that can be switched when a failure occurs. When a failure occurs in a certain blade server, a standby blade server is activated instead of the blade server in which the failure has occurred.

  First, the management software 401 requests the management module 200 to switch to a standby blade server (step S301). The management module 200 changes the association information 202 in accordance with the request from the management software 401 (step S302). Then, the management module 200 refers to the changed association information 202 and transfers the virtual HW information 201 to the BMC 301 of the standby blade server 310 (step S303). The BMC 301 stores the received virtual HW information 201 in the virtual HW information storage unit 304 (step S304). The BMC 301 transmits a reception completion notification notifying the management module 200 that reception of the virtual HW information 201 is completed (step S305). Upon receiving this, the management module 200 notifies the management software 401 that the server switching preparation has been completed. The subsequent steps are the same as the steps after step S108 described above.

  As described above, in the present embodiment, the virtual HW information 201 is set for each blade server 300. This eliminates the need to change the system settings associated with changes in hardware-specific information during server fault replacement or switching to a standby blade server. For example, when switching to the standby system, the same virtual HW information 201 as the virtual HW information 201 used by the failed server can be set in the standby blade server. Therefore, it is not necessary to change the setting of each software using the virtual HW information 201 such as OS, management software, application software, and the like. In addition, when replacing a failed server, simply installing the replacement server, the replaced server can operate as a server having the virtual HW information 201 used by the failed server. Accordingly, similarly, it is not necessary to change settings of the OS, management software, application software, and the like.

  In the first embodiment, an example of a blade server has been described. However, a rack server, a tower type server, and the like can be similarly configured. That is, the form of the server is arbitrary.

  In the first embodiment, the server system is externally managed by the management server 400 and the management software 401. However, the management module 200 may manage the server system independently without providing them. Good.

  Furthermore, in the first embodiment, the example in which the system is configured by one blade server casing has been described. However, a system including a plurality of casings may be used. In that case, the management module 200 for each case may cooperate to perform the movement of the virtual HW information 201 between the cases, the deduplication process, and the like. Next, a system including a plurality of such cases will be described.

Embodiment 2 of the Invention
FIG. 6 is a diagram showing a blade server system according to the second exemplary embodiment of the present invention. In the present embodiment shown in FIG. 6, the same components as those in the first embodiment shown in FIG. 2 are denoted by the same reference numerals, and detailed description thereof is omitted.

  The blade server according to the present embodiment has a plurality of blade server housings. Each blade server housing has a management module 200. Each management module 200 has virtual HW information 201 in its own housing and its linking information 202. The association information 202 includes information (212,..., 2 × 2) for identifying the casing in addition to the blade thread position information. In this embodiment, any one of the management modules 200 becomes the master management module 210, and the virtual HW information 201 and the association information 202 are centrally managed. Note that the link information 202 may be shared and managed by all the management modules 200.

  The master management module 200 refers to the association information 202 that is centrally managed, and checks whether or not the virtual HW information 201 is duplicated in the entire system. When each of the slave management modules 220 sets the virtual HW information 201, the slave management module 220 transfers the set contents, that is, the virtual HW information 201 and the server mounting position information to be set to the master management module 210. Based on the setting content received from the slave management module 220, the master management module 210 updates the association information 202 that is centrally managed.

  Also in this embodiment, the same effects as those of the first embodiment are obtained. That is, even when the server is replaced or replaced with a standby server, the original virtual HW information 201 can be used as it is only by changing the association information 202 as necessary. Therefore, a complicated process such as re-installation and resetting of application software and external management software when replacing a failed server is not required.

  It should be noted that the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

DESCRIPTION OF SYMBOLS 100 Blade server system 200 Management module 201 Virtual HW information 202 Linking information 210 Master management module 220 Slave management module 300 Blade server 310 Replacement blade server 301 BMC
302 BIOS
303 SMBIOS information 304 Virtual HW information storage unit 305 HW information storage unit 308 OS
401 management software 400 management server

Claims (17)

Multiple servers,
A management module for managing the plurality of servers,
The management module has a function of managing virtual HW information indicating virtual hardware specific information for all servers and association information for assigning the virtual HW information to each server,
The server has a virtual HW information storage unit that stores the virtual HW information allocated by the management module, and has a function of generating system management BIOS (Basic Input / Output System) information based on the virtual HW information. Prepared ,
The server includes a virtual HW information storage unit that stores the virtual HW information allocated by the management module, and an HW information storage unit that stores hardware specific information that is hardware information specific to the server. When there is the virtual HW information in the virtual HW information storage unit , system management BIOS (Basic Input / Output System) information is generated based on the virtual HW information, and the virtual HW information is stored in the virtual HW information storage unit A server system that generates system management BIOS (Basic Input / Output System) information based on the hardware-specific information in the HW information storage unit when there is no such information .   The server system according to claim 1, wherein the management module can communicate with a management controller that performs fault monitoring and / or power control inside the server via a network and / or serial connection. The server has a program group for controlling peripheral devices connected to the server,
The management controller of the server can communicate with the management module and the program group,
The server system according to claim 2 , wherein the management module and the program group are accessible to the virtual HW information storage unit via the management controller.   The server system according to claim 3, wherein the program group of the server generates the system management BIOS by referring to the virtual HW information in the virtual HW information storage unit.   5. The management module according to claim 1, wherein the management module has a function of generating and managing the association information so that the virtual HW information allocated to the server does not overlap between servers. Server system.   The server system according to any one of claims 1 to 5, further comprising a management server that can communicate with the server and the management module via a network, and that performs system fault monitoring and / or operation management.   The server system according to any one of claims 1 to 6, wherein the plurality of servers are blade servers, rack mount servers, or tower servers. Each server is stored in a slot
The association information includes slot position information for assigning the virtual HW information to the slot,
The management module, when detecting that a replacement server, which is a new server, is mounted, transfers the virtual HW information to the replacement server according to the association information for the mounted slot. The server system according to claim 1. Each server is stored in a slot
The association information includes slot position information for assigning the virtual HW information to the slot,
When the management module changes from the current server to the standby server, the slot information of the association information is changed from the slot where the current server is stored to the slot position information of the slot where the standby server is stored, The server system according to claim 1, wherein the virtual HW information is transferred to the standby server according to the changed association information. Having multiple blade server enclosures,
Each management module is arranged for each blade server,
The server system according to any one of claims 1 to 9, wherein the association information includes slot position information indicating a slot position of a blade and case identification information indicating which blade case. A master management module for centrally managing the virtual HW information and the association information;
The slave management module, which is a management module other than the master management module, transfers the virtual HW information and the association information in the set blade server to the master management module,
The server system according to claim 10, wherein the master management module checks whether the virtual HW information is duplicated based on received information.   The server system according to claim 10, wherein the association information is shared and managed among all management modules.   The server system according to claim 11, wherein, when the association information is updated, the slave management module transfers the updated association information to the master management module. A method for managing a server system having a plurality of servers,
For all servers, generate virtual HW information indicating virtual hardware specific information, and link information for assigning the virtual HW information to each server,
Transfer the virtual HW information to all servers according to the linking information,
Each server is transferred when said virtual HW information exists on the basis of the virtual HW information, generates a System Management BIOS (Basic Input / Output System) information, wherein when the virtual HW information does not exist the A server system management method for generating system management BIOS (Basic Input / Output System) information based on hardware specific information which is hardware information specific to a server. A program group for controlling peripheral devices connected to the server accesses a virtual HW information storage unit in which the virtual HW information is stored, via a management controller that performs fault monitoring and / or power control inside the server,
The server system management method according to claim 14, wherein the program group refers to the virtual HW information in the virtual HW information storage unit and generates the system management BIOS. Each server is stored in a slot
The association information includes slot position information for assigning the virtual HW information to the slot,
The server system according to claim 14 or 15, wherein when detecting that a replacement server which is a new server is mounted, the virtual HW information is transferred to the replacement server according to the association information for the mounted slot. Management method. Each server is stored in a slot
The association information includes slot position information for assigning the virtual HW information to the slot,
When changing from the current server to the standby server, the slot position information of the association information is changed from the slot where the current server is stored to the position information of the slot where the standby server is stored,
The server system management method according to any one of claims 14 to 16, wherein the virtual HW information is transferred to the standby server according to the changed association information.

Images