JP2007293519A - Server system and patch processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a patch processing technology suitable for a system where the operation configurations of a server coexist. <P>SOLUTION: This server system is provided with an active system server device (110) for executing a process (150); a duplex unit (101) equipped with a standby system server device (120) which is put in a standby status for an alternative operation for the active system server device; a simplex unit (102) equipped with a server device (130) for executing the same process (150) as the above process; and a maintenance server device (140) connected to each server device of both the units. The maintenance server device transmits the patch data (141) for correcting the process to the simplex unit and the duplex unit. The standby server device notifies the operation system server device of the success/failure of patch processing. The operation system server device decides the content of notification from the standby server device, and executes the patch processing of its own device from the patch data when it indicates the success/failure of the patch processing. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a server system in which a patch process for correcting a computer program is executed, and more particularly to a patch process in a system in which a single operation and a dual operation of server devices are mixed.

  Conventionally, patch processing is known as a technique for correcting a program installed in a computer. By executing this patch process, a defective part (bug) of the program or a part to be updated is automatically corrected.

  By the way, the client server system includes a server single operation, that is, an operation form in which a single server processes requests from a plurality of clients, and an operation form in which servers are duplicated. In a redundantly operated server system, a separate server as a backup is made to stand by for a server that is regularly operated. Then, when the regular server goes down due to a failure or the like, the spare server automatically starts the alternative operation.

As a technique related to patch processing in such duplex operation, for example, there is one described in Patent Document 1 described later. In the technique described in Patent Document 1, when there is a large amount of patch processing correction data to be sent to a cluster, the correction data is divided and transmitted, and the number of divisions is notified, and patch processing due to a correction data transmission error is performed. It is intended to avoid failure.
Japanese Patent Laid-Open No. 61-117645

  However, there is a case where duplication is not required depending on the use or frequency of use of the server, and there are currently systems in which the single operation and duplication operation of the server are mixed. Therefore, it is desired to propose patch processing for a system in which server operation forms are mixed.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a patch processing technique suitable for a system in which server operation modes are mixed.

  A server system according to the present invention executes a process identical to the above-described process, and a duplex unit provided with an active server apparatus that executes a process and a standby server apparatus that waits for an alternative operation for the active server apparatus And a maintenance server device connected to the single unit and the duplex unit, wherein the maintenance server device receives patch data for correcting the process as the single unit. And means for transmitting to the duplex unit.

  According to the present invention, it is possible to smoothly perform patch processing in a system in which server operation forms coexist by configuring the maintenance server device to distribute the process patch data to each server of the single unit and the dual unit. it can.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows a configuration of a server system 100 according to the embodiment. The server system 100 includes a duplex unit 101 that performs server duplex operation by the active server 110 and the standby server 120, a duplex unit 102 that performs duplex operation by the single server 130, and both units (101/102). Each server (110/120/130) includes a maintenance server 140 connected to be communicable.

  In the duplexing unit 101, the active server 110 operates regularly, and the standby server 120 waits for an alternative operation when a failure occurs while monitoring the status of the active server 110. On the other hand, the single unit 102 does not have a spare server for the stand-alone server 130 that is normally used. In the illustrated example, each server is one, but a plurality of servers may exist in actual operation.

  The processes 150a / 150b / 150c executed by the active server 110, the standby server 120, and the stand-alone server 130 are programs having the same processing contents.

  The maintenance server 140 has patch data 141 used for patch processing for correcting the process (150). The patch data 141 includes correction data for the function program described in the process (150), processing programs necessary for the correction, and the like.

  FIG. 2 schematically shows patch processing executed by each server (110/120/130). A process use area 210 in the illustrated main memory 202 indicates a storage area used for executing the process (150) in the external memory 201 such as a hard disk. The patch application area 211 in the process use area 210 indicates a storage area for correction data obtained from the patch data 141.

  When the patch data 141 is applied to the process use area 210, function correction data, that is, a correct function program is secured in the patch application area 211. Also, the process (150) is modified so that the call destination when the process (150) calls the target function is the patch application area 211. When it is confirmed that the function in the patch application area 211 functions normally, the function is overwritten on the process (150) in the external memory 201. Thereby, the patch process ends normally.

  In this way, by applying the patch data 141 to the process (150) on the main memory 202, the active server 110 and the stand-alone server 130 can perform patch processing without interrupting the service.

  The operation of the server system 100 will be described with reference to the flowchart shown in FIG. The maintenance server 140 transmits the patch data 141 of the process (150) to each server of the duplexing unit 101 and the unifying unit 102 (step S1_1). Each server (110/120/130) receives the patch data 141 from the maintenance server 140 (steps S2_1 / S3_1 / S4_1).

  The single-system server 130 of the single unit 102 starts patch processing and applies the received patch data 141 to the process 150a of its own device (step S4_2). The patch processing for the process (150) is as described with reference to FIG.

  As a result, if the application of the patch data 141 is successful (step S4_3: Yes), the external memory is overwritten (step S4_4), and the maintenance server 140 is notified that the patch processing has been completed normally (step S4_5). . If the application of the patch data 141 fails (step SS4_3: No), the maintenance server 140 is notified that the patch processing cannot be properly executed, and the process ends (step S4_5).

  On the other hand, in the duplexing unit 101, first, the standby server 120 applies the patch data 141 from the maintenance server 140 to the process 150b of its own device (step S3_2). If the application fails (step S3_3: No), this is notified to the active server 110 (step S3_4), and the maintenance server 140 is notified that the patch processing cannot be executed properly (step S3_4). S3_7).

  When the application of the patch data 141 is successful (step S3_3: Yes), the standby server 120 notifies the active server 110 to that effect (step S3_5). Then, after overwriting processing to the external memory (step S3_6), the maintenance server 140 is notified that the patch processing has ended normally (step S3_7).

  After receiving the patch data 141, the active server 110 waits for a notification from the standby server 120 regarding the success or failure of the application. As a result, when the received notification indicates application failure (step S2_2: “failure”), the patch data 141 from the maintenance server 140 is invalidated and the process is terminated.

  Also, when the standby server 120 notifies that the application has been successful (step S2_2: “success”), the active server 110 applies the patch data 141 to the process 150a of its own device (step S2_3), and the external server A memory overwriting process is performed (step S2_4), and the patch process is terminated.

  When the maintenance server 140 receives notification from the stand-alone server 130 and the standby server 120 about the result of the patch processing by the distributed patch data 141 (step S1_2), the maintenance server 140 has a predetermined format to notify the maintenance manager of the notification content. To display the screen (step S1_3). Thereby, the administrator can grasp the status of patch processing in each unit.

  According to the embodiment described above, since the maintenance server 140 is configured to distribute the patch data 141 of the process (150) to each server (110/120/130), the patch can be used in a system in which server operation modes are mixed. Processing can be performed smoothly.

  Further, the duplexing unit 101 uses the result of applying the patch data 141 in the standby server 120 to determine whether or not to perform the patch processing of the active server 110, so that the active server 110 that is providing the service. It is possible to make it difficult to influence the operation. That is, even if there is a problem in the patch data 141, it is detected by the standby server 120, so that service interruption can be avoided.

  Next, another embodiment of the present invention will be described with reference to the flowchart shown in FIG. The maintenance server 140 transmits the patch data 141 only to the standby server 120 (step S11_1). The standby server 120 that has received this (step S13_1) applies the patch data 141 to the process 150b of its own device (step S13_2), as in the above-described embodiment.

  As a result, when application of the patch data 141 has failed (step S13_3: No), the standby server 120 notifies the maintenance server 140 to that effect (step S13_4). If the application of the patch data 141 is successful (step S13_3: Yes), this is notified to the maintenance server 140 (step S13_5), the external memory is overwritten (step S13_6), and the patch process is terminated.

  The maintenance server 140 receives a notification from the standby server 120 regarding the application result of the patch data 141 (step S11_2). When the notification that the application has been successful is received (step S11_3: “success”), the maintenance server 140 updates the patch data to the active server 110 of the duplex unit 101 and the single server 130 of the single unit 102. 141 is transmitted (step S11_4).

  When the active server 110 and the stand-alone server 130 receive the patch data 141 from the maintenance server 140 (steps S12_1 and S14_1), they apply it to their own processes (150a and 150c) (steps S12_2 and S14_2). Overwrite processing to the external memory is performed (steps S12_3 and S14_3). As a result, the patch processing is completed on a server other than the standby server 120.

  On the other hand, when the notification content from the standby server 120 to the maintenance server 140 is “failure” (step S11_3: “failure”), the maintenance server 140 indicates that the current patch data 141 is inappropriate for the process (150). Is displayed (step S11_5). In this case, the patch data 141 is not transmitted to other servers (110, 130).

  As described above, in this embodiment, the application result of the standby server 120 is used to determine whether or not the patch data 141 can be transmitted to other servers (110, 130), that is, a server group that is providing a service. If the patch data 141 is unsuitable for the process (150), the service is not transmitted to the server group that provides the service, so that it is difficult to affect the service and Data transfer is made more efficient.

It is a block diagram which shows the system configuration | structure of embodiment of this invention. It is explanatory drawing for demonstrating the patch process in embodiment. It is a flowchart which shows the operation | movement procedure of embodiment. It is a flowchart which shows the operation | movement procedure of other embodiment of this invention.

Explanation of symbols

100 server system 101 duplex unit 102 single unit 110 active server 120 standby server 130 stand-alone server 140 maintenance server 141 patch data 150a, 150b, 150c process 201 external memory 202 main memory 210 process use area 211 patch application area

Claims (8)

A duplex unit provided with an active server device for executing a process and a standby server device for standby operation for the active server device, and a single device provided with a server device for executing the same process as the process And a maintenance server device connected to the single unit and the duplex unit,
The maintenance server device has means for transmitting patch data for correcting the process to the single unit and the duplex unit. The standby server device has means for notifying the operational server device of success or failure of patch processing by the patch data;
The active server device includes means for determining the content of the notification from the standby server device, and executes patch processing of the own device based on the patch data when the notification indicates successful patch processing. The server system according to claim 1. The standby server device has means for notifying the maintenance server device of success or failure of patch processing by the patch data,
The maintenance server device includes means for determining the content of the notification from the standby server device, and when the notification indicates a successful patch process, the patch is sent to the active server device and the server device of the single unit. The server system according to claim 1, wherein data is transmitted.   Each server device executes the process stored in the main memory of its own device by applying the patch data, and when the application of the patch data is successful, the part of the process in the external memory of the own device is changed to the patch data. The server system according to any one of claims 1 to 3, wherein the server system is overwritten. A duplex unit provided with an active server device for executing a process and a standby server device for standby operation for the active server device, and a single device provided with a server device for executing the same process as the process The maintenance server device connected to the
A patch processing method, wherein patch data for correcting the process is transmitted to the single unit and the duplex unit. The standby server device notifies the operational server device of success or failure of patch processing by the patch data,
The active server device determines the content of the notification from the standby server device, and executes the patch processing of the own device based on the patch data when the notification indicates a successful patch processing. 5. The patch processing method according to 5. The standby server device notifies the maintenance server device of success or failure of patch processing by the patch data,
The maintenance server device determines the content of the notification from the standby server device, and transmits the patch data to the active server device and the server device of the single unit when the notification indicates a successful patch process. The patch processing method according to claim 5, wherein:   Each server device executes the process stored in the main memory of the own device by applying the patch data, and when the application of the patch data is successful, the part of the process in the external memory of the own device is the patch data. The patch processing method according to claim 5, wherein the patch processing method is overwritten.