JP2010003022A - File updating method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a file updating method for introducing a virtual server environment for operating a plurality of logical servers in one physical server by a telecommunication server, and for preventing any influence from being given to any other virtual server on the identical physical server in performing file update accompanied by system switching in a virtual server environment. <P>SOLUTION: This file updating method includes: a step (S1) for developing a new load module according to the input of a file update preparation command; a step (S3) for creating the logical server of a new load module by an operation system according to the input of a file update start command, and for performing the first half of initialization (S2); a step (S5) for ending old file side processing (S4) according to the input of a file update system switching command, and for performing the last half of the initialization of the logical server of the new load module; and a step (S7) for performing the end processing of a standby system object logical server according to the input of a file update approval command (S6), and for creating the logical server of the new load module, and for performing initialization. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a software file update method for a telecom server, and is particularly suitable for use as a software file update method for control software for a communication server that requires service durability.

  In recent years, a telecom server, which is a server system having a function of a telephone exchange, has been used. In such a telecom server, a OS equipped with a symmetric multiprocessor (SMP) is mounted on a server including a multicore CPU, and a call processing program and other application programs are operated on the OS.

  FIG. 1 is an arrangement configuration diagram of a telecom server targeted by the present invention. In the figure, server # 11 is applied as a network node server, and server # 21 and server # 22 are also a network node server, an IP phone, a PC terminal, a PDA (Personal Digital Assistant), and the like. The IP network 100 is a peer-to-peer communication network based on an IP address, and data is delivered to a target node through various paths based on header information of the IP packet.

  The logical server # 01 and the logical server # 02 in the server # 11 are provided as functions that effectively utilize the CPU # 01 and the CPU # 02, and can be initialized in units of logical servers, and file replacement can be performed. Yes, and independence is improved for each logical server.

  In such an environment, a plurality of logical servers are operated in one physical server, and there is a need to update software files in the single system and the double system.

  In the conventional technique, when the software file of the exchange is updated, a new software file is stored in advance in the spare system of the control system device that performs call processing control of the exchange for the purpose of shortening the service interruption time. The system is switched to the new active system, call information for the connected call is taken over from the old active system to the new active system, or the processing of the existing call is continued in the old active system (for example, see Patent Document 1).

Another prior art includes a host computer that executes a plurality of virtual machines, a system resource that is organized into a plurality of resource groups, and a resource controller that associates a resource group with a virtual machine that operates on the host computer. Multiple virtual machines can be executed on the same host computer (see, for example, Patent Document 2).
Japanese Patent Laid-Open No. 06-319163 Japanese Patent Application Laid-Open No. 07-53447

  However, in the prior art (Patent Document 1), a file update for system switching is performed on a set of physical servers having a duplex configuration of an active system and a standby system, and a plurality of logical servers are allocated to one physical server. It is not the content that takes a configuration with a server. For this reason, the invention is not an invention directed to server virtualization technology that allows a plurality of logical (virtual) servers to be executed on a telecom server that has changed drastically in recent years, and a file without system switching. Cannot update.

  In the prior art (Patent Document 2), an example of a virtual machine is described. In this example, when a storage system is applied to a virtual machine, a storage device is allocated to each virtual machine and released. Is merely described, and the virtual machine is not applied to the communication field.

  The present invention introduces a virtual server environment in which a plurality of logical servers are operated in one physical server by a telecom server, and when updating files with system switching in the virtual server environment, the same physical It is an object to provide a file update method that does not affect other virtual servers on the server.

  A first invention for solving the above-mentioned problem is a file update of a duplex communication system server having a plurality of logical servers in one physical server, and each logical server having an application file and an individual process control unit. A method of deploying a new load module by inputting a file update preparation command, a step of generating a logical server of the new load module in the active system by inputting a file update start command, and performing the first half of the initial setting, When the file update system switch command is entered, the processing on the old file side is terminated, and the initial setting of the new load module logical server is completed, and when the file update approval command is entered, the standby target logical server is terminated. Steps to process, create a new load module logical server, and initialize It contains.

  According to the first aspect of the present invention, in the logical server environment, the service is interrupted for a very short time or not at all, and the connection state of the connected call is maintained by taking over the data, and the application software file is replaced. It is possible to provide a file switching method that makes it possible.

A second invention is the file update method according to the first invention, wherein:
When an abnormal state is detected during the file update, the file update state is set to unexecuted by inputting the file update forced release command, and the termination process and the extinction process are performed on the logical server running on the new file. Do.

  According to the second aspect of the present invention, it is possible to provide a file update method capable of reducing the burden on the maintenance person because the file update can be easily forcibly canceled when an abnormal state occurs during the file update.

A third invention is the file update method according to the first invention,
When an abnormal condition is detected during the file update, the file update status is set to file update aborted by entering the file update abort command, and new operation is performed by switching the active and standby systems of the redundant communication server. An operation stop signal is notified from the operation system side to perform termination processing, and the logical server is restarted after detecting the disappearance of the process control unit.

  According to the third aspect of the present invention, it is possible to provide a file update method that can reduce the burden on the maintainer because the file update can be easily canceled when an abnormal state occurs during the file update.

4th invention is the file update method as described in 1st invention,
When the file update preparation command is input, a plurality of the same type logical server groups having the same application file are collectively replaced with a new file.
According to the fourth aspect of the invention, while the application of one logical server is replaced in the logical server environment, the service can be continued in the other logical server and not every logical server, It is possible to provide a file update method that makes it possible to reduce the burden on maintenance personnel by exchanging files in a batch.

As described above, according to the file update method of the present invention,
(1) In a logical server environment, the service interruption is extremely short or not at all, and the connection state of the connected call can be maintained by taking over the data, and the application software file can be replaced.
(2) In the file update method without system switching, it is possible to replace the application software file in the logical server environment without affecting the logical server that does not replace the file at all.
(3) In the logical server environment, when the same application software is running on a plurality of logical servers for the purpose of load distribution or the like, while the application of one logical server is replaced, the service is performed on the other logical servers. Can continue.
(4) In the logical server environment, if the same application software is running on multiple logical servers for the purpose of load distribution, etc., maintenance is performed by replacing the files in batches instead of for each logical server. The burden on the user can be reduced.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 2 is a basic explanatory diagram of an existing file update method according to an embodiment of the present invention. The file update procedure in the redundant server system is described below.
[File update procedure during normal operation]
In (a), the server blade # 0 is operating as the active system and the server blade # 1 is operating as the redundant system, and the NGN basic middleware files and user application files of the active system and the standby system are the active files. You are driving with (old file).
In (b), when a file update is started, a new file is developed in the NGN basic middleware file and user application file of server blade # 1 standby system, and the operation is performed using the new file in the standby system, and separate operation is started. To do.
In (c), the server blade # 1 becomes the active system by the file update switching, and the NGN basic middleware file and the user application file start to operate as new files. On the other hand, the server blade # 0 stops the active system, enters an emergency operation waiting state, and holds the old file in preparation for switchback.
In (d), when the file update is confirmed (the operational status of the new file is normal), the old operational system is also started up with the new file and operated as a standby system, and the server blade # 1 and the operational system are switched to the duplex operation state. Become.
[File update procedure when an abnormal condition is detected]
If an abnormality is detected in the operation of the new file in (b) above, “file update abort” is set. That is, the standby system is restarted with the old file, and the operation is started as the standby system.

  If an abnormality is detected in the operation of the new file in the new active system in (c) above, “file update update cancellation” is set. That is, the system that has been operated using the new file starts up with the old file and starts operation as a standby system.

  FIG. 3 is an explanatory diagram showing a single file update procedure of the logical server in one embodiment of the present invention, and FIG. 4 is a flowchart of FIG. The single file update procedure of the logical server will be described below with reference to FIGS.

S1. By turning on the file update preparation to the logical server 1 ₁₁ of the subject, to deploy the new load module.

  S2. File update start is input.

S3. To generate a new logical server 1 _'11 of the new load module in a production system, perform the initial setting first half.

S4. When the file update switch is turned on, the termination processing of the old logic server 1 ₁₁ of the old load module.

S5. Carry out the second half of the initial setting of the new logical server 1 _'11 of the new load module.

  S6. File update approval is entered.

S7. The termination processing of the target logical server 1 ₂₁ of the standby system is performed, a new logical server 1 ′ ₂₁ of the new load module is generated, and initial setting is performed.

  FIG. 5 is an explanatory diagram showing an example of a file update preparation procedure according to the present invention.

S1. Request the file update preparation to the process control individual unit 12 ₁₁ .

S2. To notify that the system management unit 2 ₁ file update setup is requested.

S3. Whole received a notification manager 2 _1, file update preparation the issuing logical server 1 ₁₁ to notify that it has been entered.

S4. Process control individual unit 12 ₁₁ receives the notification, the new file registration directory (dir) compressed file that is located underneath deploy by copying down new directory (dir).

S5. Process control individual unit 12 ₁₁ notifies the system management unit 2 ₁ a deployment end of the file.

S6. System management unit 2 _1, which has received the notification notifies the deployment ended process control individual unit 12 _11.

S7. Process control individual unit 12 ₁₁ that has received the notification transmits that the file update preparation is input to the process control individual unit 12 ₂₁ of the standby system.

  S8-S12. Processing equivalent to S2-S6 is performed in the standby system.

S13. A response is returned from the standby process control individual unit 12 ₂₁ to the active process control individual unit 12 ₁₁ .

  FIG. 6 is an explanatory diagram showing an example of the same-type server file update preparation procedure according to the present invention.

S1. Request the file update preparation to the process control individual unit 12 ₁₁ .

S2. To notify that the system management unit 2 ₁ file update setup is requested.

S3. Whole received a notification manager 2 _1, a compressed file that is placed under the new file registration directory, copy under the new file registration directory of the same type logical server.

S4. After copying system management unit 2 ₁ notifies that the same type server file update preparation is input to the target logical server.

S5. Process control individual unit 12 ₁ receives the notification, to expand and copy the compressed files that are placed under the new file registration directory under the new directory.

S6. Process control individual unit 12 ₁₁ notifies the system management unit 2 ₁ a deployment end of the file.

S7. System management unit 2 ₁ that has received the notification, after receiving the expanded termination notification from all process control individual unit, and notifies the expanded exit request issuing process control individual unit 12 _11.

S8. Process control individual unit 12 ₁₁ that has received the notification transmits that allogeneic server file update preparation is input to the process control individual unit 12 ₂₁ of the standby system.

  S9-S14. Processing equivalent to S2 to S7 is performed in the standby system.

S15. A response is returned from the standby process control individual unit 12 ₂₁ to the active process control individual unit 12 ₁₁ .

  FIG. 7 is an explanatory diagram showing an example of a file update start procedure (no system switching) according to the present invention.

S1. Requests the process control individual unit 12 ₁₁ to start file update.

S2. To notify that the system management unit 2 ₁ file update start has been requested.

S3. System management unit 2 _1, the issuing logic server 1 ₁₁ to a file update start to notify that it has been entered.

S4. Process control individual units 12 ₁₁ which has received the notification, to notify the update query events to file update query event notification registered users.

S5. Process control individual unit 12 _11, and transmits a termination process in the system management unit 2 _1.

S6. Processing system management unit 2 ₁ receiving is completed, the target logical server from the new path of the target logical server launching an initial installation type "duplicated switching without file updating initialization".

S7. "Redundant switching without file update the initial setting" new logical server 1 _'11 that was started in, set the current path under the new file directory, by executing the initial setting up to the middle of the initial setting, the old logical server 1 ₁₁ The shared memory data can be taken over.

S8. And it notifies the system management unit 2 ₁ initialization termination.

S9. System management unit 2 ₁ sends a process end to the issuing logical server 1 _11.

  FIG. 8 is an explanatory diagram showing an example of data transfer at the start of file update according to the present invention. (1) The top address of the shared memory area for key management is acquired.

  (2) The new logical server uses a different aspect from that used by the old logical server.

(3) Of the key management data used by the old logical server, the shared memory control and exclusive control function data to be inherited are copied to the key management data of the new logical server.
Key management data corresponding to other functions that are not to be inherited, such as thread control functions and inter-process communication functions, are initialized.

  (4) Shared memory management acquires the ID of the shared memory for shared memory management using key management and attaches it. Since data is shared with the old logical server, resource initialization and semaphore regeneration are not performed.

  (5) Exclusive control acquires the ID of the shared memory for exclusive control using key management, and performs attachment. Since the data is shared with the old logical server, the resource is not initialized.

  (6) An application process is generated.

  (7) The application process performs only the initial setting for communication with the orbital source, and transmits the application promise generation success to the activation source process.

  FIG. 9 is an explanatory diagram showing an example of a file update start procedure (with system switching) according to the present invention.

S1. Requests the process control individual unit 12 ₁₁ to start file update.

S2. To notify that the system management unit 2 ₁ file update start has been requested.

S3. System management unit 2 ₁ notifies that all logical server file update start is input.

S4. Process control individual units 12 ₁₁ which has received the notification, to notify the update query event if the file being updated to the file update query event notification registered users.

S5. Process control individual unit 12 ₁₁ transmits the process end to the system management unit 2 _1.

S6. System management unit 2 _1, after receiving responses from all of the process control individual unit that sent the message, and sends the processing completion to the requesting process control individual unit 12 _11.

S7. Requesting process control individual unit 12 ₁₁ requests the standby disconnect the system management unit 2 ₁ to the cluster system.

S8. System management unit 2 ₁ inquires the disconnect enable / disable the standby to all logical server 1 _11, 1 _12.

S9. Process control individual unit 12 _11, and returns whether the standby disconnecting the system management unit 2 _1.

S10. System management unit 2 ₁ requests the standby disconnect the cluster system 3 _1.

S11. Cluster system 3 ₁ by disconnecting the standby, the standby system stop signal from the cluster system 3 ₂ standby is notified to the system management unit 2 _2.

S12. System management unit 2 ₂ transmits a standby stop for all logical server.

S13. If the file is being updated, the process control individual unit ₁₂₂₁ sets a file update flag and performs an end process to switch the file path of the working directory to the new file storage directory. If the file is not being updated, only end processing is performed.

S14. Process control individual unit 12 _21, and transmits processing completion to the system management unit 2 _2.

S15. The system management unit 2 _2, after waiting a completion notification from all of the process control individual unit, and restarts to perform the standby system startup in the new file logical server 1 ₂₁ in the file update.

S16. In operation system, the process control individual unit 12 ₁₁ detects a link establishment between the operation system / standby system.

S17. Requesting process control individual unit 12 ₁₁ inquires whether the new file launch standby for the entire management unit 2 ₁ is completed.

S18. System management unit 2 ₁ queries the entire process control individual unit, and notifies the result to the requesting logical server 1 _11.

  FIG. 10 is an explanatory diagram showing an example of a file update system switching procedure (no system switching) according to the present invention.

S1. Requests file update system switching to the individual process control unit 12 ₁₁ .

S2. The system management unit 2 ₁ is switched file update system notifies the input.

S3. System management unit 2 ₁ receives the notification, the logical server 1 ₁₁ commands input source, and notifies that the switch file updating system is input.

S4. Logical server 1 ₁₁ receives the notification, to file updating state.

S5. Process control individual unit 12 ₁₁ notifies the completion of the process in the system management unit 2 _1.

S6. System management unit 2 ₁ sends a response issuing source logical server 1 ₁₁ in the S2.

S7. Command issuing logical server 1 ₁₁ requests a new file launch for the entire management section 2 _1.

S8. Command issuing logical server 1 ₁₁ inquires the disappearance whether the own logical server 1 ₁₁ to the system management unit 2 _1.

S9. System management unit _{2 1} sends a response of the S8.

S10. Upon receiving the response, the logical server 1 ₁₁ performs termination processing, performs release processing for resources that are not to be inherited, switches the file path to a new file, and then performs erasure processing for no duplexed file update switching. switch the ₁₁ to the new file, and then perform the disappearance processing for without switching duplicated file update, to terminate the logical server 1 _11.

S11. System management unit 2 ₁ detects the disappearance of the process control individual unit 12 _11.

S12. To notify the initial set continued in the new logical server 1 _'11.

S13. New logical server 1 _'11 continues the initialization from the middle initial setting raises the application (APL) process. In the latter half of the initial setting, the key information of the shared memory that is not the target of inheritance is deleted.

S14. And it notifies the system management unit 2 ₁ a completion of the initial settings.

  S15. The active server that was started up with “file update initialization without duplexing switching” at the start of file update does not perform inter-system matching processing until file update approval is entered.

  FIG. 11 is an explanatory diagram showing an example of data takeover at the time of file update system switching according to the present invention.

  In the latter half of the initial setting, the key information of the shared memory that is not the target of inheritance is deleted.

  FIG. 12 is an explanatory diagram showing an example of a file update system switching procedure (with system switching) according to the present invention.

S1. Requests the process control individual unit 12 ₁₁ to switch the file update system.

S2. The system management unit 2 ₁ is switched file update system notifies the input.

S3. System management unit 2 ₁ which has received the notification notifies that all logical server 1 _11, 1 ₁₂ to switch file update system is input.

S4. Logical server 1 ₁₁ which has received the notification, is a file being updated, in the case of there data conversion, to send a file update query events to the user.

S5. Process control individual unit 12 ₁₁ notifies the completion of the process in the system management unit 2 _1.

S6. System management unit 2 ₁ sends to the issuing logical server 1 ₁₁ receives a notification from all the logical server a response to switching file update system.

S7. Requesting logical server 1 ₁₁ requests the switching system for the entire management section 2 _1.

S8. Discharging the application (APL), within the process control individual unit 12 ₁₁ performs the stopping of the communication unit.

S9. System management unit 2 ₁ requests the switching system to the cluster system 3 _1.

S10. Cluster system 3 ₁ by switching the system, operation system switching signal from the cluster system 3 ₂ standby is notified to the system management unit 2 _2.

S11. Operation system standby system management unit 2 ₂ which has received the switching signal, the total logical server 1 _21, 1 _22, and transmits the switching operation system, the logical server 1 _21, 1 ₂₂ is the active system.

S12. System management unit 2 ₁ is the old active system, from the cluster system 3 ₁ receives the operation system stop signal.

S13. System management unit 2 ₁ sends an operation system stop notification to each process control individual unit 12 _11, 12 _12.

S14. The process control individual unit 12 ₁₁ that has received the operation system stops, 12 ₁₂ transmits processing completion to the system management unit 2 ₁ shifts the state to the emergency operation waiting.

  FIG. 13 is an explanatory diagram showing an example of a file update approval procedure (no system switching) according to the present invention.

S1. To request a file update approval to process control individual units 12 _'11.

S2. Process control individual unit 12 _'11 to the standby system 1 _21, and notifies that the file update approval is input.

S3. Process control individual unit 12 ₂₁ receives the notification, to notify that the system management unit 2 ₂ file update approved inputted.

S4. From system management unit 2 ₂ to the notification source logical server 1 _21, and notifies that the file update approval is input.

S5. Receiving the notification, the individual process control unit ₁₂₂₁ sets the file updating status.

S6. And it notifies the completion of processing to the system management unit 2 _2.

S7. System management unit 2 ₂ transmits a process end to the issuing logical server 1 _21.

S8. Exit notified of the received issuing logical server 1 _21, returns a response to the operational system.

S9. After returning the response, to the entire management unit 2 _2, to notify that it has returned the response to the operational system.

S10. Process control individual unit 12 ₂₁ inquires the disappearance possibility for the entire management section 2 _2.

S11. System management unit _{2 2} returns the response to the S10 to logic server 1 _21.

S12. Logical server 1 ₂₁ receives the response to the S10, after performing the annihilation process and end the process and disappear by switching the starting path to the new path.

S13. System management unit 2 ₂ restarts the logical server 1 _'21 disappeared. Logical server 1 _'21 is restarted collects the backup file.

S14. The process control of the active received a response from the standby system individual unit 12 _'11 notifies that to the target logical server file updating approved entered for the system management unit 2 _1.

S15. System management unit 2 _1, to the issuing process control individual unit 12 _'11 file update approval notifying that input.

S16. Logical server 1 _'11 which has received the notification, send the file update completion of the file update query event notification registered users, to collect the backup file.

S17. Logical server 1 _'11 collected the backup file, notifies the completion of processing to the system management unit 2 _1.

S18. System management unit 2 ₁ 'receives a response from the _11, file update approval issuing server 1' logical server 1 to _11, returns a response to the file update approval.

  S19. Inter-system matching processing is started upon establishment of a link with the standby server.

  FIG. 14 is an explanatory diagram showing an example of a file update approval procedure (with system switching) according to the present invention.

S1. Requests file update approval from the individual process control unit 12 ₂₁ .

S2. To the entire management unit 2 _2, informing you that the file update approval has been input.

S3. The system management unit 2 ₂ from all the logical server 1 _21, 1 ₂₂ file update approval notifying that input.

S4. Process control individual unit 12 ₂₁ receives the notification, performs backup collection.

S5. It notifies the system management unit 2 ₂ the backup collection completion.

S6. System management unit 2 _2, after receiving the notification from all target servers, returns a response to the requesting process control individual unit 12 ₂₁ with respect to the S2.

S7. Requesting process control individual unit 12 ₂₁ requests the disconnection of the standby server for the entire management section 2 _2.

S8. System management unit 2 ₂ requests the disconnection of the standby server to the cluster system 3 _2.

S9. Logical server 1 ₂₁ in the file update, after the processing in S5, with respect to the system management unit 2 ₂ inquires whether the notification to the application (APL).

S10. System management unit 2 ₂ After receiving a response from all of the process control individual unit sends a response to the S9.

S11. Logical server 1 _21, after receiving the above S10 is a response of the S9, transmits the file update completion to the file update query event notification registration user.

S12. Cluster system 3 ₂ by the S8 is by disconnecting the standby, the standby system stop signal from the cluster system 3 ₁ standby is notified to the system management unit 2 _1.

S13. System management unit 2 ₁ sends a standby stop for all logical server.

S14. Process control individual units 12 _11, if the file being updated, making a file update flag, a termination process is performed, it switches the file path to the new directory. If the file is not being updated, only end processing is performed.

S15. It notifies the end of the process to the system management unit 2 _1.

S16. The system management unit 2 _1, after waiting a completion notification from all of the process control individual unit, and restarts to perform the standby system startup in the new file logical server 1 _'11 in the file update.

  S17. After restarting, collect the backup file.

  FIG. 15 is an explanatory diagram showing an example of a file update forced release procedure (no system switching) according to the present invention.

S1. Requests the process control individual unit 12 ₁₁ to forcibly release the file update.

S2. The system management unit 2 _1, notifying that the file update forcibly released is input.

S3. From system management unit 2 ₁ to the issuing logical server 1 ₁₁ file update forcibly released notifies the input.

S4. Receiving the notification, the process control individual unit ₁₂₁₁ sets the file update status to “unupdated file”.

S5. Processing and notifies the system management unit 2 ₁ the completion.

S6. Logical server 1 ₁₁ notifies the stop processing end notification to the system management unit 2 _1.

S7. System management unit 2 _1, if logical server 1 _'11 running on the new file, and transmits an end notification.

S8. New file start logical server 1 _'11 which has received the notification, performs the annihilation process and end the process, the process disappears.

S9. System management unit 2 _1, 16 to transmit a response of the above-described Step S6 the issuing logical server 1 ₁₁ is an explanatory diagram showing an example of the file update Forced release procedure according to the present invention (with the system switching).

S1. Requests the process control individual unit 12 ₁₁ to forcibly release the file update.

S2. The system management unit 2 _1, notifying that the file update forcibly released is input.

S3. System management unit 2 ₁ from all the logical server 1 _11, 1 ₁₂ to file update forcibly released to notify that it has been entered.

S4. Process control individual unit 12 ₁₁ receives the notification, if the file being updated to set the file update status file updated unexecuted transfers the file update status file on the standby server 1 _'21.

S5. Processing and notifies the system management unit 2 ₁ the completion.

S6. System management unit 2 _1, waiting the notification from all target servers 1 _11, 1 ₁₂ and returns a response to the requesting process control individual unit 12 ₁₁ in the S2.

S7. Requesting process control individual unit 12 _11, to the system management unit 2 _1, and requests the disconnection of the standby server.

S8. System management unit 2 _1, to the cluster system 3 ₁ requests the disconnection of the standby server 1 _'21.

S9. Cluster system by disconnecting the standby, the standby system stop signal from the cluster system 3 ₂ standby is notified to the system management unit 2 _2.

S10. System management unit 2 ₂ transmits a standby stop for all logical server 1 _'21.

S11. Process control individual unit 12 _'21, which has received the standby system stops, if the file updating performed to set the file update status file updated unexecuted end processing, switches the path. If the file is not being updated, only end processing is performed.

S12. And it notifies the completion of processing the system management unit 2 _2.

S13. System management unit 2 ₂ waits for notification from all process control individual unit, performing the reboot.

S14. In a production system, with respect to the cluster system 3 _1, after performing the standby disconnection, and it sends the processing completion to the requesting process control individual unit 12 _11.

  FIG. 17 is an explanatory diagram showing an example of a file update cancellation procedure (no system switching) according to the present invention.

S1. It requests a file updating cancellation to the process control individual unit 12 _'11.

S2. The system management unit 2 _1, notifying that the file update cancellation is input.

S3. From system management unit 2 ₁ to the issuing logical server 1 _'11 file updating cancellation is notified that it has been entered.

S4. Process control individual part 12 _'11 which has received the notification sets the file update status during a file update cancellation.

S5. Processing and notifies the system management unit 2 ₁ the completion.

S6. System management unit 2 ₁ is returned to the command issuing source logical server 1 _'11 response to the S2.

S7. Requesting process control individual unit 12 _'11, with respect to the system management unit 2 ₁ requests the switching system.

S8. System management unit 2 ₁ requests the switching system to the cluster system 3 _1.

S9. Cluster system 3 ₁ by performing switching system, operating system switching signal from the cluster system 3 ₂ standby is notified to the system management unit 2 _2.

S10. Operation system switching signal the new active system system management unit 2 ₂ which has received the send operation system switching signal to all logical server 1 _21, 1 ₂₂ 1 _2, logical server 1 _21, 1 ₂₂ is operated based .

S11. The whole operation system stop signal from the cluster system 3 ₂ management unit 2 ₁ is received.

S12. System management unit 2 ₁ notifies the operation system stop for the process control individual unit 12 _'11.

S13. Process control individual unit 12 _'11 sets the file update status if file updating a file update unexecuted, performs an end process to change the start-up directory. If the file is not being updated, only end processing is performed.

S14. It notifies the end process throughout the administrative unit 2 _1.

S15. System management unit 2 ₁ performs restart and receive processing end notification from all of the process control individual unit.

  FIG. 18 is an explanatory diagram showing an example of a file update cancel procedure (with system switching) according to the present invention.

S1. It requests a file updating cancellation to the process control individual unit 12 _'21.

S2. The system management unit 2 ₂ notifies that the file update cancellation is input.

S3. From system management unit 2 ₂ All logical server file updating cancellation is notified that it has been entered.

S4. Process control individual unit 12 _'21 which has received the notification sets if file updating a file update status during a file update cancellation.

S5. Processing and notifies the system management unit 2 ₂ the completion.

S6. System management unit 2 ₂ returns a response to the requesting process control individual unit 12 _'21 with respect to the S2.

S7. Requesting process control individual unit 12 _'21, with respect to the system management unit 2 ₂ requests the switching system.

S8. System management unit 2 ₂ requests the switching system to the cluster system 3 _2.

S9. Cluster system 3 ₂ By performing the system switching, the operation system switching signal from the cluster system 3 ₁ standby is notified to the system management unit 2 _1.

S10. The whole operation system stop signal from the cluster system 3 ₁ management unit 2 ₂ receives.

S11. System management unit 2 ₂ notifies the operating system stops to the process control individual unit 12 _'21.

S12. Process control individual units 12 _'21, if the file being updated, the termination process by changing the startup directory. If the file is not being updated, only end processing is performed.

S13. It notifies the end process throughout the administrative unit 2 _2.

S14. System management unit 2 ₂ performs restart and receive processing end notification from all of the process control individual unit.

S15. Whole new active system received a switching operation based management unit 2 ₁ sends an operating system changes with respect to the total process control individual unit 12 _11, 12 _12.

S16. Process control individual unit 12 ₁₁ changes state to file update unexecuted state if the file being updated.

S17. Thereafter, the process control individual unit 12 ₁₁ performs the disappearance own process.

S18. System management unit 2 ₁ that has detected the disappearance of process control individual unit 12 _11, to restart the logical server 1 _11, 1 _12.

  The present invention relates to software for controlling a communication server (telecom server) that requires service sustainability, particularly application software for a single logical server in a virtual server environment in which a plurality of logical servers are operated in a physical server. This file can be used to replace the files while taking over the data, and can be used to reduce maintenance work of the communication device by replacing the application software files of a plurality of logical servers at once.

It is the arrangement block diagram of the telecom server which this invention makes object. It is a basic explanatory view of the existing file update method in one embodiment of the present invention. It is explanatory drawing which shows the single file update procedure of the logical server in one Embodiment of this invention. It is a flowchart of FIG. It is explanatory drawing which shows an example of the file update preparation procedure by this invention. It is explanatory drawing which shows an example of the homogeneous server file update preparation procedure by this invention. It is explanatory drawing which shows an example of the file update start procedure (without system switching) by this invention. It is explanatory drawing which shows an example of the data taking over at the time of the file update start by this invention. It is explanatory drawing which shows an example of the file update start procedure (with system switching) by this invention. It is explanatory drawing which shows an example of the file update type | system | group switching procedure (no system switching) by this invention. It is explanatory drawing which shows an example of the data taking over at the time of file update type | system | group switching by this invention. It is explanatory drawing which shows an example of the file update type | system | group switching procedure (with system switching) by this invention. It is explanatory drawing which shows an example of the file update approval procedure (no system switching) by this invention. It is explanatory drawing which shows an example of the file update approval procedure (with system switching) by this invention. It is explanatory drawing which shows an example of the file update forced cancellation procedure (no system switching) by this invention. It is explanatory drawing which shows an example of the file update forced cancellation procedure (with system switching) by this invention. It is explanatory drawing which shows an example of the file update cancellation procedure (no system switching) by this invention. It is explanatory drawing which shows an example of the file update cancellation procedure (with system switching) by this invention.

Explanation of symbols

1 ₁₁ Operational logical server # 1
1 _'11 operating system new logical server # 1
1 ₁₂ Operational logical server # 2
1 ₂₁ standby logical server # 1
1 _'21 standby new logical server # 1
1 ₂₂ standby logical server # 2
2 ₁ OS of the active system management unit 2 ₂ whole of the standby system management section 3 ₁ operating system cluster 3 ₂ standby cluster 4 ₁ operational system of the
42 ₂ OS
5 ₁ Active server basic part 5 ₂ Standby server basic part 11 ₁ Active application (APL)
11 ₂ Standby application (APL)
12 ₁ Process control individual unit for the active system 12 ₂ Process control individual unit for the standby system 100 IP network 110 Server # 11
111 Logical server 112 CPU
210 Server # 21
211 CPU
310 Server # 31
311 CPU

Claims (4)

A file update method for a duplex communication system server having a plurality of logical servers in one physical server, and each logical server having an application file and an individual process control unit,
The step of expanding the new load module by inputting the file update preparation command,
Generate a new load module logical server in the active system by inputting the file update start command, and perform the first half of the initial setting,
The process of terminating the processing of the old file side by inputting the file update system switching command and performing the latter half of the initial setting of the logical server of the new load module;
The process of terminating the standby target logical server by inputting the file update approval command, generating a logical server for the new load module, and initial setting,
A file update method comprising: The file update method according to claim 1,
When an abnormal state is detected during the file update, the file update state is set to unexecuted by inputting the file update forced release command, and the termination process and the extinction process are performed on the logical server running on the new file. A file update method characterized by performing. The file update method according to claim 1,
When an abnormal state is detected during the file update, the file update state is set to file update aborted by inputting the file update abort command, and new operation is performed by switching the active and standby systems of the redundant communication server. An operation system stop signal is notified from the operation system side to perform termination processing, and the logical server is restarted after the disappearance of the process control unit is detected. File update method. The file update method according to claim 1,
A file update method characterized in that when a file update preparation command is input, a plurality of logical server groups having the same application file are collectively replaced with a new file.

Images