KR100316841B1 - Method for restarting an agent and apparatus thereof - Google Patents

Abstract

The present invention can minimize the construction time for the management object (MO) and the resulting inter-processor communication (IPC) load of the communication network component (exchange) when re-executing the management agent (Agent) of the communication management network (TMN) It is to provide a method and apparatus for reexecuting a management agent. In the method according to the present invention, in a communication management network in which the information structure is divided into a manager, a management agent, and a communication network element, when a redo of the management agent is required, the management agent itself initializes the basic management object. ; Initializing an interface mode between a management object managed in a separate area from the basic management object in the management agent; The management agent reads only the changed portion of the management agent for all the management objects of the management agent during the period of execution interruption; Analyzing the read content to update the corresponding management object; When the construction of all managed objects is completed, the management agent performs a step of reporting completion of redo to the administrator. Therefore, it is possible to minimize the management object construction time in the management agent and the communication load on the communication network components.

Description

METHOD FOR RESTARTING AN AGENT AND APPARATUS THEREOF}

The present invention relates to a management agent (or remote network management agent) in a telecommunication management network (hereinafter, referred to as TMN), and in particular, when restarting the management agent (Network Element, The present invention relates to a rerun method and apparatus for minimizing the construction time and load of the NE for Managed Objects (hereinafter, referred to as NE).

As operation systems or NEs that manage telecommunications networks use specific information models and interfaces for each supplier, it becomes more difficult to integrate and manage at the network-wide level while maintaining interoperability as the telecommunications network becomes larger. have. Accordingly, it was required to establish a communication network operation management system that can cope with various types of operating systems while overcoming various and heterogeneous NEs.

TMN was first proposed in 1988 by the Consultative Committe International Telegraph and Telephone (CCITT) in response to the above-mentioned requirements. It is an independent type of network with an interface function. The information structure of the TMN is composed of a manager, a management agent, and a network element corresponding to a network resource (Network Element, hereinafter abbreviated as NE).

In order to manage the network of the NE remotely, the above-described management agent manages the MO corresponding to the managed entity existing in the NE in a tree structure, and informs the manager of the management information. It serves to instruct the management operation to be taken by the management agent based on the management information notified from the. In other words, the manager supports remote object management and general management information structure management functions, and the management agent provides management information to the manager in the form of a management object, and the manager analyzes when the message is delivered to the management object located in a specific management agent. Take the appropriate management action.

The redo action of these management agents is either a failure on their own or at the request of an administrator. However, in order to build a MO for management information that must be provided to the administrator when redo is performed, the management agent must refer to the NE's database for each MO.

1 is a block diagram illustrating a relationship between an existing management agent 100 and an exchange 110 corresponding to an NE. The management agent 100 includes a management agent processor 101 and all MOs corresponding to management information. Memory 103 stored in a structure, a Central Control Processor (CCP) interface unit 105 for matching processing with the switch 110, the failure detection unit for detecting the occurrence of failure in the management agent 100 107, the switch 110 includes a management agent interface unit 111, a CCP 112 and a database 113 corresponding thereto, and n line matching processors (ASPs 0 to n) ( 114_0 to 114_n, corresponding databases (abbreviated as DB hereinafter) 115_0 to n, location register processor (hereinafter referred to as LRP) 116, and corresponding DB 117 It consists of.

Referring to Figure 1 describes the existing re-execution process for the management agent 100 as follows.

First, a request for rerun due to a failure of itself is made by the failure detection unit 107. That is, the failure detecting unit 107 checks the operation state of the management agent processor 101 in the management agent 100 or transmits specific data to the exchange 110 through the CCP interface unit 105 and then to If it is recognized that a failure has occurred according to the result of checking whether the response message has been received, the redo request signal is sent to the management agent processor 101. If a redo is requested from the manager not shown to the management agent 100, the management agent 100 is directly authorized to the management agent processor 101. The redo request for the management agent 100 generated from the manager not shown is generally generated when the manager (not shown) interrupts the execution of the corresponding management agent 100 and tries to execute it again.

The management agent processor 101 may, if re-execution is requested from the failure detecting unit 107 or an administrator not shown, to reconstruct the MO of all management information to be provided to the administrator (not shown) in the memory 103. The log data request signal corresponding to the class is transmitted to the exchange 110 through the CCP interface 105.

The exchange 110 transmits the log data request signal transmitted from the management agent 100 to the CCP 112 through the management agent interface unit 111, and the CCP 112 analyzes the received log data request signal. The management agent 100 determines the processor in which the desired management entity exists. And it requires the management entity that corresponds to the identified processor. Accordingly, the corresponding processor reads the requested management entity from the database and transmits it to the CCP 112. For example, as a result of analysis of the CCP 112, when the corresponding management entity exists in the ASP 0 (114_0), the CCP 112 requests the management entity corresponding to the ASP 0 (114_0), ASP 0 (114_0) ) Reads the requested management entity from the corresponding database 115_0 and sends it to the CCP 112.

The CCP 112 transmits the management entity received from the corresponding processor to the management agent 100 through the management agent interface unit 111 as log data, and the management agent 100 transmits the received log data to the CCP interface unit ( 105 is stored in the memory 103 in the form of MO.

The above-described process is repeatedly performed until all the MOs are built in the memory 103.

As described above, in order to rebuild all the MOs of the management agent 100 by the redo request, the MO has to go through the process of accessing the corresponding database in the exchange for each MO. The problem is that the management service is delayed by that much. In addition, the exchange 110 also provides inter-processor communication (hereinafter abbreviated as IPC) to provide log data of the management entity corresponding to the management agent 100 until all MOs of the management agent 100 are rebuilt. Since messages continue to be generated, the IPC load in the exchange is increased due to re-execution, resulting in a decrease in processing efficiency in the exchange.

The present invention has been made to solve the above problems, and when re-executing the management agent (Agent) of the Telecommunication Management Network (hereinafter referred to as TMN), the construction time for the managed object and accordingly It is an object of the present invention to provide a method and apparatus for reexecuting a management agent that can minimize an IPC load of an exchange.

In order to achieve the above object, the management agent redo method according to the present invention is a communication management network in which an information structure is divided into a manager, a management agent, and a communication network element, and when a redo of the management agent is required, the basic management object Initializing itself for the management agent; Initializing an interface mode between a management object managed in a separate area from the basic management object in the management agent; The management agent reads only the changed portion of the management agent for all the management objects of the management agent during the period of execution interruption; Analyzing the read content to update the corresponding management object; When the construction of all the managed objects is completed, the management agent is characterized in that it comprises the step of reporting the completion of the redo to the administrator.

In order to achieve the above object, a management agent redo apparatus according to the present invention is a communication management network in which an information structure is operated by a manager, a management agent, and a communication network component structure, and when a redo is executed, the management agent may initialize itself. A memory for storing managed objects; a database for storing managed objects, except for basic managed objects, among managed objects that should be built in the management agent when reexecuted; If a redo is requested, the default managed object to be stored in memory is initialized, the interface mode with the database is initialized, and only the changed information during the period in which the execution is stopped is performed. A management agent processor for receiving information from communication network components and updating memory and database information; A disk provided in the communication network component and storing information about management objects generated in the communication network component during a period in which execution is stopped; And a central control processor (CCP) provided in the communication network component and configured to read information corresponding to the arbitrary management object from a disk and to transmit the information to the management agent when the changed information about the management object is requested from the management agent. It features.

1 is a block diagram showing a relationship between an existing management agent and an exchange period,

2 is a block diagram of a part of a communication management network having a redo apparatus of a management agent according to the present invention;

3 is a flowchart illustrating an operation of a management agent in a method of reexecuting the management agent according to the present invention;

4 is an operation flowchart of the exchange side in the re-execution method of the management agent according to the present invention.

<Description of the symbols for the main parts of the drawings>

200: management agent 201: management agent processor

202: memory 203: database

204: failure detection unit 205: CCP interface unit

210: switch 211: management agent interface unit

212: Central Control Processor (CCP) 213: Disk

214: Line registration processor (ASP) 215: Position register processor (LRP)

220: Operator terminal

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a partial functional block diagram of a communication management network (hereinafter referred to as TMN) with a redo device for management agents in accordance with the present invention. FIG. 2 is a diagram of a management agent 200 and a network element (hereinafter referred to as NE). It consists of a corresponding switch 210 and an operator terminal 220 connected to the switch 210.

The management agent 200 includes a management agent processor 201, a memory 202 for managing a basic managed object (hereinafter referred to as MO), and a database 203 for managing all other MOs except the basic MO described above. A failure detection unit 204 for detecting a failure in the management agent 200 and a central control processor (CCP) interface unit 205 for matching with the switch 210. The exchanger 210 is provided with a new command of description (COD) through the management agent interface unit 211, the CCP 212, and the operator terminal 220 that match with the management agent 200. And a disk 213 for storing M_PODs (Mo_Printing Of Description) generated in the &lt; RTI ID = 0.0 &gt;), &lt; / RTI &gt; The above-described database 203 is composed of a disk or the like.

3 is an operation flowchart of the management agent side in the reexecution method of the management agent according to the present invention, and FIG. 4 is an operation flowchart of the exchange side in the reexecution method of the management agent according to the present invention.

2 to 4, the reexecution operation of the management agent according to the present invention will be described as follows.

First, a case in which a redo request signal is generated from the failure detection unit 204 to the management agent processor 201 is a situation as described with reference to FIG. 1. That is, according to the result of checking whether the operation state of the management agent processor 201 is normal and the result of checking whether the CCP interface unit 205 is operating normally, whether the corresponding management agent 200 is identified or not is detected. When it is recognized that a failure has occurred, a redo request signal is generated to the management agent processor 201. The manager (not shown) also generates a redo request signal to the management agent 200 in the same situation as mentioned in FIG. 1.

When the redo request signal is applied to the management agent processor 201 by the above-described situation, the management agent processor 201 proceeds to step 301 to initialize the basic MO managed by the memory 202. Basic MOs are those that the management agent 200 does not need to be initialized through interworking with the exchange 210, and is frequently used by an administrator (not shown) or received from the exchange 210 after the management agent 200 is re-executed. MOs may be included. As a specific example, the performance management MO of the exchange 210 corresponds to this. The number of such base MOs is about 1 to several tens.

After the base MO is initialized, the management agent processor 201 proceeds to step 302 to interface with the database 203 so as to quickly provide the administrator (not shown) with the MO stored in the database 203. Initialize That is, a message is transmitted between the manager (not shown) and the management agent 200 in the form of a common management information service (sometimes referred to as CMIS), while the management agent processor 201 and the database ( Since messages are transmitted in a structured query language (SQL) format between 203), a format conversion mode is initialized to enable fast data transfer between them.

The management agent processor 201 then proceeds to step 303 where the management agent 200 manages in the base MO and the database 203 managed by the memory 202 from the time when the management agent 200 stopped execution to the present time. The MMC (Man Machine Command) log data (Log Data) request signal is transmitted to the switch 210 through the CCP interface unit 205 to receive the changed contents of the MO.

The MMC log data request signal transmitted through the CCP interface unit 205 includes a time point and a current time when the corresponding management agent 200 stops executing, and a printing of description (POD) included in the corresponding MO and MO (hereinafter M_POD). Information) such as desired POD is included. Here, the MO is a network device managed by a network management protocol in network management, and is a function (process) provided in the switch 210, and the POD is a condition that may exist in each MO through the operator terminal 220. Operation result data according to authorized command of description (COD).

Therefore, by transmitting the above-mentioned MMC log data request signal, the management agent 200 updates or updates in the POD of the function to be reported to the administrator (not shown) from the time when execution is stopped to the corresponding switch 210 to the present time. The newly generated information is required. When the MMC log data request signal is sent, the management agent processor 201 determines the MO information to be carried in the request signal based on the class of the MOs previously managed. The above-mentioned MO includes the base MO and all the MOs managed in the database 203.

After sending the MMC log data request signal to the switch 210, the management agent processor 201 checks whether M_POD is received from the switch 210 (step 304). If M_POD is received, the management agent processor 201 proceeds to step 305 and analyzes the received M_POD (Parsing). As a result of analysis, when the received M_POD corresponds to the MO managed by the memory 202, the corresponding MO is updated to be stored in the memory 202. However, as a result of analysis, if the received M_POD corresponds to the MO managed by the database 203, the corresponding MO is updated to be stored in the database 203 (step 306).

The management agent processor 201 then checks whether it has requested log data of all MOs. If it is checked that log data for all MOs is not required, the management agent processor 201 returns to step 303 to repeat the above-described process.

However, if log data of all the MOs is required, the management agent processor 201 proceeds from step 307 to step 308 to notify the administrator (not shown) of the rerun completion, and terminates the rerun process.

If an error message is received when M_POD is not received in step 304, the management agent processor 201 proceeds to step 307 through step 309 and operates in the same manner as described above. If both the M_POD and the error message have not been received, the process proceeds to step 310 to check whether a predetermined time has elapsed. As a result of the check, if the predetermined time has not elapsed, the process returns to step 304 to repeat the above-described process. If the predetermined time elapses, the process proceeds to step 307 and operates as described above.

On the other hand, when the exchange 210 receives the MMC log data request signal for any MO through the management agent interface 211, and transmits to the CCP (212). The CCP 212 searches whether the M_POD corresponding to the disk 213 exists according to the result of analyzing the received MMC log data request signal.

The disk 213 stores M_POD for each MO in the form of a file. The M_POD is an operation result according to the COD authorized from the operator terminal 220 while the management agent 200 is stopped. Therefore, the presence of the M_POD corresponding to the disk 213 means that the corresponding M_POD has been updated or newly created during the period in which the management agent 200 is stopped. At this time, there may be an M_POD belonging to the corresponding MO but not requested by the management agent 200 on the disk 213. This M_POD is information that does not need to be reported to a manager (not shown). For this reason, when the management agent 200 transmits a log data request signal, it designates up to M_POD to receive.

As a result of checking in step 401, if the corresponding M_POD exists in the disk 213, the CCP 212 proceeds to step 402 to read the corresponding M_POD stored in the corresponding log data file of the disk 213. In step 403, the read M_POD is sent to the management agent 200 through the management agent interface unit 211, and the sending operation according to the MMC log data request is completed.

However, if the M_POD corresponding to the corresponding file on the disk 213 does not exist as a result of checking in step 401, the updated or newly generated information does not exist for the corresponding M_POD while the management agent 200 is stopped. Since the CCP 212 notifies the error message through the management agent interface unit 211, and finishes the sending operation according to the MMC log data request.

As described above, the present invention separates the base MO that can be initialized by the management agent (Agent) of the communication management network (TMN) and the MO except this, and manages the MO except the base MO even in a state where execution is stopped. Access to the network components when rebuilding managed objects due to the management agent's redo is implemented by allowing the agent to be stored in the agent so that only the changed contents are read from the network component during the period of execution interruption. There is an effect that can minimize the number of times, reduce the number of IPC generated in the network component to improve the processing efficiency of the network component.

Claims (3)

In a communication management network in which an information structure is divided into managers, management agents, and communication network components, Initializing the management agent itself with respect to a basic management object when redo is required for the management agent; Initiating an interface mode with management objects managed in a separate area from the basic management object in the management agent; Reading, by the management agent, only the changed portion of all the management objects of the management agent from the communication component during the period of interruption of execution; Analyzing the read content and updating a corresponding managed object; And when the construction of all the management objects is completed, the management agent reporting the completion of the redo to the manager. The method of claim 1, wherein the management agent re-execution method repeats the reading and updating steps in a management object unit until all the management objects are constructed in the management agent. In a communication management network in which an information structure is operated by a manager, a management agent, and a communication network component structure, A memory storing a basic management object that the management agent can initialize itself when re-executing; A database storing management objects except for the basic management object among management objects to be built in the management agent when the redo is executed; When the redo is requested, the basic management object to be stored in the memory is initialized, the interface mode with the database is initialized, and execution of the information about all management objects to be constructed in the management agent is stopped. A management agent processor for receiving only information changed during the period from the communication network component to update information in the memory and the database; A disk provided in the network component, the disk storing information about the management objects generated in the network component during a period in which the execution is stopped; A central control processor (CCP) provided in the communication network component and configured to read information corresponding to the arbitrary management object from the disk and to transmit the information to the management agent when the changed information on any management object is requested from the management agent. Re-execution apparatus of the management agent comprising a.