JP2885123B2 - Name conversion method in network management system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a name conversion method in a network management system, and more specifically, when a plurality of agents are managed by a manager, the names of objects to be managed under each agent are uniquely identified by the manager. The name conversion method for

[0002]

2. Description of the Related Art In a network management system that implements OSI management and a manager integrally manages a plurality of agents, each agent manages the names of managed objects managed by the agents, their attributes, and the inclusive relation between objects. It holds a defined management information base (hereinafter, referred to as MIB). Conventionally, the highest managed object name in the instance tree defined in the MIB, that is, the highest managed object name in the inclusion relationship of the managed objects managed by the agent is duplicated between agents. It was not allowed to do so, and it was necessary for the manager to have the same unique managed object name as the unique managed object name given to them.

Conventionally, as shown in FIG. 7, when a new agent is connected to a manager, prior to the connection, a management object name defined in the MIB of the agent and an object instance type are used. The work of converting all the top-level managed object names in a certain attribute value or the like into unique managed object names on the manager side has been performed.

[0004] For example, if the highest managed object name of a certain agent is / networkID / ec, the agent can be directly connected to a manager that can be uniquely managed by the managed object name. For example, in order to connect to a manager that must be managed with a unique name of / networkID / tokyo because it overlaps with the top-level managed object name of another agent already connected, for example, The top-level managed object name must be changed to the above name. The name of a managed object lower than the highest managed object includes a part of the highest managed object name, such as / networkID / ec / xNameID / x1, and is defined in the MIB. The attribute value of each managed object includes many values of the object instance type including the name of the managed object. Therefore, the attribute value of “/ networkID / ec” in the data defined by MIB is included. All parts are "/ n
After converting to "networkID / tokyo", it is necessary to connect to the manager.

[0005]

As described above, in the conventional network management system, the name of the highest managed object in the inclusion relation of the managed objects managed by the agents is not allowed to be duplicated among the agents. Therefore, prior to connection, it is necessary to convert values such as attribute values including managed object names and object instance types defined in the MIB of the agent, and the number of objects managed by the agent. When there are many, there is a problem that the conversion process takes much time and connection is delayed. In addition, the agent cannot perform an operation using the same managed object name as before the connection after the connection with the manager.

Accordingly, an object of the present invention is to allow the top-level managed object name in the inclusion relationship of the managed objects managed by the agent to overlap among the agents, and to allow the manager to manage the managed object names under each agent. An object of the present invention is to provide a name conversion method in a network management system that can be uniquely identified.

[0007]

According to the present invention, in order to achieve the above object, in a network management system which implements OSI management and a manager integrally manages a plurality of agents, a management target object managed by the agents is included. If managed object name the uppermost but overlap between agents, which the manager side manages to impart unique managed object name to them in relation to the manager, the agent
Acts as a point of contact for operation results and event notifications returned from
A correspondence describing the correspondence between the endpoint object for each agent, the top-level managed object name on the agent side, and the managed object name given to it on the manager side for each agent ID of each agent The end point object receiving the data receives a table and a portion corresponding to the highest managed object name on the agent side included in the data received from the agent through the end point object.
Corresponding to the agent ID of the agent
Refers to the correspondence described in the correspondence table, converts it to the managed object name given by the manager, and converts it to the managed object name given by the manager included in the data to be sent to the agent. Match the corresponding part to the agent ID of the agent.
In response to this, there is provided a name conversion means for referring to the correspondence described in the correspondence table and converting it to the highest-level managed object name on the agent side.

Here, when the manager requests the agent to perform the operation, the name conversion means sets the name of the operation target object and the value of the object instance type included in the operation time information as a target of the name conversion processing, and executes the operation result from the agent. At the time of acquisition, the name of the operation target object and the value of the object instance type included in the operation result information are subject to name conversion processing, and at the time of event notification from the agent, the event occurrence object name and the value of the object instance type included in the occurrence event information And are subject to name conversion processing.

[0009]

In the name conversion method in the network management system of the present invention, the highest managed object name in the inclusion relation of the managed objects managed by the agents is duplicated among the agents, but the manager side has a unique name for each of them. If you manage to impart Do managed object name correspondence table provided in the manager, the management highest object name and manager side at the agent side for each agent ID of each agent contrast And the name conversion means provided in the manager is used by the agent to convert the information contained in the data received from the agent when the operation result is obtained or when an event occurs. The equivalent of the top-level managed object name And by referring to the correspondence table, it converts to a managed object name assigned by the manager side for it. Conversely, the location corresponding to the managed object name given by the manager included in the data to be sent to the agent at the time of the operation request is referred to the correspondence table, and the highest managed object on the agent side Convert to object name.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described in detail with reference to the drawings.

Referring to FIG. 1, in an example of a network management system to which the present invention is applied, a manager 1 and a plurality of agents 2 and 3 are connected by communication paths 4 and 5.

The agents 2 and 3 have MIBs 6 and 7 that define the names of managed objects managed by themselves, their attributes, and the inclusion relations between the objects, and the management objects defined by the MIBs 6 and 7. In response to an operation request for the object from the manager 1, the managed object is actually operated, and an event occurred in the managed object is notified to the manager 1. On the other hand, the manager 1 is a part that manages the entire network, issues operation requests to the managed objects under the agents 2 and 3 to obtain attribute values and the like, and notifies the agents 2 and 3 of events from the managed objects. , Etc., are received from the agents 2 and 3 to perform processing such as grasping the state of the network.

Although not shown in FIG. 1, OSI management is implemented in the manager 1 and the agents 2 and 3. In other words, the manager 1 implements a manager application and a protocol stack for OSI management, and the agents 2 and 3 implement an agent administration application and a protocol stack. Here, the protocol stack means a set of protocols of the first to seventh layers of OSI, and particularly, a network management protocol such as CM
IS / CMIP (Common Management Information Service / Common Management Information Protocol) is incorporated. This CMIS / CM
The IP is a command (GET request and its response) for acquiring attribute values of managed objects held in MIBs 6 and 7 of agents 2 and 3, and a command for rewriting the contents of MIBs 6 and 7 (SET request and response). Various commands are supported, such as a response (response thereof), and a command (EVENT-REPORT request and response thereof) for notifying the manager 1 of an event occurring in the managed object from the agents 2 and 3. The management business applications of the manager 1 and the agents 2 and 3 send and receive these commands, so that the agents 2 and 3 provide management information and notify events to the manager 1, and the manager 1 uses the agents 2 and 3 The network management is performed in cooperation with each other by requesting management information, setting various attribute values, and the like.

In this embodiment, the manager 1 in the network management system as described above is provided with a correspondence table 11 and name conversion means 12.

The correspondence table 11 describes, for each agent, the correspondence between the highest-level managed object name on the agent side and the managed object name given to it on the manager side. For example, assume that the inclusion relationship of the managed object defined in MIB 6 of agent 2 is shown in FIG. 2B, and the inclusion relationship of the managed object defined in MIB 7 of agent 3 is shown in C of FIG. Then, the top-level managed object names overlap each other with / networkID / ec. And on the manager side,
As shown in A of FIG. 1, the highest managed object on the agent 2 side is assigned with / networkID / tokyo, and the highest managed object on the agent 3 side is assigned with / networkID / oka and managed. , The correspondence table 11 includes
For example, as shown in FIG. 3, the highest managed object name (/ n) on the agent 2 side corresponds to the agent ID (= tokyo) assigned to the agent 2.
networkID / ec) and the manager 1
Object name (/ network
kID / tokyo) is set, and the highest managed object name (/ networkID / ec) on the agent 3 side and the manager corresponding to the agent ID (= osaka) assigned to the agent 3 are set. A set is set with the management target object name (/ networkID / oka) assigned on the first side.

The name conversion means 12 performs such a correspondence table 1
1, it corresponds to the name of the highest-level managed object on the agent 2, 3 side included in the data received from the agents 2, 3 (data at the time of obtaining operation results, data at the time of event notification). With reference to the correspondence table 11, the location is converted into a managed object name given by the manager 1, and the manager 1 included in data to be transmitted to the agents 2 and 3 (data at the time of operation request). The part corresponding to the managed object name given by the agent is referred to the corresponding table of the agent 2,
This is a means for converting to the top-level managed object name on the third side.

The name conversion process by the name conversion means 12 includes (1) when the manager 1 requests an operation to the agents 2 and 3, (2) when a response from the agents 2 and 3 to the operation request, and (3) ) Agent 2,
3 when the event is notified. Less than,
The operation of this embodiment will be described.

(1) From manager 1 to agent 2,
When an operation is requested to the manager 3, when performing any operation (for example, acquiring or changing an attribute value) on the managed object managed by the agents 2 and 3, the manager 1 An operation request including operation-time information describing whether such an operation is desired is issued from the business management application on the manager 1. It should be noted that “operation” in the present specification is a concept including “operation” in OSI management.

When an operation request is issued, the name conversion means 12 executes processing as shown in FIG. First, an agent to be transmitted is specified based on the operation target object name (S1). For example, the operation target object name specified in the operation request is: / networkID / tokyo / xNameID /
If it is x1, since it contains the agent ID "tokyo" of the agent 2, it is determined that it should be transmitted to the agent 2.

Next, the name conversion means 12 determines whether or not an object instance type is defined in the operation information (S2). If the object instance type is not defined, the process proceeds to step S4. If the object instance type is defined, the name of the value of the object instance type is converted in step S3. The name conversion process in step S3 is performed as follows.

First, all values of the object instance type are searched from the operation time information. Next, name conversion is performed on the retrieved values of the individual object instance types in accordance with the correspondence shown in the correspondence table 11. For example, if the current operation request is sent to the agent 2 and the operation time information includes data such as SEQUENCE {a ObjectInstance b INTERGER} a = / networkID / tokyo / xNameID / x1. "A Object
Since there is a definition of the object instance type “tInstance”, all portions of the value “a” of the object instance type are searched. Then, referring to the correspondence table 11, the object name “/ net” on the manager 1 side is displayed.
Since “workID / tokyo” corresponds to the object name “/ networkID / ec” on the agent 2 side, name conversion is performed as follows. SEQUENCE ｛a ObjectInstance b INTAGEＴＥ a = / networkID / ec / xNameID / x1

It should be noted that where the value of the object instance type appears in the operation time information tends to be largely determined by the combination of the type of operation (type of attribute acquisition, change, etc.) and the attribute name. Therefore, for each combination of the name of the operation and the attribute name of the operation requesting this time, a function for performing the name conversion processing in step S3 is prepared in advance, and the corresponding function is called to perform the name conversion. By doing so, the efficiency of the name conversion process is high. Such a method of selectively using the functions can also be used for conversion of the operation target object name, name conversion in operation result information described later, and name conversion in event notification information.

Next, the name conversion means 12 converts the operation target object name specified in the current operation request according to the correspondence shown in the correspondence table 11 (S4). That is, the current operation request is sent to the agent 2, and the operation target object name is / networkID / tokyo / xNameID /
If it is x1, it is converted as follows: / networkID / ec / xNameID / x1.

The operation request subjected to the name conversion processing in the name conversion means 12 as described above is transmitted to the agent 2 specified in step S1.

(2) Agent 2 for operation request
At the time of response from 3, the agents 2 and 3 operate the corresponding managed object based on the operation request transmitted from the manager 1. The operation result including the name of the operated managed object (operation object name) and information indicating the result of the operation (operation result information) is stored in the manager 1.
Return to. Here, generally, the manager 1 generates an object called an end point object for each of the agents 2 and 3 at the time of startup, and this end point object plays a role of a window for an operation result returned from the agents 2 and 3 and an event notification. Therefore, the operation results transmitted from the agents 2 and 3 are passed to the name conversion means 12 through the end point object.

When the operation results are returned from the agents 2 and 3, the name conversion means 12 of the manager 1 starts the processing shown in FIG. First, it is determined whether an object instance type is defined in the operation result information (S11). When the object instance type is not defined, the process proceeds to step S13. When the object instance type is defined, the name of the value of the object instance type is converted in step S12. That is, as in the case of the operation request, first, all the values of the object instance type are searched from the operation result information.
Performs name conversion according to the correspondence indicated by. At this time, the set in the correspondence table 11 to be referred to is the set corresponding to the agent to which the operation result has been transmitted. Therefore, when the operation result is returned from the agent 2, the object name “/ netw” on the agent 2 side among the values of the object instance type included in the operation result information.
The “orkID / ec” part is replaced with the object name “/ networkID / tokyo” on the manager 1 side.

Next, the name conversion means 12 converts the operation target object name specified by the current operation result according to the correspondence shown in the correspondence table 11 (S13). That is, assuming that the current operation request is transmitted from the agent 2 and the operation target object name is / networkID / ec / xNameID / x1, / networkID / tokyo / xNameID /
x1.

The operation result subjected to the name conversion processing by the name conversion means 12 is notified to the management business application which is the operation request source.

(3) Event Notification from Agents 2 and 3 When the occurrence of an event is notified from the managed object managed by itself, the agent 2 or 3 sends an event notification to the manager 1. The event notification includes the management target object name (event occurrence object name) where the event has occurred, and event information indicating the event in which the event has occurred.

Upon receiving the event notification from the agents 2 and 3 through the end point object, the name conversion means 12 of the manager 1 starts the processing shown in FIG. First, it is determined whether an object instance type is defined in the event information (S21). If the object instance type is not defined, step S23
If it is defined, the name of the value of the object instance type is converted in step S22. That is, as in the case of the operation request, all the values of the object instance type are searched from the event notification, and then the names of the searched individual object instance types are determined in accordance with the correspondence shown in the correspondence table 11. Perform the conversion. At this time, the set of the correspondence table 11 to be referred to is the set corresponding to the agent to which the event notification has been transmitted.
Therefore, when the event notification is transmitted from the agent 2, the part of the object name “/ networkID / ec” on the agent 2 side in the value of the object instance type included in the event notification is Object name "/ networkID / to
kyo ".

Next, the name conversion means 12 converts the event occurrence object name specified in the current event notification according to the correspondence shown in the correspondence table 11 (S23). That is, if the event notification of this time is transmitted from the agent 2 and the event occurrence object name is / networkID / ec / xNameID / x1, then / networkID / tokyo / xNameID /
x1.

The event notification having undergone the name conversion processing as described above is notified to the management business application.

[0033]

As described above, in the name conversion method in the network management system of the present invention, the name of the highest-level managed object on the agent side and the name assigned to the manager on the agent side are provided for each agent ID of each agent. Using the correspondence table describing the correspondence relationship with the managed object names, the name conversion unit determines the part corresponding to the highest managed object name on the agent side included in the data received from the agent, On the other hand, it is converted to the managed object name given by the manager, and conversely, the part corresponding to the managed object name given by the manager included in the data to be sent to the agent is converted to the minimum at the agent. It is now converted to a higher-level managed object name. There can be managed object name the uppermost be duplicated between agents, which uniquely identifies the managed object name under each agent manager in a containment relationship of managed objects managed. Therefore, prior to connection, M
There is no need to change the IB.

The agent can operate using the same managed object name as before the connection with the manager.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an example of a network management system to which the present invention has been applied.

FIG. 2 is a diagram illustrating an example of an inclusion relationship between management targets on a manager and an agent side.

FIG. 3 is a diagram showing an example of a correspondence table.

FIG. 4 is a flowchart illustrating a processing example of a name conversion unit when an operation is requested.

FIG. 5 is a flowchart illustrating a processing example of a name conversion unit when acquiring an operation result.

FIG. 6 is a flowchart illustrating a processing example of a name conversion unit at the time of event notification.

FIG. 7 is an explanatory diagram of a conventional technique.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Manager 11 ... Correspondence table 12 ... Name conversion means 2, 3 ... Agent 4, 5 ... Communication path 6, 7 ... MIB

──────────────────────────────────────────────────続 き Continuation of the front page (56) References Keiko Arima, et al., “Unified management method of management target in network management system”, IEICE Technical Report, November 1992, Vol. 92, No. 317, p. 7-11 Mamoru Maekawa 2 authors, "What comes next after distributed operating system UNIX", first edition, Kyoritsu Shuppan Co., Ltd., December 25, 1991, p. 277-281 (58) Field surveyed (Int. Cl. ⁶ , DB name) G06F 12/00 G06F 13/00

Claims (4)

(57) [Claims] In a network management system that implements OSI management and a manager integrally manages a plurality of agents, the highest managed object name in the inclusion relationship of the managed objects managed by the agent is duplicated among the agents. When the manager side manages them by assigning unique managed object names to them, the manager is the point of contact for operation results and event notifications returned from agents.
Endpoint objects for each agent
When a correspondence table describing the correspondence between a managed object name given to it in the managed object name and manager side uppermost on the agent side for each agent ID of each agent, the agent through the end point object the portion corresponding to the managed object name of the top level at the agent side included in the received data from the data
Of the agent responsible for the endpoint object that received the
It is described in the correspondence table corresponding to the agent ID.
That by referring to the correspondence relationship, contrast into a managed object name assigned by the manager side, the portion corresponding to the managed object name assigned by the manager side contained in the data to be transmitted to the agent, the A
Corresponding to the Agent ID of the Agent
A name conversion means for referring to the described correspondence and converting it to the highest-level managed object name on the agent side. 2. The method according to claim 1, wherein the name conversion unit sets the name of the operation target object and the value of the object instance type included in the operation time information as a target of the name conversion processing at the time of the operation request from the manager to the agent. A name conversion method in the network management system according to claim 1. 3. The method according to claim 1, wherein the name conversion unit sets the name of the operation target object and the value of the object instance type included in the operation result information as a target of the name conversion processing when obtaining the operation result from the agent. Item 1
Or a name conversion method in the network management system according to 2. 4. The name conversion unit according to claim 1, wherein, when an event is reported from the agent, the name of the event occurrence object and the value of the object instance type included in the occurrence event information are subjected to name conversion processing. , 2 or 3, the name conversion method in the network management system.