JP2011035753A - Network management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To manage a network without using a standby apparatus for redundancy, and to improve the reliability of the network. <P>SOLUTION: In a network management system 100, a manager 3 obtains the management information of network elements 6a, 6b connected to a LAN-I, and at the same time, and obtains the management information of network elements 6c, 6d connected to a LAN-II through a manager 4. Similarly, a manager 4 also obtains the management information of the LAN-I and the LAN-II. A naming server 5 monitors each status of the managers 3, 4, and determines the manager 3 or the manager 4 of a connection destination receiving a request from a client terminal 1. Thereby, since it is possible to manage a wide area network even if any of the managers 3, 4 installed in every network is utilized, the wide area network can be managed by the other managers even if failure occurs in one manager, thus improving the reliability of the network. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

The present invention relates to a network management system for managing devices connected to a network, and particularly to managing a wide area network composed of a plurality of local networks.

Conventionally, a network management system for managing a single local area network (LAN) is generally a network element (such as a server or a router connected to the LAN) (
Device) and a management server that integrally manages these network elements. In addition,
As appropriate, the management server is called a manager.

A configuration example of a network management system 300 according to a conventional example will be described with reference to FIG. A network management system 300 shown in FIG. 7 includes a manager 30 and network elements 6a and 6b, and manages a single LAN. There are various types of LAN connection forms (network topologies) such as a star type, a ring type, and a bus type, and any topology may be used. In this example, the manager 30 and the network elements 6a, 6
b constitutes a ring-type LAN.

Network management agents (software) are installed in the network elements 6a and 6b. Each agent has a configuration of network elements 6a and 6b,
Collect management information on performance, failures, etc. For example, the agent collects management information of the network elements 6a and 6b with reference to a management information base called MIB (Management Information Base) of the network elements 6a and 6b. There are various MIBs such as vendors and device-specific ones. RFC (Re) issued by an organization that establishes technology standards related to the Internet as MIBs to be implemented by devices connected to the network
“quest For Comment” defines “standard MIB”.

The manager 30 acquires management information of the network elements 6a and 6b via the agent. The communication between the manager 30 and the agent is SNMP (Simple Network).
rk Management Protocol) is often used.

The manager 30 requests management information of the network elements 6 a and 6 b from the agent at a certain frequency, and the agent returns the management information to the manager 30. When an event such as a failure or a state change occurs in the network elements 6a and 6b, the agent notifies the manager 30 of the event. Based on the event received from the agent, the manager 30 displays the network configuration and state on the display, and notifies the administrator of failure information by mail or an alarm function. In this way, the manager 30 is connected to the LAN.
Since the network elements 6a and 6b connected to the network are managed, the manager can grasp the state of the network by using the manager 30. Patent Document 1 describes that a management apparatus acquires history information stored in an extended MIB of a terminal apparatus connected via a network using SNMP.

On the other hand, a wide area network (WAN) consisting of multiple LANs
When managing a large number of network elements with a single manager, there are problems such as the reliability of the manager itself and the difficulty of providing management suitable for each LAN. In order to solve these problems, there is a method of performing distributed management by installing a manager for each LAN. Here, a manager that manages individual LAN information is called a sub-manager. Furthermore, an integrated manager for managing information on a plurality of LANs is installed through these sub-managers.

A configuration example of a network management system 400 according to a conventional example will be described with reference to FIG. A network management system 400 shown in FIG. 8 is a hierarchical system composed of sub-managers 61 and 62 and an integrated manager 63, and manages a wide area network composed of a plurality of LAN-Is and LAN-IIs. The ring type LAN-I includes a sub-manager 61 and network elements 6a and 6b. The ring LAN-II includes a sub-manager 62 and network elements 6c and 6d.

Network management agents (software) are installed in the network elements 6a to 6d. Each agent has a configuration of network elements 6a-6d,
Collect management information on performance, failures, etc. The sub-manager 61 acquires management information of the network elements 6a and 6b via the agent, and the sub-manager 62 acquires management information of the network elements 6c and 6d via the agent.

The integrated manager 63 acquires LAN-I and LAN-II management information by communicating with the sub-managers 61 and 62 via the WAN. For example, if the administrator is an integrated manager 6
3 and the integrated manager 63 acquires the management information of LAN-I and LAN-II, the following operation is performed. First, the integrated manager 63 sends a LAN-I to the sub-manager 61.
Request management information. Next, the sub-manager 61 collects the management information of the network elements 6a and 6b in real time via each agent or refers to the management information collected periodically. Next, the sub-manager 61 aggregates the collected management information of the network elements 6a and 6b, and transmits the collected management information to the integrated manager 63 as LAN-I management information.

Similarly, the integrated manager 63 requests LAN-II management information from the sub-manager 62. Next, the sub-manager 62 passes through each agent to the network element 6c.
, 6d is collected, and the collected management information of the network elements 6c, 6d is aggregated and transmitted to the integrated manager 63 as LAN-II management information.

The integrated manager 63 receives the LAN-I and LAN-II received from the sub-managers 61 and 62.
Based on the management information, the configuration and status of the network composed of LAN-I and LAN-II are displayed on the display, or the failure information is notified to the administrator by e-mail or an alarm function. As described above, the integrated manager 63 receives a plurality of LAN-Is via the sub-managers 61 and 62.
, LAN-II management information can be referred to, so that the administrator can use the integrated manager 63 to grasp the status of the entire network including a plurality of LAN-Is and LAN-IIs.

For example, when an event such as a failure or a state change occurs in the network element 6 a, the agent of the network element 6 a may notify the integrated manager 63 of the event via the sub-manager 61.

Japanese Patent Application Laid-Open No. 2004-280592

By the way, to manage a wide area network consisting of multiple LAN-I and LAN-II,
Distributed management is realized by installing sub-managers 61 and 62. However, these sub-managers 61 and 62 are managed by the integrated manager 63. For this reason,
When the integrated manager 63 fails, even if all the sub-managers 61 and 62 are normal, the administrator cannot obtain management information of the network elements 6a to 6d connected to the LAN-I and LAN-II, and the network There is a risk that the reliability of the system will be reduced. In order to deal with this problem, a method of duplicating by setting up an integrated manager on the standby side is also conceivable, but there are problems of increase in equipment and cost increase.

SUMMARY OF THE INVENTION The present invention solves the above-described problems, and an object thereof is to provide a network management system that manages a network without using a duplex standby device and improves the reliability of the network. To do.

In order to solve the above-described problem, a network management system according to the present invention is installed for each network and acquires management information of devices connected to its own network.
A management server that acquires management information of devices connected to other networks, a client terminal that can communicate with each of the management servers, a naming server that monitors each of the management servers and can communicate with the client terminals, The client terminal requests a connection destination management server from the naming server, and the naming server that has received the request from the client terminal determines a connection destination management server based on a predetermined connection order. ,
The management server determined by the naming server transmits the acquired management information to the client terminal.

In the present invention, the naming server that has received a request from the client terminal determines a management server as a connection destination based on a predetermined connection order. For example, the naming server determines the connection order to connect to the management server based on the failure and load of the management server. The management server determined by the naming server transmits the acquired management information to the client terminal. For example, the management server integrates and manages management information of devices connected to its own network and management information of devices connected to other networks, and transmits the integrated management information to the client terminal. The client terminal acquires network management information from the management server and displays the network configuration and status on the display.

According to the network management system of the present invention, the management server that acquires the management information of the device connected to its own network and the management server that acquires the management information of the device connected to the other network is monitored, and from the client terminal And a naming server that determines a connection destination management server in response to the request.

As a result, the wide area network can be managed by using any of the management servers installed in each network, so even if a failure occurs in one management server, the wide area network can be managed by another management server. Can be improved. Moreover, the naming server eliminates the need for the administrator to be aware of which management server to use among the plurality of management servers, thereby reducing the burden on the administrator.

It is a block diagram which shows the structural example of the network management system 100 which concerns on this invention. It is an activity diagram which shows the operation example of the network management system 100 at the time of normal. It is an activity figure which shows the operation example of the network management system 100 when the manager 3 is no response. 3 is a block diagram illustrating a configuration example of managers 3 and 4 and a naming server 5. FIG. It is a sequence diagram which shows the example which collects the management information of a network. It is a block diagram which shows the structural example of the large-scale job management system 200 as 2nd Example based on this invention. It is a block diagram which shows the structural example of the network management system 300 which concerns on a prior art example. It is a block diagram which shows the structural example of the network management system 400 which concerns on a prior art example.

Next, an embodiment of a network management system according to the present invention will be described with reference to the drawings. The present invention acquires management information of network components (devices) connected to its own LAN, and also acquires a plurality of managers that acquire management information of network components connected to other LANs, and By providing a naming server that monitors each state and determines a connection destination manager in response to a request from a client terminal,
The network reliability can be improved and the burden on the network administrator can be reduced.

Referring to FIG. 1, a network management system 100 as a first embodiment according to the present invention.
An example of the configuration will be described. A network management system 100 shown in FIG. 1 includes a client terminal 1, managers 3 and 4 and a naming server 5 that can communicate with the client terminal 1, and manages a wide area network composed of a plurality of LAN-Is and LAN-IIs. Ring type L
AN-I includes a manager 3 and network elements 6a and 6b. The ring-type LAN-II includes a manager 4 and network elements 6c and 6d. L
The connection form (network topology) of the AN includes a star type and a bus type in addition to the ring type, and any topology may be used.

Network management agents (software) are installed in the network elements 6a to 6d. Each agent has a configuration of network elements 6a-6d,
Collect management information on performance, failures, etc. For example, the agent collects management information with reference to the MIB in the network elements 6a to 6d. The manager 3 acquires management information of the network elements 6a and 6b via the agent, and the manager 4 acquires management information of the network elements 6c and 6d via the agent. In this example, Manager 3
, 4 and the agent use SNMP as a communication protocol.

The manager 3 is an example of a management server, is installed in the LAN-I, acquires management information of the network elements 6a and 6b connected to its own LAN-I, and is connected to other LAN-II. The management information 6c and 6d is acquired via the manager 4. For example, the manager 3 requests management information of the network elements 6 a and 6 b from the agent at a certain frequency, and the agent returns the management information to the manager 3. When an event such as a failure or a state change occurs in the network elements 6a and 6b, the agent notifies the manager 3 of the event. The manager 3 requests the manager 4 for the management information of the network elements 6 c and 6 d, and the manager 4 returns the management information to the manager 3.
The manager 3 acquires LAN-I and LAN-II management information and integrates and manages them (integrated management function).

Similarly, the manager 4 is an example of a management server and is installed in the LAN-II and has its own LA.
Acquire management information of network elements 6c and 6d connected to N-II and
The management information of the network elements 6a and 6b connected to N-I is acquired via the manager 3. For example, the manager 4 sends a network element 6c to the agent at a certain frequency.
, 6d, and the agent returns the management information to the manager 4. When an event such as a failure or a state change occurs in the network elements 6c and 6d, the agent notifies the manager 4 of the event. The manager 4 requests the manager 3 for management information of the network elements 6 a and 6 b, and the manager 3 returns the management information to the manager 4. The manager 4 acquires management information of LAN-I and LAN-II and integrates and manages them (integrated management function).

As described above, both the manager 3 and the manager 4 acquire and integrate management information of LAN-I and LAN-II, and therefore, if a failure occurs in the integrated management function of the manager 3. Even if it exists, the management information of LAN-I and LAN-II can be acquired via the manager 4. As a result, the reliability of the network can be improved. The client terminal 1 requests management information of the network elements 6 a to 6 d from the naming server 5.

The naming server 5 monitors the state of each of the managers 3 and 4, and the client terminal 1
In response to the request from the manager 3, the manager 3 or 4 is determined as the connection destination. For example, the naming server 5 registers connection information transmitted from the managers 3 and 4 when the managers 3 and 4 are activated. For example, the naming server 5 selects one manager 3 from a plurality of managers 3 and 4, and returns connection information of the manager 3 to the client terminal 1 in response to a request from the client terminal 1. Thereby, the administrator who operates the client terminal 1 does not need to be aware of which manager among the plurality of managers 3 and 4 and can reduce the burden on the administrator.

The naming server 5 keeps (ke) to confirm that it is operating normally.
ep alive) signal is periodically transmitted to the managers 3 and 4. The managers 3 and 4 respond to the keep alive signal when operating normally, and do not respond to the keep alive signal in the case of a failure or an overload condition. As a result, the naming server 5
It can be determined whether or not is operating normally. Therefore, the naming server 5 can switch the manager of the connection destination provided to the client terminal 1 to a manager that is operating normally.

The client terminal 1 can always access a normal manager by inquiring the connection destination to the naming server 5. Furthermore, even if one of the managers is accessed and that manager fails and no response is made, the manager of the connection destination is switched by inquiring the naming server 5 again, and the management information of LAN-I and LAN-II is integrated. And continue to manage functions.

The client terminal 1 receives the LAN-I and L received from the manager 3 or 4
Based on the management information of AN-II, the network configuration and status are displayed on the display, and the failure information is notified to the administrator by an alarm function or the like. Thus, since the client terminal 1 acquires the management information of the network elements 6a to 6d from the manager 3 or the manager 4, the administrator accesses the manager 3 or the manager 4 determined by the naming server 5 by the client terminal 1. Thus, it is possible to grasp the status of the network composed of LAN-I and LAN-II.

Next, an operation example of the network management system 100 at the normal time will be described with reference to FIG. In this example, the naming server 5 registers the connection information transmitted from the managers 3 and 4. The naming server 5 is set to connect to the manager 3 when both the managers 3 and 4 are operating normally. Of course, this manager 3
, 4 are normally operating, the management information to be acquired is the same regardless of which of the manager 3 and the manager 4 is connected, so that the manager 4 may be set to be connected.

In step ST1 shown in FIG. 2, when starting the client application, the client terminal 1 requests a manager connection destination from the naming server 5 and starts the step ST2.
Migrate to

In step ST2, the naming server 5 selects one manager 3 from the plurality of managers 3 and 4, returns the connection information of the manager 3 to the client terminal 1, and returns to step ST3.
Migrate to In step ST3, the client terminal 1 connects to the manager 3 based on the acquired connection information, requests network management information from the manager 3, and moves to step ST4.

In step ST4, the manager 3 manages the network element 6a in the LAN-I that it manages.
, 6b is collected from the agent, and the process proceeds to step ST5. Step ST
5, the manager 3 sends a network element 6 c, from the manager 4 that manages the LAN-II.
The management information 6d is acquired, and the process proceeds to step ST6.

In step ST6, the manager 3 integrates the management information of the LAN-I collected by itself and the management information of the LAN-II acquired from the manager 4, returns it to the client terminal 1 as network management information, and proceeds to step ST7. To do. In step ST7, the client terminal 1 displays the acquired network management information on the display. As a result, the administrator can grasp the status of the network composed of LAN-I and LAN-II only by accessing the manager 3 that manages LAN-I.

Subsequently, referring to FIG. 3, the network management system 1 when the manager 3 is not responding.
An operation example of 00 will be described. In this example, a case will be described in which a failure occurs in the function of integrated management of network management information in the manager 3 connected to the client terminal 1, and the connected manager 3 becomes unresponsive.

In step ST11 shown in FIG. 3, the client terminal 1 requests the latest management information from the connected manager 3, and proceeds to step ST12. At step ST12, manager 3
Does not respond because a failure has occurred in the function for integrated management of network management information.
Subsequently, the process proceeds to step ST13.

In step ST13, when there is no response from the manager 3 for a certain time, for example, for 4 to 5 seconds, the client terminal 1 determines that a failure has occurred in the integrated management function of the manager 3 and connects to the naming server 5 Request the manager's connection destination. Then step ST1
4

In step ST14, the naming server 5 returns the connection information of the manager 4. For example, the naming server 5 transmits a keep-alive signal for confirming that it is operating normally to the managers 3 and 4 at a certain frequency. In this example, the naming server 5
Has received a response from the manager 4 and confirmed that it is operating normally, but the manager 3 has not received a response from the manager 3 because the integrated management function has failed. . Accordingly, the naming server 5 returns the connection information of the manager 4 to the client terminal 1 in response to the request from the client terminal 1. Subsequently, the process proceeds to step ST15.

In step ST15, the client terminal 1 connects to the manager 4 from the connection information acquired from the naming server 5, and requests network management information. Then step ST
Move to 16.

In step ST16, the manager 4 collects management information of the network elements 6c and 6d in the LAN-II managed by the manager 4 from the agent, and proceeds to step ST17. In step ST17, the manager 4 acquires the management information of the network elements 6a and 6b from the manager 3 that manages LAN-I. If Manager 3 is completely down due to a failure,
The manager 4 cannot acquire LAN-I management information. Subsequently, the process proceeds to step ST18.

In step ST18, the manager 4 integrates the management information of the LAN-II collected by itself and the management information of the LAN-I acquired from the manager 3, returns it to the client terminal 1 as network management information, and proceeds to step ST19. To do. In step ST19, the client terminal 1 displays the acquired network management information on the display. This
The administrator can acquire management information of LAN-I and LAN-II via the manager 4 even if a failure occurs in the integrated management function of the manager 3. Therefore, the reliability of the network can be improved.

2 and 3, the managers 3 and 4 acquire network management information in real time when a request is received from the client terminal 1, but the manager 3
, 4 may collect network management information at a certain frequency and return the collected management information when a request is received from the client terminal 1.

Next, configuration examples of the managers 3 and 4 and the naming server 5 will be described with reference to FIG. The manager 3 shown in FIG. 4 includes a communication unit 31, a network information collection unit 32, an LA
N-to-N connection relation information storage unit 33, manager information storage unit 34, LAN information collection unit 35
, A LAN information storage unit 36, a remote LAN information acquisition unit 37, a LAN information provision unit 38, an activation communication unit 39, and a keep alive reception unit 310 are provided.

The communication unit 31 transmits network management information to the client terminal 1 in response to a request from the client terminal 1. The network information collection unit 32 collects network management information in response to a request from the client terminal 1.

The connection relationship information storage unit 33 between the LANs stores connection relationship information of a plurality of LANs connected to the network. In this example, the LAN-to-LAN connection relationship information storage unit 33 stores LAN-II connection relationship information. The manager information storage unit 34 stores connection information of a plurality of managers provided for each LAN. In this example, the manager information storage unit 34 is the manager 4
Stores connection information.

The in-LAN information collecting unit 35 collects management information from network elements to be managed in the LAN. In this example, the LAN internal information collection unit 35 collects management information of the network elements 6a and 6b. The LAN internal information storage unit 36 stores the management information collected by the LAN internal information collection unit 35. In this example, the in-LAN information storage unit 36 stores management information of the network elements 6a and 6b.

The remote LAN information acquisition unit 37 is an LA stored in the connection relation information storage unit 33 between LANs.
Based on the N connection relation information and the manager connection information stored in the manager information storage unit 34, communication with the manager in the remote LAN is performed to obtain management information of the remote LAN.
In this example, the remote LAN information acquisition unit 37 communicates with the manager 4 and acquires the management information of the manager 4.

The LAN information providing unit 38 responds to the request from the manager in the remote LAN by its own L
Provides management information for AN. In this example, the in-LAN information providing unit 38 provides its own LAN-I management information in response to a request from the manager 4.

The start communication unit 39 notifies the naming server 5 of connection information of the manager 3 when the manager 3 is started. The keep alive receiving unit 310 responds to the keep alive signal transmitted from the naming server 5.

The manager 4 shown in FIG. 4 has the same configuration as the manager 3 described above.
That is, the manager 4 includes a communication unit 41, a network information collection unit 42, a connection relation information storage unit 43 between LANs, a manager information storage unit 44, an in-LAN information collection unit 45, an in-LAN information storage unit 46, and remote LAN information acquisition. Unit 47, in-LAN information providing unit 48, activation communication unit 49
And a keep-alive receiving unit 410. Since the manager 4 has the same configuration as the manager 3, its detailed description is omitted.

The naming server 5 shown in FIG. 4 includes a manager registration unit 51 and a keep-alive transmission unit 52.
A manager information storage unit 53, a connection destination determination unit 54, and a connection destination provision unit 55. The manager registration unit 51 registers the manager connection information (IP address and the like) transmitted when the managers 3 and 4 are activated in the manager information storage unit 53.

The keep alive transmission unit 52 reads the connection information of the managers 3 and 4 from the manager information storage unit 53 and transmits a keep alive signal to the managers 3 and 4 at a certain frequency.
A failure in the integrated management function of the managers 3 and 4 is detected based on whether or not there is a response from the managers 3 and 4. The manager information storage unit 53 stores information related to the managers 3 and 4, for example, connection information.

The connection destination determination unit 54 determines one manager from the registered managers 3 and 4 according to the connection order (policy) and outputs the connection information to the connection destination provision unit 55. In this example, the connection order to connect to the manager is determined based on the manager failure and load. For example, the connection destination determination unit 54 acquires the status of the managers 3 and 4 from the keep alive transmission unit 52,
For example, when a failure occurs in the integrated management function of the manager 3 and the manager 4 is operating normally, the connection order is determined so that the manager 4 is connected. Further, when both the managers 3 and 4 are normally driven, the connection destination determination unit 54 gives priority to a manager with a light resource load based on a resource load such as a manager CPU usage rate or a memory usage rate. The connection order is determined so as to connect to.

When the connection destination providing unit 55 receives a connection request from the client terminal 1, the connection destination determining unit 5
The manager connection information (IP address) determined by 4 is returned to the client terminal 1. The client terminal 1 connects to either the manager 3 or the manager 4 based on the connection information acquired from the connection destination providing unit 55 of the naming server 5. After the connection, the client terminal 1 displays the network configuration and state on the display based on the management information received from the manager 3 or the manager 4, or notifies the administrator of the failure information using an alarm function or the like. The client terminal 1 may temporarily store connection information of the manager connected last time, and may connect to the manager based on the stored manager connection information when the naming server 5 cannot be used due to a failure. Thereby, even if the naming server 5 fails, it becomes possible to connect to the manager.

Next, a process in which the manager 3 collects network management information will be described with reference to FIG. In this example, the manager 3 collects management information of its own LAN-I and the LAN-II network of the manager 4.

The network information collection unit 32 of the manager 3 illustrated in FIG. 5 requests LAN-I management information from the in-LAN information collection unit 35. The LAN internal information collection unit 35 requests management information from the agents of the network elements 6a and 6b connected to the LAN-I. The agents of the network elements 6a and 6b return management information to the LAN information collection unit 35. LAN
The internal information collection unit 35 returns the management information of the network elements 6 a and 6 b to the network information collection unit 32. In this way, the network information collecting unit 32 of the manager 3 has its own LA.
Obtain N-I management information.

Next, the network information collection unit 32 sends a LAN-II to the remote LAN information acquisition unit 37.
Request management information. The remote LAN information acquisition unit 37 requests LAN-II management information from the in-LAN information providing unit 48 of the manager 4 connected to the LAN-II. The LAN internal information providing unit 48 requests LAN-II management information from the LAN internal information collecting unit 45. LA
The intra-N information collection unit 45 requests management information from the agents of the network elements 6c and 6d connected to the LAN-II.

The agents of the network elements 6c and 6d return management information to the LAN internal information collection unit 45. The LAN internal information collection unit 45 returns the management information of the network elements 6c and 6d connected to the LAN-II to the LAN internal information providing unit 48 via the LAN internal information storage unit 46. The LAN internal information providing unit 48 sends the network element 6c, the remote LAN information acquiring unit 37 of the manager 3 to the network element 6c,
6d management information is returned. The remote LAN information acquisition unit 37 of the manager 3 returns the management information of the LAN-II network elements 6 c and 6 d to the network information collection unit 32.

The network information collection unit 32 creates network management information obtained by collecting LAN-I and LAN-II management information. The communication unit 31 transmits the network management information created by the network information collection unit 32 to the client terminal 1 in response to a request from the client terminal 1. The client terminal 1 receives the LAN-I and LAN received from the manager 3
-Based on the management information of II, the network configuration and status are displayed on the display, and the fault information is notified to the administrator by an alarm function.

In the same manner as the manager 3, the manager 4 collects its own LAN-II management information, collects LAN-I management information from the manager 3, and aggregates these management information to create a network. Create management information for.

As described above, according to the network management system 100 according to the present invention, the manager 3 acquires management information of network elements connected to its own network and acquires management information of network elements connected to other networks. , 4 are monitored, and a naming server 5 is provided for determining a connection destination manager in response to a request from the client terminal 1.

As a result, the wide area network can be managed by using any of the managers 3 and 4 installed for each network. Therefore, even if one manager fails, the wide area network can be managed by another manager, and the network trust can be managed. Can be improved. Furthermore, it is possible to realize a redundant configuration capable of continuing the function of integrating and managing network management information without using a standby device. In addition, since the naming server 5, the manager does not need to be aware of which manager among the plurality of managers 3 and 4, and the burden on the manager can be reduced.

In addition, the naming server 5 automatically switches the connection destination provided according to the failure of the managers 3 and 4 and the load state, so that the manager is aware of the change of the connection destination even when the manager in use fails. Since the business can be continued as it is not necessary, the reliability of the network management system is improved.

In the above-described network management system 100, an example of managing two LAN-Is and LAN-IIs has been described. However, the present invention is not limited to this example, and two or more LANs, that is, LAN-I and L-L
A system for managing a network having AN-II, LAN-III,..., LAN-N (N is an integer) may be used.

Next, an example in which the present invention is applied to a job management system in addition to a communication network management system will be described. The job management system is a system that controls activation and termination of a plurality of jobs, for example, batch processing, and reports job execution and termination.

In general, a job management system includes a job execution server that executes scripts and batch files as jobs, a device that manages job execution status via an agent of the job execution server (hereinafter referred to as a job management manager), and a job operation management screen. It consists of client terminals that provide The administrator checks the job execution status in real time from the client terminal.

When a large-scale job management system is formed by linking a plurality of tasks, the job network is often divided into several sub job networks, and a job management manager is provided for each sub job network to perform distributed management.

For example, FIG. 6 shows a large-scale job management system 200 as the second embodiment according to the present invention.
It is a block diagram which shows the example of a structure. The job management system 200 illustrated in FIG. 6 includes a job network 11 including three sub job networks 11A to 11C, and a naming server 5.
A and the client terminal 1A. The sub job network 11A includes a job management manager 7 and job execution servers 7a and 7b. The sub job network 11B includes a job management manager 8 and job execution servers 8a and 8b. The sub job network 11C includes a job management manager 9 and a job execution server. 9a and 9b. The job management managers 7 to 9 are examples of management servers.

The naming server 5A monitors the status of each of the job management managers 7 to 9, and determines a connection destination as one of the job management managers 7 to 9 in response to a request from the client terminal 1A. For example, the naming server 5A registers connection information transmitted from the job management managers 7 to 9 when the job management managers 7 to 9 are activated. Naming server 5A
For example, one job management manager 7 is selected from the job management managers 7 to 9, and connection information of the job management manager 7 is returned to the client terminal 1A in response to a request from the client terminal 1A. Thereby, the administrator who operates the client terminal 1A does not need to be aware of which job management manager among the plurality of job management managers 7 to 9, and can reduce the burden on the administrator.

The naming server 5A periodically transmits a keep alive signal for confirming that it is operating normally to the job management managers 7-9. Job management manager 7 ~
No. 9 responds to the keep alive signal when operating normally, and does not respond to the keep alive signal in the case of failure or overload. Thereby, the naming server 5A can determine whether or not the job management managers 7 to 9 are operating normally. Accordingly, the naming server 5A can switch the connection-destination job management manager provided to the client terminal 1A to a job management manager that is operating normally. The client terminal 1A provides a job operation management screen based on the execution status received from any of the job management managers 7-9.

The job management manager 7 collects the execution status of the job execution servers 7a and 7b in the sub job network 11A managed by the job management agent from the job agent. Further, the job management manager 7 acquires the execution status of the job execution servers 8a and 8b in the sub job network 11B from the other job management manager 8, and also from the job management manager 9 to the job execution servers 9a and 9b in the sub job network 11C. Get the execution status of.

Similarly, the job management manager 8 collects the execution status of the job execution servers 8a and 8b in the sub job network 11B managed by the job management agent 8 from the job agent, and acquires the execution status of the job execution servers 7a and 7b from the job management manager 7. The execution status of the job execution servers 9a and 9b is acquired from the job management manager 9.

Further, the job management manager 9 collects the execution status of the job execution servers 9a and 9b in the sub job network 11C managed by itself from the job agent and acquires the execution status of the job execution servers 7a and 7b from the job management manager 7. The execution status of the job execution servers 8a and 8b is acquired from the job management manager 8. As a result, the job management managers 7 to 9 can acquire the execution statuses of all the job execution servers 7a, 7b, 8a, 8b, 9a, and 9b connected to the sub job networks 11A to 11C.

The client terminal 1A connects to the job management manager determined by the naming server 5A, acquires the execution status of all job execution servers, and provides the administrator with a job operation management screen. As a result, even if one of the job management managers has a failure in the function that manages the job execution status in an integrated manner, the execution status of all job execution servers can be acquired by accessing other job management managers. become. Therefore, the execution status of the job execution server can be grasped by switching to an active job management manager without installing a standby device.

In the above-described job management system 200, the three sub job networks 11A to 11A
Although an example of managing 11C has been described, of course, the present invention is not limited to this, and a job management system having three or more sub job networks may be used.

The present invention is extremely suitable when applied to a network management system that manages devices connected to a network and a job management system that performs batch processing.

1, 1A ... Client terminal, 3, 4 ... Manager (management server), 5 ...
Naming server, 6a to 6d ... network element, 7-9 ... job management manager (management server), 7a, 7b, 8a, 8b, 9a, 9b ... job execution server, 1
00 ... Network management system 200 ... Job management system

Claims (5)

A management server that is installed for each network and acquires management information of devices connected to its own network, and acquires management information of devices connected to other networks;
A client terminal capable of communicating with each of the management servers;
A monitoring server that monitors each of the management servers and can communicate with the client terminal;
The client terminal requests a connection destination management server to the naming server,
The naming server that has received the request from the client terminal determines a management server as a connection destination based on a predetermined connection order,
The network management system, wherein the management server determined by the naming server transmits the acquired management information to the client terminal. The management server integrates and manages management information of devices connected to its own network and management information of devices connected to other networks, and transmits the integrated management information to the client terminal The network management system according to claim 1, further comprising: The said naming server is provided with the connection destination determination part which determines the said connection order | rank connected to the said management server based on the failure and load of this management server, It is characterized by the above-mentioned.
The network management system described in 1. The naming server is
A manager information storage unit for storing information about the management server;
A manager registration unit for registering connection information of the management server transmitted when the management server is started in the manager information storage unit;
Reads connection information of the management server from the manager information storage unit, transmits a keep alive signal to the management server, a keep alive transmission unit that detects a failure of the management server based on the presence or absence of a response from the management server,
A connection destination providing unit that returns connection information of the management server determined by the connection destination determination unit to the client terminal;
The connection destination determination unit obtains the status of the management server from the keep-alive transmission unit, determines one management server from the registered management servers based on the connection order, and outputs the connection information. The network management system according to claim 3, wherein: The management server
A connection relation information storage unit between LANs for storing a plurality of connection relation information connected to other networks;
A manager information storage unit for storing connection information of a plurality of management servers;
A LAN information collection unit that collects management information from devices to be managed in its own network;
A LAN information storage unit for storing management information collected by the LAN information collection unit;
Based on the connection relation information stored in the connection relation information storage section between the LANs and the connection information stored in the manager information storage section, communication is performed with a management server in another network, and management information of the other network is obtained. A remote LAN information acquisition unit to acquire;
In-LAN information providing unit that provides management information of its own network in response to a request from a management server in another network;
When the management server is activated, an activation communication unit that notifies the naming server of connection information of the management server;
A keep-alive receiving unit that responds to a keep-alive signal transmitted from the naming server;
In response to a request from the client terminal, the network information collection unit
The network management according to any one of claims 2 to 4, wherein the management information of the own network is collected from the N-internal information collection unit and the management information of another network is collected from the remote LAN information acquisition unit. system.

Images