JP3534645B2 - 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 network management system, and more particularly to a network management system including an agent, an active system manager, and a standby system manager, and a failure occurrence in the agent is detected by the active system manager or the standby system manager. The present invention relates to a technique for reducing the amount of communication between the active system manager and the standby system manager in a network management system that detects and notifies an external device of the occurrence of the failure.

[0002]

2. Description of the Related Art A network agent (managed device, here simply referred to as "agent") and a network manager (managed device, here simply referred to as "manager") are communicatively connected by a network, and a fault occurring in the agent is managed by a manager. 2. Description of the Related Art A network management system having a function of detecting the above-mentioned information and notifying it to the outside is conventionally known. At this time, a configuration in which the manager is duplicated in the active system and the standby system and the management function is made redundant is often adopted.

FIG. 6 is a diagram showing the overall configuration of a network management system having dual managers. As shown in the figure, the network management system 100 has a configuration in which a plurality of agents 102, an active system manager 108, and a standby system manager 110 are connected to a network 106. Such network management system 1
According to 00, when the active system manager 108 manages the agent 102, the active system manager 108
If a failure occurs in the
Manages the agent 102. Thus, the network management system 100 ensures the reliability of the network system.

As a method of switching the active system manager 108 to the standby system manager 110, for example, Japanese Patent Laid-Open No. 3-25
In Japanese Patent No. 5556, the active system manager 108 is provided with a failure detection means, and this means is used as the active system manager 108.
When the occurrence of a failure is detected, the spare system manager 110 is notified to that effect, so that
The technology for launching

[0005]

However, in this technique, for example, when the failure detecting means itself provided in the active manager 108 fails, or when the power of the active manager 108 is cut off, the active manager 108 When a serious failure occurs in 108, the standby system manager 110 may not be notified of the failure occurrence. In this case, the active system manager 108 and the standby system manager 110 are not reliably switched, and the reliability of the network system cannot always be sufficiently ensured.

In this regard, in the management system according to Japanese Patent Laid-Open No. 6-197112, a failure recognition section is provided in the active system manager 108 and a failure detection section is also provided in the standby system manager 110, so that the failure of the standby system manager 110 occurs. The detection unit periodically inquires of the active system manager 108 whether or not a failure has occurred. By doing so, it is determined that the active system manager 108 has a failure not only when the active system manager 108 responds to the inquiry, but also when the active system manager 108 does not respond even after a predetermined timeout time elapses. be able to. Therefore, even if a serious failure occurs in the active manager 108, the active manager 108 and the standby manager 1
It is possible to reliably switch between 10 and 10, and it is possible to ensure the reliability of the network system.

However, the above-mentioned Japanese Laid-Open Patent Publication No. 6-19711.
In the management system according to Japanese Patent No. 2, the standby system manager 1
Since the inquirer 10 periodically inquires of the active system manager 108 whether or not a failure has occurred, there is a problem that the inquiries increase the traffic of the network 106.

The present invention has been made in view of the above problems, and an object thereof is to reduce a communication amount between an active system manager and a standby system manager while causing a serious failure in the active system manager. Even in the case, it is to provide a network management system capable of reliably switching managers.

[0009]

In order to solve the above problems, in the network management system according to the first aspect of the present invention, the network management system includes an agent, an active system manager and a standby system manager, and a fault occurs in the agent. In the network management system for detecting the occurrence of the failure to the external device by the active system manager or the standby system manager, the agent includes an agent failure occurrence detecting unit for detecting a failure occurrence in the agent, Fault occurrence information transmitting means for transmitting fault occurrence information to the active system manager and the standby system manager when a fault occurrence is detected by the agent fault detection means, wherein the active system manager is An active system failure that receives the failure occurrence information from the agent The standby system manager includes an occurrence information receiving unit and an active system external notification unit that notifies the external device of a failure occurrence in the agent when the failure occurrence information is received. Standby system failure occurrence information receiving means for receiving information; manager failure occurrence detecting means for detecting failure occurrence in the active system manager when the failure occurrence information is received from the agent; A standby system external notification means for notifying the external device of the failure occurrence in the agent when the failure occurrence in the system manager is detected.

A network management system according to a second invention is the network management system according to the first invention, wherein a plurality of the agents are provided, and the manager failure occurrence detecting means has already detected the occurrence of a failure in the active manager. If the failure information is further received from one of the agents, it is confirmed whether the active manager has recovered from the failure, and the active manager has recovered from the failure. If it is confirmed, the notification by the standby system external notification means is stopped, and the active system notification means notifies the external device of the failure occurrence in the agent when the active system manager recovers from the failure. It will restart.

A network management system according to a third aspect of the present invention is the network management system according to the first or second aspect of the present invention, wherein the standby external notification means further notifies the external device of a failure occurrence in the active manager. To do.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a diagram showing the configuration of a network management system according to an embodiment of the present invention. As shown in the figure, in the network management system 10, the agents 12a and 12b, the active system manager 20, and the standby system manager 28 are connected by a network 18 so that they can communicate with each other. Network 18 is LA
It is configured to include N, WAN, public telephone network, or the like. The agents 12a and 12b include agent failure occurrence detection units 14a and 14b and a failure occurrence information transmission unit 16
a, 16b, and computers, network devices, etc. configured to include them, which are managed devices in the network management system 10. The agent failure detection units 14a and 14b detect the failure occurrences in the agents 12a and 12b, respectively. In addition, the failure occurrence information transmission units 16a and 16b are configured so that the agent failure occurrence detection units 14a and 14b use the agents 12a and 12b.
When it is detected that a failure has occurred, the failure occurrence information indicating that a failure has occurred is transmitted to both the active system manager 20 and the standby system manager 28 accordingly. At this time, the agent failure occurrence detection units 14a, 1
It is desirable to detect the type of failure occurring in 4b so that the failure occurrence information transmitted to the active system manager 20 or the standby system manager 28 also includes information indicating the type of failure.

The active system manager 20 comprises a manager failure occurrence recognizing section 22, a failure occurrence information receiving section 24, and an external communication section 26. The active system manager 20 mainly manages the agents 12a and 12b, and when a failure occurs in the active system manager 20, the standby system manager 28 replaces the agent 1 with the agent 1a.
Manages 2a and 12b.

Here, the failure occurrence information receiving section 24 receives the failure occurrence information transmitted from the agent 12a or 12b via the network 18. Also, the external notification unit 2
When the failure occurrence information receiving unit 24 receives the failure occurrence information from the agent 12a or 12b, 6 notifies the external device 36 that a failure has occurred in the agent 12a or agent 12b. The external device 36 is configured to include, for example, a display and a speaker, and according to the notification from the external notification unit 26, the agent 12a or 12b.
The administrator may be notified of the occurrence of a failure in. Further, the manager failure occurrence recognizing unit 22 checks whether or not a failure has occurred in itself upon receiving an inquiry from the standby system manager 28, and the check result is used as the standby system manager 28.
Reply to. Note that the active manager 20 automatically recovers the agents 12a and 12b when it recovers from the failure.
It has a function to restart the management of. Therefore, when the failure occurrence information is received from the agent 12a or 12b after recovery from the failure, the external notification unit 26 notifies the external device 36 of that fact again.

On the other hand, the backup system manager 28 comprises a failure occurrence information receiving section 30, an external notification section 32, and a manager failure occurrence detecting section 34. As described above, the standby system manager 28 is the active system manager 20.
In the event of a failure, the agents 12a and 12b are managed in place of the active system manager 20. Specifically, when a failure occurs in the agent 12a or 12b, normally, the active system manager 20 notifies the external device 36 of the failure occurrence, but the active system manager 20 also fails. Instead, the standby system manager 28 notifies the external device 36 that a failure has occurred in the agent 12a or 12b. At this time, the standby system manager 28 may also notify the external device 36 that the active system manager 20 has a failure. When notifying the occurrence of a failure, it is desirable to also notify the type of the failure. In this way, the administrator who uses the external device 36 can respond to the failure surely and quickly.

The failure occurrence information receiving section 30 is similar to the failure occurrence information receiving section 24 provided in the active system manager 20.
The failure occurrence information transmitted from the agent 12a or 12b is received via the network 18. At this time, the failure occurrence information receiving unit 30 uses the agent 12a or 12
When the failure occurrence information is received from b, the manager failure occurrence detection unit 34 is notified of the reception. When the manager failure occurrence detecting unit 34 is informed that the failure occurrence information is received from the failure occurrence information receiving unit 30, the failure occurrence recognizing unit 22 of the active system manager 20 is notified of the occurrence of a failure in the active system manager 20. About Network 18
Contact us via. 1) Fault occurrence recognition unit 2
Not only when a response indicating that a failure has occurred is received from the active system manager 20 from 2 but 2) when there is no desired response from the failure occurrence recognition unit 22 even after a predetermined time has elapsed,
It is determined that the active manager 20 has a failure. If the active manager 20 determines that a failure has occurred, the manager failure detection unit 3
4 is the agent 12a or 1 for the external notification unit 32
In 2b, a control signal for instructing to notify that a failure has occurred is transmitted. In response to this, the external notification section 32 reads the received failure occurrence information from the failure occurrence information receiving section 30, and based on the information, the agent 12a or 12
In step b, the external device 36 is notified that a failure has occurred. On the other hand, when the manager failure occurrence detection unit 34 determines that no failure has occurred in the active system manager 20, the manager failure occurrence detection unit 34 determines that the agent 1
A control signal instructing not to notify the occurrence of a failure in 2a or 12b is transmitted to the external notification unit 32. In this case, the external notification unit 32 does not notify the external device 36 of the reception, although the failure occurrence information reception unit 30 has received the failure occurrence information. Also, the transmission of the control signal,
Even when the notification from the external notification unit 32 to the external device 36 has already started. Therefore, the external notification unit 32
Stops the notification when it has already notified the failure occurrence in the agent 12a or 12b.

The operation of the network management system 10 having the above configuration will be described. First, the operation of the agents 12a and 12b will be described. FIG. 2 is a flowchart illustrating the operation of the agent 12a or 12b. Here, the flow chart will be described by taking the agent 12a as an example, but the same applies to the agent 12b. The flowchart shown in the figure shows a process executed by the agent 12a at a predetermined cycle corresponding to a machine cycle. In this process, the agent failure occurrence detection unit 14a checks whether or not a failure has occurred in the agent 12a (S101). ). When the agent failure occurrence detection unit 14a detects the failure occurrence in the agent 12a (S102), the failure occurrence information transmission unit 16
In a, the agent 12a transmits failure occurrence information indicating that a failure has occurred to the active system manager 20 (S103), and also sends the same information to the standby system manager 28 (S104).

Next, the operation of the active manager 20 will be described. FIG. 3 is a flow chart for explaining processing of failure occurrence information in the active system manager 20. As shown in the figure, in the active system manager 20, the failure occurrence information receiving unit 24 waits for the failure occurrence information transmitted from the agent 12a or 12b (S201). And
Upon receiving the information, the external notification unit 26 causes the agent 12 to
The external device 36 is notified of the occurrence of a failure in a or 12b (S202).

FIG. 4 is a diagram for explaining a failure occurrence inquiry process in the active system manager 20. As shown in the figure, in this processing, the manager failure occurrence recognition unit 2
2 is waiting for an inquiry from the manager failure occurrence detection unit 34 provided in the standby system manager 28 (S
301). Then, when the manager failure occurrence detection unit 34 makes an inquiry about the presence or absence of a failure in the active system manager 20, the manager failure occurrence recognition unit 22 investigates the existence or nonexistence of a failure in the active system manager 20 (S302). It responds to the manager failure occurrence detection unit 34 (S303). For example, when a failure occurs in the active system manager 20, information indicating the type of failure is returned, and when no failure occurs, information indicating that there is no abnormality is returned.

Further, the operation of the standby system manager 28 will be described. FIG. 5 is a flowchart illustrating the operation of the standby system manager 28. As shown in the figure, in the standby system manager 28, the failure occurrence information receiving unit 30 waits for the failure occurrence information transmitted from the agent 12a or 12b (S401). When the failure occurrence information is received, the manager failure occurrence detecting unit 34 is activated, and the manager failure occurrence recognizing unit 22 is notified to the active system manager 20.
Inquiries are made regarding whether or not the failure has occurred. When the manager failure occurrence recognizing section 22 returns information indicating the kind of failure or does not respond to this inquiry even after a predetermined time elapses, the manager failure occurrence detecting section 34 indicates that a failure has occurred in the active manager 20. Determine that
When the information indicating that there is no abnormality is returned, it is determined that the active system manager 20 is functioning normally. In this way, the standby system manager 28 diagnoses the active system manager 20 (S402). And the active manager 20
If it is determined that the error is normal (S403), the standby system manager 28 sends a control signal to the external notification unit 32 to instruct the external device 36 not to notify the external device 36 of the failure occurrence,
External notification unit 3 even if notification of failure has already been issued
The operation of No. 2 is stopped (S404). On the other hand, if it is determined that a failure has occurred in the active system manager 20 (S4
03), the standby system manager 28 is the active system manager 20.
Instead of this, a control signal instructing to notify the external device 36 of the occurrence of the failure is transmitted to the external notification unit 32. As a result, the external notification unit 32 notifies the external device 36 of the failure occurrence (S405).

The network management system 1 described above
According to 0, even when a serious failure occurs in which the active system manager 20 cannot notify the standby system manager 28 of the failure, the active system manager 20 ensures that the active system manager 20 can Since the management entity can be migrated to 28, the reliability of the network system can be improved. Further, since the number of times the standby system manager 28 inquires of the active system manager 20 about the occurrence of a failure can be reduced, the communication volume between the two managers can be reduced. Of course, since the agents 12a and 12b transmit failure occurrence information to both the active system manager 20 and the standby system manager 28, the traffic of the network 18 increases at this part. However, agent 1
Since the failure occurrence information is transmitted from 2a and 12b only when a failure occurs in the agent 12a or 12b,
Such increase in traffic is not a big problem.

Inquiries about whether or not the active system manager 20 is restored are also made to the manager failure occurrence recognizing unit 22 from the manager detecting unit 34 to the minimum necessary amount. 28
Therefore, the number of inquiries to the active system manager 20 can be reduced, and the traffic volume of the network 18 can be reduced.

The network management system 10 described above can be modified in various ways. For example, the inquiry from the manager failure occurrence detecting unit 34 to the manager failure occurrence recognizing unit 22 as to whether there is a failure does not need to go through the network 18, and a dedicated communication line is provided.
Other means may be adopted as the communication means as long as bidirectional communication is secured. In addition, the external notification unit 26,
Various means may be used for communication from 32 to the external device 36, and may be shared with the network 18, for example.

In the above description, the active manager 2
0 or a function provided in the standby system manager 28, the failure occurrence in the agent 12a or 12b is detected by the external device 3
6 has been mainly described, but in addition, the active manager 20 and the standby manager 28 inquire of the agents 12a and 12b about the current status, or the agents 12a and 12b periodically report the current status. You may add the function of displaying the state of (3) on a display not shown.

[0026]

As described above, according to the network management system of the first aspect of the present invention, when an agent detects a fault occurrence, the fault occurrence information is sent from the agent to the active system manager and the standby system manager. Since the standby system manager detects the failure occurrence in the active system manager upon receiving the failure occurrence information, it is possible to reduce the opportunity for the standby system manager to detect the failure occurrence in the active system manager. You can Therefore, it is possible to reduce the amount of communication between the standby system manager and the active system manager. In addition, since the standby manager detects the failure of the active manager from the outside, even if a serious failure occurs such that the active manager cannot notify the standby manager of its own failure. It is possible to reliably switch the management from the active manager to the standby manager.

Further, according to the network management system of the second aspect of the present invention, when the standby system manager starts managing the agent and then receives the failure occurrence information from the agent again, whether the active system manager has recovered from the failure? If it is recovered, the backup system manager stops the notification to the external device, while the active system manager restarts the notification to the external device. It is possible to switch the management entity from the standby system manager to the active system manager while reducing the number of times of checking the presence or absence of recovery as much as possible to reduce the communication volume between both managers.

Further, according to the network management system of the third aspect of the present invention, since the standby system manager further notifies the external device of the failure occurrence in the active system manager, the active system manager has a failure. The effect can be known on the external device side.

[Brief description of drawings]

FIG. 1 is a diagram showing an overall configuration of a network management system according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating the operation of an agent.

FIG. 3 is a flowchart illustrating the operation of the active manager.

FIG. 4 is a flowchart illustrating the operation of the active manager.

FIG. 5 is a flowchart illustrating the operation of the standby system manager.

FIG. 6 is a diagram showing a configuration of a network management system according to a conventional technique.

[Explanation of symbols]

10 network management system, 12a, 12b agent, 14a, 14b agent failure occurrence detection unit, 16a, 16b failure occurrence information transmission unit, 18 network, 20 active system manager, 22 manager failure occurrence recognition unit, 24, 30 failure occurrence information reception Department,
26, 32 external notification unit, 28 standby system manager, 3
4 manager failure occurrence detection unit, 36 external device.

Front page continuation (72) Inventor Shinsuke Yamamoto 2-3-3 Marunouchi, Chiyoda-ku, Tokyo Mitsubishi Electric Corporation (56) Reference JP 10-327150 (JP, A) JP 2-71644 ( JP, A) JP-A-6-197112 (JP, A) JP-A-6-348628 (JP, A) JP-A-8-163165 (JP, A) JP-A-11-27266 (JP, A) (58 ) Fields surveyed (Int.Cl. ⁷ , DB name) H04L 12/24 G06F 13/00

Claims (3)

(57) [Claims] 1. An agent, an active system manager, and a standby system manager are included, and a failure occurrence in the agent is detected by the active system manager or the standby system manager, and the external device is notified of the failure occurrence. In the network management system for notifying, the agent includes an agent failure occurrence detection unit that detects a failure occurrence in the agent, and the active system manager and the standby system manager when a failure occurrence is detected by the agent failure detection unit. A failure occurrence information transmitting means for transmitting failure occurrence information, wherein the active system manager receives an active system failure occurrence information receiving means for receiving the failure occurrence information from the agent, and the failure occurrence information. If received, the failure occurrence in the agent An active external notification means for notifying an external device, the standby system manager receiving a backup system failure occurrence information receiving means for receiving the failure occurrence information from the agent, and receiving the failure occurrence information for the agent In this case, manager failure occurrence detection means for detecting a failure occurrence in the active system manager, and when the manager failure occurrence means detects a failure occurrence in the active system manager, notifies the external device of the failure occurrence in the agent. A standby system external notification means, and a network management system comprising: 2. A plurality of the agents are provided, and the manager failure occurrence detecting means receives further failure occurrence information from one of the agents when a failure occurrence in the active manager has already been detected. Checking whether the active system manager has recovered from the failure, and when it is confirmed that the active system manager has recovered from the failure, stop the notification by the standby system external notification means, The system notification means restarts notification of a failure occurrence in the agent to the external device when the active system manager is recovered from the failure.
Network management system described in. 3. The network management system according to claim 1, wherein the standby system external notification means further notifies the external device of a failure occurrence in the active system manager.