JP2001067292A - State informing system for network management system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an informing system for a network management system for informing a managing device of state change information detected by a device to be managed while improving real time performance and reliability. SOLUTION: A device 1 to be managed on which the agent function of an SNMP protocol is mounted is provided with a means for transmitting TRAP information at the time of detecting the change of the inside state of its own device. A managing device 2 on which the manager function of the SNAP protocol is mounted is provided with a means for transmitting a GET-REQUEST command for requesting the transfer of detail information at the time of receiving the TRAP information from the device 1 to be managed. The device 1 to be managed is also provided with a means for transmitting a response command in response to the GET-REQUEST command in which the detail information of the detected state change is included.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a status notification method for a network management system that notifies status change information detected by an unmanaged device in the network management system to the management device.

[0002]

2. Description of the Related Art Conventionally, in a communication network, a network management system (NETWORK MANAGE) has been used.
MENT SYSTEM: NMS is a management apparatus in which a plurality of network devices (NETWOR) configuring a communication network are connected.
K ELEMENT: NE), various types of settings and controls are exchanged according to a prescribed message format and management protocol in advance, and operation and maintenance of the communication network are performed. In a network management system, a plurality of managed devices are managed by one or a small number of management devices.

In such a network management system,
There are two methods by which the management device manages the status change of the managed device. One is a method in which the management apparatus polls the managed apparatus at regular intervals to manage state changes, and the other is that the managed apparatus spontaneously detects a state change of its own apparatus. This is a method of notifying the management device of a status change.
The above method is applied to the network management protocol SN
MP (SIMPLENETWORK MANAGEME
When using NT PROTOCOL), the former sends a GET-REQUEST at regular intervals to the SNMP agent installed on the managed device to the SNMP agent installed on the managed device in order to manage state changes. GET-RESPONSE, which is the response, to manage the state change.

[0004] SNMP is a typical network management protocol, and has five basic commands used for communication between a manager and an agent: GET-REQUEST.
T, GET-NEXT-REQUEST, SET-RE
Various functions are realized by QUEST, GET-RESPONSE, and TRAP.

[0005] Here, GET-REQUEST is a command for extracting information from the managed device.
T-NEXT-REQUEST is GET-REQUEST
A command for extracting the next information after extracting the information once in ST, SET-REQUEST is a command for changing and setting the information of the managed device, and is all issued from the management device to the managed device. . In contrast, G
The ET-RESPONSE is a response command for returning information requested by the three REQUEST commands from the managed device to the management device. On the other hand, TRAP
This is a command called event information for notifying the management device of an event that has occurred on the managed device side.

Here, when SNMP is used as a network management protocol, a lower layer protocol is generally UDP (USER DATAGRAM PROTOC).
OL) is used. This UDP is a method in which data is transmitted from the own station apparatus to another station apparatus without establishing a session (connection) with a communication partner, and high-speed communication is possible because protocol processing is easy. However, on the other hand, reliability is low because retransmission control and the like are not performed. UDP
GET-REQUEST, GET-NEX of operations other than TRAP in SNMP which is an upper layer of
In T-REQUEST and SET-REQUEST, a REQUEST type command as a request and a RE
SPONSE type commands are used as a pair. That is, when the own station device transmits a request to the other station device, the own station device can receive a response from the other station device,
The own station device can confirm the reception status of the other station device, and compensates for the low reliability which is a disadvantage of UDP.

In the latter method of spontaneously notifying the management device of a state change, when the managed device detects a change in the state of the device due to, for example, the occurrence of a failure or the like, the management device notifies the management device of T.
By transmitting the RAP, the management device can manage the status change of the managed device in real time. Issuance of this interrupt is performed by the network management protocol SNM.
In P, it is called TRAP.

[0008] Here, in the invention described in "Network management method and system, management apparatus and program storage medium used in the system" described in Japanese Patent Application Laid-Open No. Hei 10-340411, a network management system using SNMP is used. When a state change occurs in the sub-network management apparatus, the network management apparatus transmits simplified TRAP information simply notifying that the state change has occurred, and the network management apparatus Issues a GET command to the management device. In the network management device, GET-RESP
The reception of the ONSE command is monitored, and if not received within a predetermined time, the GET command is reissued.

According to the invention described in Japanese Patent Application Laid-Open No. H11-154950, when a device status of a network managed device changes, a TRAP command is issued to the network management device. GET-REQUES triggered by TRAP
T command is sent and GET-RESPO
If no NSE command is received, the GET-R
It retransmits the EQUEST command.

However, in these inventions, when a backbone failure of the network or a power failure occurs, a number of managed devices issue a TRAP command to the management device. As a result, the GET-REQUEST command and a process of waiting for the monitoring of the response due to the GET-REQUEST command occur in parallel, and thus there is a problem that the load on the management apparatus is not reduced.

[0011]

SUMMARY OF THE INVENTION In a method in which a management device manages a state change of a managed device, the former GET-
In the polling method using a pair of REQUEST and RESPONSE as a response, it is difficult for the management device to manage a large number of managed devices in real time. This is because the polling interval time is at least delayed until the management device recognizes the state of the managed device. Furthermore, there is a situation in which the polling interval has to be lengthened as the number of managed devices increases. However, when the polling interval is shortened, there is a problem that the management device is overloaded and exceeds the processing capacity of the management device.

In the latter method, in which a status change is spontaneously notified from the managed device to the management device by detecting the status change, the RAP which is a response to the TRAP command is transmitted.
Since there is no ESPONSE, the managed device only notifies the management device unilaterally.
The managed device has no means for confirming that the management device has received the notification of the state change, and thus has a problem that information arrival confirmation and reliability are lacking. Therefore, it is necessary to improve the real-time property of state change collection by periodic polling and to improve the reliability of state change collection by interruption.

[0013]

According to the present invention, the TR
GET using AP operation in request / response pairs
-Used in combination with REQUEST and RESPONSE operations. According to the present invention, in a network management system using the SNMP protocol, a management device recognizes a state change such as a failure that has occurred in a managed device, and the management device automatically triggers a TRAP operation generated by the managed device as a trigger. From GET-REQ
Activate the UEST operation.

Thus, by combining the real-time property of the TRAP operation with the reliability of the GET-REQUEST operation, it is an object of the present invention to provide a state change generated in the managed apparatus by the management apparatus in real time and with high reliability. Further, in consideration of the case where the management device monitors a large number of managed devices, a timer value for confirming the delivery of TRAP and the corresponding GET-REQUEST operation, and By changing the number of retransmissions, it is possible to collect optimal network information even at the time of centralized reception of TRAP.

According to the status notification method of the network management system of the present invention, a plurality of managed devices constituting a communication network and a network management system for monitoring and controlling the managed devices using an SNMP communication protocol are used in an SNMP protocol. In a managed device equipped with an agent function, a detecting device for detecting a change in the state of the device itself, a TRAP information transmitting device, and a management device equipped with a manager function of the SNMP protocol include TRA.
P information analysis means, means for transmitting a GET-REQUEST command to the managed device triggered by reception of TRAP information, and monitoring for monitoring whether a response command is returned within a predetermined time. And means.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS (1) An embodiment of the present invention will be described. In the present invention, the GET-
Using REQUEST, state changes are collected. However, the management device sends a GET-REQUEST to the managed device.
Rather than periodically issuing the ST, the managed device transmits a GET-REQUEST upon receiving a TRAP operation transmitted to the management device.

Thus, the management device can manage the status change of the managed device in real time. In addition, since the managed device receives the GET-REQUEST operation as a response to the TRAP operation, the reliability of the TRAP operation can be improved. That is, it is possible for the management device to perform the state change management of the managed device in real time and with high reliability.

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

The management device and the managed device are usually connected by a control network for transferring management information. Through this control network, the management device analyzes and analyzes data provided from each managed device. A series of operations are executed, such as making a determination and instructing a specific operation to each NE. Here, the managing side is generally called a manager, and the managed side is called an agent.

FIG. 1 is a system configuration diagram of an embodiment of the present invention. Referring to FIG. 1, a managed device 1 equipped with an SNMP agent and a management device 2 equipped with an SNMP manager are connected via a communication path 3, and a TRAP or GET-RE is sent from the agent side to the manager side.
SPONSE is transmitted, and GET-REQUEST is transmitted from the manager side to the agent side.

Next, a functional configuration of the managed device 1 will be described with reference to FIG. FIG. 2 is a configuration diagram of the managed device according to the embodiment of the present invention. Managed device 1
Is a state change detection unit 11 that detects a state change of the own device.
A TRAP processing unit 12, a timer unit 16 for detecting a timeout of a TRAP timer, and a retransmission counting unit for performing retransmission processing, a GET-REQUEST / R
The main configuration is a plurality of functional blocks for performing ESPONSE processing, and a communication processing unit 15 for performing communication processing with the management device 2.

Hereinafter, each functional block will be described in detail.

The state change detector 11 detects a change in the state of the communication path to be connected or the apparatus itself. A state change storage unit 13 for storing the state change information at this time.
In the TRAP processing unit 1
Notify 2. The TRAP processing unit 12 transmits the notified state change information from the managed device 1 to the TRA for the management device 2.
The information is converted as P information and transmitted to the TRAP transmission unit 14.

The TRAP transmitting section 14 transmits TRAP to the communication processing section 15, and the communication processing section 15 transmits TRAP to the management device 2 via the communication path 3.

The TRAP processing unit 12 transmits the TRAP information to the TRAP
At the same time as sending to the AP transmission unit 14, the timer unit 16 operates to detect a timeout of the TRAP timer transmitted from the managed device 1 to the management device 2. Here, the timeout time is set to an appropriate value in advance.
And it has the TRAP storage part 17 which stores TRAP information.

Further, the managed device 1 has a retransmission counting section 18 for counting the number of retransmissions due to the occurrence of a timeout of the timer section 16. Here, the number of retransmissions due to the timeout of the timer section 16 is counted up. When the timeout of the timer unit 16 occurs, the retransmission counting unit 18 repeatedly requests the TRAP processing unit 12 to perform retransmission if the current number of retransmissions counted is equal to or less than the upper limit value.

The TRAP processing section 12 retransmits the TRAP information stored in the TRAP storage section 17 in response to the retransmission request. When the number of retransmissions exceeds the upper limit of the number of retransmissions, the retransmission counting unit 18
The value of the retransmission count unit 18 is cleared, and the TRAP operation is forcibly terminated.

Conversely, the GET is transmitted from the management device 2 to the transmission line 3.
The managed device 1 that has received the -REQUEST transmits the GET-REQUEST via the communication processing unit 15 to the GET-REQUEST.
Transmit to the QUEST receiver 19. GET-REQUEST
The ST receiver 19 receives the GET-REQUEST sent.
To the GET-REQUEST processing unit 20.

The GET-REQUEST processing unit 20 is a timer unit based on the received GET-REQUEST information.
6, clear the TRAP storage unit 17 and the retransmission count unit 18, complete the TRAP operation, read the status change information from the status change storage unit 13, and execute GET-RES.
It is transmitted to the PONSE processing unit 21.

At this time, the status change information stored in the status change storage unit 13 is stored in the status database 23 of the apparatus.
To be stored.

The GET-RESPONSE processing unit 21
The read state change information is transmitted to the management device 2 via the GET-RESPONSE transmitting unit 22 and the communication processing unit 15 as GET-RESPONSE information.

FIG. 3 is a configuration diagram of the management device according to the embodiment of the present invention.

The management device 2 includes a communication processing unit 31 for recognizing the TRAP transmitted from the managed device 1, a TRAP receiving unit 32 for receiving only the TRAP information, and a TRAP processing unit for actually analyzing the TRAP information. 33.

The TRAP processing section 33 sends a GET-REQUEST from the management apparatus 2 to the managed apparatus 1 by TRAP analysis.
It has a function to request transmission of ST.

When the GET-REQUEST processing unit 34 determines that the GE
At the same time as sending the T-REQUEST information to the GET-REQUEST transmitting unit 35, the timer unit 36 for detecting the timeout of the GET-REQUEST timer operates. Then, the transmitted information is referred to as GET-REQUEST.
The data is stored in the storage unit 37.

The retransmission counting section 38 for counting the timeout of the timer section 36 counts up when the timeout of the timer section 36 occurs. Then, if the number of retransmissions is equal to or less than the upper limit,
It requests the T-REQUEST processing unit 34 for retransmission.

The GET-REQUEST processing unit 34 retransmits the GET-REQUEST information stored in the GET-REQUEST storage unit 37 in response to the retransmission request.
In addition, when the retransmission count value exceeds the retransmission upper limit value, the retransmission count unit 38 sets the timer unit 36, the GET-REQUEST storage unit 3
7. Clear the retransmission count section 38 and get the GET-REQU
It has a function to forcibly terminate the EST operation.

The GET sent from the managed device
-When the GET-RE is received,
Timer unit 36, TRAP storage unit 3 based on SPONSE
7. Clear the retransmission count section 38 and get the GET-REQU
Complete the EST operation. Then, the device status is stored in a database 41 for storing the status of each device.

(2) Description of the operation of the embodiment In the following five cases of the sequence of the present invention, (A) in the case of a normal state change collection sequence, and (B) abnormality in the managed device which does not exceed the retransmission upper limit value In the case of (C), if the managed device has an error exceeding the retransmission upper limit value, (D) if the management device has an error not exceeding the retransmission upper limit value, (E)
A description will be given of five cases in which an abnormality exceeds the retransmission upper limit value in the management device. In addition, each sequence diagram,
These are shown in FIGS. 4, 5, 6, 7, and 8, respectively.

(A) The case of a normal state change collection sequence.

The normal state change collection sequence will be described below. FIG. 4 is a normal state change collection sequence diagram in the embodiment of the present invention. The managed device is
Upon detecting the occurrence of a status change in the own device, the management device transmits a TRAP (status change occurrence notification) to notify the management device that the status change has occurred.

At this time, the content of the TRAP information notifies the managed device that only a status change has occurred. TRA
The management device that has received P sends a GET to the managed device.
-Send a REQUEST (state collection request) to request acquisition of detailed state change information of the managed device. This GET
-The managed device that has received the REQUEST transmits detailed status change information to the management device as a GET-RESPONSE. The management device recognizes detailed status change information of the managed device by receiving the GET-RESPONSE.

Next, a normal state change collection process in the managed device and the management device will be specifically described with reference to FIGS.

When the managed device 1 detects a change in the status of its own device, it notifies the TRAP processing unit 12 of the occurrence of the status change and stores it in the status change storage unit 13 to store the latest status of the device. And update.

The TRAP processing unit 12 sends the status change information of the managed device to the TRAP transmitting unit 14 and, at the same time, detects a timeout of the TRAP timer transmitted from the managed device 1 to the management device 2. The part 16 operates. Then, the TRAP information is stored in the TRAP storage unit 17. Further, the TRAP transmission unit transmits the TRAP to the management device 2 via the communication processing unit 15. In the management device 2, the information transmitted from the managed device 1 is transmitted to the TRAP receiving unit 32 via the communication processing unit 31, and the TRAP processing unit 33 analyzes the received TRAP information.

Then, the GET-REQUEST information is transmitted to the managed device 1 corresponding to the received TRAP, and the information is stored in the GET-REQUEST storage unit 37 at the same time. Further, the timer unit 36 for monitoring a response to the GET-REQUEST is operated.

When the non-management device receives the GET-REQUEST, the GET-REQUEST processing unit 20 executes the timer unit 1 based on the received GET-REQUEST information.
6. Clear the TRAP storage unit 17 and retransmission count unit 18 to complete the TRAP operation. Then, the state change is read from the state change storage unit 13 and the GET-R
Send to the ESPONSE processing unit 21.

The GET-RESPONSE processing unit 21 uses the state change information as GET-RESPONSE information, and the communication processing unit 15 sends the G to the management device 2 via the communication path 3.
Send ET-RESPONSE.

In the management apparatus of FIG. 3, the received GET-R
Based on ESPONSE, the timer section 36, the TRAP storage section 37, and the retransmission count section 38 are cleared, and the GET-REQ
Complete the UEST operation. Then, the status of the managed device is stored in the database 41 for storing the status of each device. (B) In the case of an abnormality that does not exceed the retransmission upper limit value in the managed device.

Next, a description will be given of a process in the case where the managed device does not exceed the retransmission upper limit value when an abnormality occurs. FIG.
FIG. 7 is a state change collection sequence diagram at the time of TRAP abnormality when the retransmission upper limit is not exceeded in the embodiment of the present invention.

Referring to FIG. 5, when the managed device detects the occurrence of a status change of its own device, transmitting a TRAP indicating that the status change has occurred to the management device is the same as in case (A). It is.

At this time, TRAP (state change occurrence notification)
The content is information for notifying only that a state change has occurred in the managed device. On the other hand, the management device
Is not successfully received, the management device cannot transmit a GET-REQUEST (state collection request) to the managed device. Therefore, the managed device cannot receive the GET-REQUEST from the managed device, the TRAP timer of the managed device times out, and retransmits the TRAP to the managed device.

The retransmission is repeated, and the monitoring device normally receives the TRAP from the managed device within the specified number of retransmissions.
Send EQUEST and request detailed state change information.

The managed device that has received the GET-REQUEST sends detailed status change information to the management device as G
Transmit as ET-RESPONSE. Management device is G
By receiving the ET-RESPONSE, detailed status change information of the managed device is recognized.

The state of FIG. 5 will be described with reference to FIGS. The managed device 1 detects the status change of its own device and performs TRAP.
Is the same as (A).

Here, the TRA transmitted to the management device
If P cannot be transmitted to the management device for some reason, a timeout occurs in the timer unit 16 and the retransmission counting unit 18 is counted. Retransmission count unit 18
If the retransmission upper limit value is not reached in
2 requests retransmission. The TRAP processing unit 12 repeats the retransmission of the TRAP information stored in the TRAP storage unit 17 in response to the retransmission request from the retransmission counting unit 18.

In the management apparatus 2, the communication path 3
When the communication processing unit 31 recognizes the TRAP transmitted via the communication processing unit 31, the communication processing unit 31 transmits the TRAP to the TRAP receiving unit 32.
And the TRAP receiving unit 32 sends the transmitted TRAP to TR
It is sent to the TRAP processing unit 33 as AP information.

The TRAP processing unit 33 sends a G
Sending the ET-REQUEST is a GET-REQUEST.
The QUEST processing unit 34 is notified and the GET-REQUES
The managed device 1 via the T transmission unit 35 and the communication processing unit 31
, The GET-REQUEST timer is started by the timer unit 36 at the same time as the GET-REQUEST is transmitted. Then, the GET-REQUEST information is
It is stored in the T-REQUEST storage unit 37.

The non-management device 2 sends a GET-
GET-RE sent when REQUEST is received
The timer section 16, the TRAP storage section 17, and the retransmission count section 18 are cleared based on the QUEST information, and the TRAP operation is completed. Thus, G within the specified number of retransmissions
Since the ET-REQUEST has been received, the managed device 2
The transmission processing of the TRAP ends, and the received GET-R
A predetermined response process to EQUEST, that is, a state change is read out from the state change
Send PONSE.

In the management device, the received GET-RE
Timer unit 36, TRAP storage unit 3 based on SPONSE
7. Clear the retransmission count section 38 and get the GET-REQU
Complete the EST operation. Then, the status of the managed device is stored in the database 41 for storing the status of each device.

(C) In the case of an abnormality in the managed device exceeding the retransmission upper limit value.

Next, a description will be given of a state change collection sequence in the case where an abnormality exceeds the retransmission upper limit value in the managed device.

FIG. 6 is a state change collection sequence diagram when the TRAP retransmission upper limit value is exceeded in the managed device 1 according to the embodiment of the present invention.

Referring to FIG. 6, when the managed device detects the occurrence of a state change in the managed device, the managed device transmits a TRAP to the management device to notify the management device that the state change has occurred. . At this time, the TRAP content is information for notifying only that a state change has occurred in the managed device. On the other hand, when the management device has not normally received TRAP, the management device cannot transmit a GET-REQUEST (state collection request) to the managed device. Therefore, the managed device cannot receive the GET-REQUEST from the managed device, the TRAP timer of the managed device times out, and retransmits the TRAP to the managed device. While repeating this retransmission, if the number of retransmissions of the managed device exceeds a preset upper limit, the managed device forcibly terminates the TRAP operation.

The specific processing of each managed device and management device is omitted.

(D) A state change collection sequence in the case of an abnormality that does not exceed the retransmission upper limit value in the management device will be described.

FIG. 7 is a state change collection sequence diagram when the GET-REQUEST is abnormal when the retransmission upper limit is not exceeded according to the embodiment of the present invention.

Referring to FIG. 7, when the managed device detects the occurrence of a status change in the managed device, the managed device sends a TRAP to the management device to notify the management device that the status change has occurred. . At this time, the TRAP content is information for notifying only that a state change has occurred in the managed device. On the other hand, the management device that has received the TRAP sends a GET-REQUEST to the managed device.
T to request detailed status change information of the managed device.

When the managed device cannot normally receive the GET-REQUEST, the managed device cannot transmit a GET-RESPONSE (status collection response) to the managed device. Therefore, the management device cannot receive the GET-RESPONSE from the managed device, the GET-REQUEST timer of the management device times out, and retransmits the GET-REQUEST. When this retransmission is repeated, the monitored device normally receives the GET-REQUEST from the management device, and the managed device transmits detailed status change information to the management device as a GET-RESPONSE. The management device recognizes detailed status change information of the managed device by receiving the GET-RESPONSE.

(E) Abnormality Exceeding the Retransmission Upper Limit in the Management Apparatus Next, a state change collection sequence in the case of an abnormality exceeding the retransmission upper limit in the management apparatus will be described. FIG.
GE when the retransmission upper limit of the embodiment of the present invention is exceeded
It is a state change collection sequence figure at the time of T-REQUEST abnormality. Referring to FIG. 8, when the managed device detects the occurrence of a status change in the managed device, the managed device transmits a TRAP to the management device to notify the management device of the occurrence of the status change. At this time, TRAP contents are
This is information for notifying only that a state change has occurred in the managed device. On the other hand, the management device that has received the TRAP transmits a GET-REQUEST to the managed device and requests detailed status change information of the managed device. When the managed device cannot normally receive the GET-REQUEST, the managed device sends a GET-R to the management device.
ESPONSE (status collection response) cannot be transmitted. Therefore, the management device receives the GET-R from the managed device.
Since ESPONSE cannot be received, the GET-R of the management device
EQUEST timer generates timeout and GET-
Retransmit REQUEST. When the retransmission is repeated and the number of retransmissions of the management device exceeds a preset upper limit, the management device performs GET-REQUEST.
Terminate the T operation.

FIG. 9 is a flow chart showing a series of steps up to now. FIG. 9 is a state change detection flowchart according to the embodiment of the present invention. Referring to FIG. The managed device 2 monitors whether there is a state change (step 1). If there is no state change, monitoring for the state change is continued again. (Step 1: NO) When there is a state change (Step 1: YES), TRAP is transmitted to the management device 1 (Step 2), and TRAP is transmitted.
The timer for use is started (step 3). Next, whether or not a timeout has occurred is monitored (step 104). If a timeout has occurred (step 104: YES), the retransmission counter is counted up (step 10).
9). If no timeout has occurred (Step 1
04: NO), it monitors whether a GET-REQUEST has been received (step 105).

If no GET-REQUEST has been received (step 105: NO), timeout is monitored again (step 104).

When the GET-REQUEST has been received (step 105: YES), the timer and the retransmission counter are cleared (step 106), and the GET-RESPON is received.
The SE is transmitted to the management device 1 (step 107). Next, the operation of the management device 1 will be described. The management device 1 monitors whether or not TRAP has been received (step 201). When no TRAP is received (step 20)
1: NO), the presence or absence of TRAP reception is monitored again (step 201).

When TRAP is received (step 20)
1: YES) is GET-REQ to the managed device 2
Send UEST (step 202) and GET-REQ
A UEST timer is started (step 203). After monitoring whether or not a timeout has occurred for GET-REQUEST, the management device 1 counts up a retransmission counter when a timeout has occurred (step 204: YES). (Step 209).

If a timeout has not occurred (step 204: NO), reception of GET-RESPONSE is monitored (step 205), and it is monitored whether the number of retransmissions exceeds the upper limit (steps 209 and 208).

If the number of retransmissions exceeds the upper limit (step 208: YES), the GET-REQUEST operation is forcibly terminated (step 210).

If the number of retransmissions does not exceed the upper limit (step 208: NO), GET-REQUEST is retransmitted (step 202), and a GET-REQUEST timer is started (step 203).

As described above, when the GET-RESPONSE is not received, the timeout is monitored again.
When T-RESPONSE is received (step 20)
5: YES), the timer and the retransmission counter are cleared (step 206), and the state change of the managed device 2 is recognized (step 207).

(Other Embodiments of the Invention) Next, other embodiments of the present invention will be described. In the present embodiment, TRAP in FIG.
A priority table for each monitored device is prepared in the AP processing unit 33, and processing is performed from an important managed device. In other words, priority is given to processing for the TRAP command for each managed device.

Hereinafter, an outline of the processing of another embodiment will be described. The management device 2 is configured by a plurality of managed devices 21 and 2
, 2N (not shown), and when a TRAP command is transmitted from the managed device 21 and the managed device 22, the management device receives the plurality of TRAPs. The TRAP processing unit 33 has a priority order table for each managed device, identifies the source managed device from the received TRAP commands, and sets the priority based on the managed device information. A series of state change detection processes are started in the order of commands related to the devices with the highest numbers.

Hereinafter, another embodiment of the present invention will be described with reference to FIGS. Management device 2 and managed device 1
In response to a command transmitted from the own device, a response command is transmitted within a predetermined time, and if no response command is received within the predetermined timer value, the minimum processing is repeated a predetermined number of times. In the embodiment of the present invention, further, a timer, a retransmission number counter,
The timer and the retransmission counter of the management device can be set arbitrarily.

The processing operation of each of the management device and the managed device will be described below. When a TRAP command is received from a plurality of managed devices at substantially the same reception timing,
The TRAP receiving unit 32 of the management device receives the plurality of received TRs.
For the AP information, the priority of the corresponding managed device is read from the priority table, and based on the priority, the TRAP information is processed, and the command of the high priority managed device is GET-processed. The transmission process of the REQUEST command is started. Further, the managed device having the second highest priority is processed at the next processing timing.

Here, the management device can process commands for a plurality of devices, but it is assumed that the number of processes that can be processed simultaneously has a certain limit due to its processing capability.

On the other hand, when the number of TRAPs received in a certain time period exceeds a certain value, the TRAP receiving unit 32
The corresponding GE via the ET-REQUEST processing unit 34
The value of the timer value for monitoring the T-RESPONSE command is changed to a longer value. In this way, the load on the management device can be reduced by changing the retransmission interval longer for the processing of the device with the lower priority. Also, T
When the number of TRAP commands received by the RAP receiving unit 32 further increases to a predetermined value or more, processing related to TRAP of a specific low-priority managed device is temporarily stopped, and retransmission processing is also stopped. Then, when the number of TRAPs observed by the TRAP receiving unit 32 becomes equal to or less than a certain value, the processing of the suspended TRAP command for the low-priority managed device is started.

When the managed device 1 detects a status change in its own device, it sends a TRAP command to the management device,
GET-R as a response corresponding to the transmitted TRAP
It has timer means for monitoring whether an EQUEST command is transmitted and retransmission processing means for retransmitting when a timeout occurs. The timer value and retransmission count number can be set from the management device. Management device T
When the RAP receiving unit 32 observes the number of TRAPs equal to or more than a certain value, the managed device 1 that transmitted the TRAP changes the timer value and the retransmission count.
By sending a REQUEST command, it is designated and notified. The managed device that has received the notification by the SET-REQUEST command reads the specified value, changes the settings of the timer value and the retransmission count, and performs a series of state change notification operations based on the values.

In this way, by setting the processing priority in accordance with the number of occurrences of TRAP and variably setting the timeout time of the timer of the managed device and the upper limit of the number of retransmissions, the network scale can be improved. Also, flexible network management corresponding to the processing load of the devices in the network becomes possible.

In a large-scale network, there is a high possibility that traffic is concentrated on the management device, and the time required for the management device to process the TRAP issued from the managed device becomes longer. Timeout time of the managed device according to the number of commands issued,
By changing the number of retransmissions, it is possible to reduce the number of retransmission TRAPs per time and reduce traffic in the communication network.

[0088]

The effect of the present invention is that, in the SNMP network management, the real time performance is superior to the status monitoring by only polling of the GET-REQUEST operation performed by the management device to recognize the status change of the managed device. .

Further, the reliability is superior to the status monitoring by only the TRAP operation performed by the management device to recognize the status change of the managed device.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram of an embodiment of the present invention.

FIG. 2 is a configuration diagram of a managed device according to an embodiment of the present invention.

FIG. 3 is a configuration diagram of a management device according to the embodiment of this invention.

FIG. 4 is a state change collection sequence diagram in a normal state according to the embodiment of the present invention.

FIG. 5 is a state change collection sequence diagram at the time of TRAP abnormality when the retransmission upper limit value is not exceeded in the embodiment of the present invention.

FIG. 6 is a state change collection sequence diagram at the time of TRAP abnormality when the retransmission upper limit is exceeded according to the embodiment of the present invention.

FIG. 7 is a state change collection sequence chart when a GET-REQUEST is abnormal when the retransmission upper limit is not exceeded according to the embodiment of the present invention.

FIG. 8 is a state change collection sequence diagram when a GET-REQUEST is abnormal when the retransmission upper limit is exceeded according to the embodiment of the present invention.

FIG. 9 is a state change detection flowchart according to the embodiment of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 managed device 2 management device 11 state change detection unit 12 TRAP processing unit 13 state change storage unit 14 TRAP transmission processing unit 15 communication processing unit 16 timer unit 17 TRAP storage unit 18 retransmission count unit 19 GET-REQUEST reception unit 20 GET- REQUEST processing unit 21 GET-RESPONSE processing unit 22 GET-RESPONSE transmitting unit 23 Device status database 31 Communication processing unit 32 TRAP receiving unit 33 TRAP processing unit 34 GET-REQUEST processing unit 35 GET-REQUEST transmitting unit 36 Timer unit 37 GET- REQUEST storage unit 38 Retransmission count unit 39 GET-RESPONSE reception unit 40 GET-RESPONSE processing unit 41 State database of each device

Claims (9)

[Claims] 1. A network management system for monitoring and controlling a plurality of managed devices constituting a communication network using an SNMP communication protocol, wherein the managed devices have an agent function of an SNMP protocol. Has a means for transmitting TRAP information for notifying the detection of the change in the state when the change in the state of the own apparatus has been detected. The management apparatus has a manager function of the SNMP protocol, Upon receiving the information, the receiving device has means for transmitting, to the managed device, a command for instructing to collect a state, with the reception of the TRAP information as a trigger, wherein the managed device includes a command for instructing to collect the state. Further comprising means for returning the detected state change information as a response command when the A status notification method for a network management system. 2. When the managed device transmits the TRAP information to the management device, the managed device stores the TRAP information, and receives a command instructing collection of the status from the management device within a predetermined time. 2. The status notification method according to claim 1, further comprising a unit that reads out the stored TRAP information and transmits the read TRAP information to the management device again when the management information cannot be obtained. 3. The managed device includes: a state change detecting unit that detects a change in a state of the managed device; a TRAP information processing unit that generates the TRAP command from the detected information; After sending to
If a command instructing the collection of the state cannot be received from the previous management device within a predetermined time,
Retransmission means for performing the retransmission processing of the TRAP command, means for repeating the retransmission processing until there is a response from the management device and counting the number of retransmissions, and when the number of retransmissions exceeds a predetermined upper limit, And means for stopping the retransmission processing.
2. The status notification method for a network management system according to item 1. 4. The management apparatus according to claim 1, wherein the management unit receives and analyzes the TRAP command received from the managed device, and collects detailed information with respect to a source managed device identified from the TRAP command. GET-R indicating
Transmitting means for transmitting an EQUEST command;
Monitoring means for monitoring whether a response to the command is returned within a predetermined time together with the transmission of the T-REQUEST command, and a predetermined number of times when the response to the command is not received within the predetermined time by the monitoring means. GET-REQUEST
4. The status notification method for a network management system according to claim 3, further comprising retransmission means for repeating retransmission of the ST command. 5. The management device receives the TRAP command received from the managed device, analyzes the received device, and collects detailed information with respect to the source managed device identified from the TRAP command. GET-R indicating
Transmitting means for transmitting an EQUEST command;
A timer unit for monitoring whether a GET-RESPONSE command which is a response to the T-REQUEST command is returned within a predetermined time together with transmission of the T-REQUEST command.
A retransmission count unit that repeats retransmission of a T-REQUEST command, and a GET-RESPO from the managed device.
4. The network management system according to claim 3, further comprising an update unit configured to extract status change information included in the NSE command when the NSE command is received, and update a status database corresponding to the managed device. method. 6. When the management device receives the TRAP command from a number of managed devices exceeding a predetermined value, the management device preferentially processes the TRAP information of the managed device having a predetermined high processing priority. The status notification method for a network management system according to claim 4, wherein the status notification is performed. 7. The management device, when receiving the TRAP commands from a number of managed devices exceeding a predetermined value, converts the TRAP command from a managed device having a predetermined low processing priority into a TRAP command. Corresponding G
7. The status notification method for a network management system according to claim 6, further comprising means for changing a setting of a retransmission interval and the number of retransmissions of an et-REQUEST command. 8. The management device, when receiving the TRAP command from a number of managed devices exceeding a predetermined value, sends the TRAP command to a managed device having a predetermined low processing priority. 8. The status notification method for a network management system according to claim 7, further comprising means for instructing a setting change of a timer value to be transmitted and a retransmission count. 9. The management apparatus further comprises means for transmitting a SET-REQUEST command for instructing the timer unit of the managed apparatus to set a timeout time and an upper limit of the number of retransmissions to arbitrary values. Item 9. A status notification method for a network management system according to Item 8.