JPH09172435A - Distributed managing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To distribute the load of a network by allowing a representative agent to pole in the state with each group closed and allowing agents to mutually monitor so as to automatically alternate the representative agent. SOLUTION: When an agent 5 transfers a frame to the agents 5 of high priority orders by referring to a monitoring table 7 and there is an agent 5 which does not respond to it but is restored, information that the agent 5 of the high priority order returns to be the representative agent is given to all of the other agents 5 within the group and a manager 2. On the other hand, when the authentication of a password and an access right in a frame is OK at the time of receiving the frame to which an access request from the manager 2 is set, data is set to the frame, transmitted toward the network by addressing its own agent and received. When difference between received data and specified data is over a prescribed value, the manager 2 is informed of abnormality. A polling function is provided for each agent 5 like this.

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 for managing agents connected to a network.

[0002]

2. Description of the Related Art Conventionally, in a system in which a manager and a plurality of agents are connected to a network, the manager performs all frame transmission for polling confirmation of each agent and frame polling for MIB polling for reading data. Was there.

[0003]

Conventionally, the manager individually carries out polling for existence confirmation and polling for reading data for each agent connected to the network, so that the network to which the manager is connected is enormous. There was a problem that it would be overloaded.

[0004] The present invention solves these problems,
The agent has a polling function, and the representative agent polls in a closed state in each group.
The purpose is to notify the manager only when a status change or an abnormality occurs and to perform agent mutual monitoring to automatically substitute the representative agent to distribute the network load and ensure reliability.

[0005]

Means for solving the problem will be described with reference to FIG. In FIG. 1, the manager 2 monitors a plurality of agents 5 connected to the network, and is composed of a monitoring information setting means 3 and an agent management table 4 here.

The monitoring information setting means 3 detects the agents 5 connected to the network and registers them in the agent management table 4, divides the registered agents 5 into groups, and selects a representative agent from the agents in the group. It is something that is decided.

The agent management table 4 is for registering information about the agent 5 to be monitored, which is connected to the network. The agent 5 is monitored (a terminal) by the manager 2 connected to the network, and is composed of a polling means 6, an agent management table 4, a monitoring table 7 and the like here.

The polling means 6 is for polling the frame to monitor the agent.
The monitoring table 7 is for registering agents to be monitored, priorities, monitoring time intervals T, and the like.

Next, the operation will be described. A frame received by the representative agent 5 when the monitoring information setting means 3 of the manager 2 divides the agents connected to the network into groups, sets the agents in the group, the representative agent, and the priority in a frame and transmits the frames to the representative agent. Table 7 for monitoring the addresses and priorities of agents in the group set in
When a frame is sent to the agent set in the monitoring table 7 and there is no response, or when an abnormal notification is received from the agent 5 in a frame, the manager is set The frame 2 notifies the agent 5 of the abnormality.

The monitoring information setting means 3 of the manager 2
Group the agents 5 connected to the network, set the agents 5 in the group and the representative agent in the frame and send the frames to the representative agent, and the representative agent 5 in the group set in the frame received When the address of the agent 5 and the determined priority are set in the monitoring table 7 and transmitted to the other agents 5 in the group to be saved, and a frame is transmitted to the agent 5 set in the monitoring table 7 and there is no response. Alternatively, the manager 2 receives a frame error notification from the agent 5.
The frame 5 notifies the agent 5 of the abnormality.

At this time, the agent 5 uses the monitoring table 7
When a frame is transmitted to the agent 5 having a higher priority by referring to, the agent 5 having the next priority is determined to be the representative agent, and all the other agents 5 and managers 2 in the group I am trying to notify you.

When the agent 5 refers to the monitoring table 7 and sends a frame to the agent 5 having a higher priority and there is no response from the frame, the agent 5 having a higher priority becomes the representative agent. All other agents 5 and managers 2 in the group are notified of the return.

Further, when a frame in which an access request is set from the manager 2 is received, and the password and access right authentication in the frame are OK, the data is set in the frame and transmitted to the self agent toward the network. The manager 2 is notified of the abnormality when the difference between the received data and the designated data exceeds a predetermined value or is less than the predetermined value.

Therefore, each agent 5 is provided with a polling function, the agent polls, and the manager 2 is notified only when a state change or an abnormality occurs, whereby the load of the network can be distributed.

[0015]

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

FIG. 1 shows a system configuration diagram of the present invention.
Here, groups 1, 2, and 3 surrounded by a dotted line indicate ranges of groups that are arbitrarily determined in advance. Here, it is assumed that the following determination is made.

[0017] Next, according to the order shown in the flowchart of FIG. 2, the procedure when the manager 2 detects the addresses of all the agents 5 connected to the network by broadcasting and sets them in the agent management table 4 will be described in detail.

FIG. 2 shows a setting flowchart (No. 1) of the present invention. In FIG. 2, S1 is a frame (Ec
ho) is created. This is, like a frame (Echo) of FIG. 5A described later, this is a destination address (ff: represents a frozen cast), a source address (# 00: an address of the manager 2), control data (Ech).
o: Create a frame in which response request data) is set.

S2 is transmitted by broadcasting. This is the broadcast frame (E
cho) to the network. The agent receives S3. In this, the agent 5 receives the frame (Echo) transmitted to the network in S2.

In step S4, it is determined whether a response is required. In the case of YES, the process proceeds to S5. In the case of NO, the own agent has determined that no response is necessary, so the received frame is ignored (discarded).

In S5, since it is determined that a response is required in S4, a frame (EchoReply) is created. This is the frame (EchoReply) of FIG. 5B described later.
, The destination address (# 00: the address of the manager 2), the source address (# 11: the address of the agent 5), the control data (EchoReply:
Create a frame with response data).

At step S6, the data is transmitted. This sends the frame (EchoReply) created in S5 to the network. In S7, the frame (Echo transmitted in S6
Reply) is received by the destination manager 2.

In step S8, registration in the management table is performed. This registers the address of the agent 5 extracted from the received frame (EchoReply) and the presence of the response of the frame (Echo) in the agent management table 4 of FIG. 1 (see FIG. 6).

By the above S1 to S8, the manager 2 broadcasts a frame, and the addresses of all the agents 5 to be monitored and the presence of the frame (Echo) will be described later by the response frame from the agent 5 connected to the network. It becomes possible to search and register as shown in the agent management table 4 of FIG.

In step S9, a frame (SNMP request) is created. This is the frame (SN) of FIG.
Like the MP request), the destination address (# 11: the address of the agent 5), the source address (# 00: the address of the manager 2), the protocol data (GetReq).
(uest PDU: PDU request data) is created.

At step S10, the data is transmitted. This sends the frame (SNMP request) created in S9 to the network. The agent receives S11. This is S1
The destination agent 5 receives the frame (SNMP request) transmitted to the network at 0.

In step S12, it is determined whether a response is required. In the case of YES, it progresses to S13. In the case of NO, the own agent has determined that no response is necessary, so the received frame is ignored (discarded).

In S13, a frame (SNMP response) is created because it is determined that a response is required in S12. This is the destination address (# 00: address of the manager 2), source address (# 11: address of the agent 5), protocol data (as shown in a frame (SNMP response) of FIG. GetResponseP
A frame in which DU: PDU response data) is set is created.

At step S14, the data is transmitted. This sends the frame (SNMP response) created in S13 to the network. S15 is the frame (SNM
The manager 2 of the destination receives the (P response).

In step S16, registration in the management table is performed.
This registers the address of the agent 5 extracted from the received frame (SNMP response) and the PDU (protocol data unit) in the agent management table 4 of FIG.

At step S17, the recognition of the agent is completed. Through S9 to S17, the manager 2 completes the recognition of the agent 5 connected to the monitored network by transmitting the frame to the monitored agent 5, receiving the reply of the PDU data and registering the frame. . Then, FIG. 3 and FIG. 4 are performed subsequent to (A).

FIG. 3 shows a flow chart of priority determination by the manager of the present invention. This is a flow chart when the manager 2 determines the priority order of the agents in the group, notifies the representative agent, and sets it.

In FIG. 3, S21 is divided into groups. This divides the agent 5 into groups 2 and 3, for example, as shown in FIG. S2
2 determines a representative agent for each group. For example, the representative agent of group 2 in FIG. 1 is determined as # 13.

In step S23, an alternative agent is determined for each group. For example, as substitute agents for the representative agent # 13 of group 2 in FIG.
1 is determined.

In step S24, a frame (agent) is created. This creates, for example, the following frame shown in the frame (agent) of FIG.

Destination address: representative agent # 13 Source address: manager # 00 Agent data: representative agent # 13 Alternative agent # 12 ... Priority order: # 13, # 12, # 11 Monitoring time interval: T1 S25 transmits to the network.

The representative agent receives S26.
S27 saves in the monitoring table. This is taken out from the frame received in S26 and stored as described below in the monitoring table 7 of FIG. 7 which will be described later.

The agent priority order monitoring time interval # 13 1 T1 # 12 2 T1 # 11 3 T1 S28 is transmitted by the representative agent to the alternative agent. This sends the contents of the monitoring table 7 set in S27 to the alternative agents (all agents other than the representative agent).

In step S29, the alternative agent saves it in the monitoring table. As described above, the manager 2 divides the agents connected to the network into groups, determines the representative agent, the priority order, and the monitoring time interval for each group, and transmits them to the representative agent of the group. Each representative agent saves this in its own monitoring table 7 and sends it to all other alternative agents to save the monitoring table 7. As a result, all agents have the same monitoring table 7 for each group, the alternative agent substitutes when the representative agent described later stops operating, and the representative agent accepts the abnormality information in the group when an abnormality occurs. It is possible to reduce the load between the networks and the load of the manager 2 by transmitting them to the manager 2 collectively.

FIG. 4 shows a priority order determination flowchart by the representative agent of the present invention. This is a flow chart when the representative agent notified from the manager 2 determines and sets the priority order of the agents in the group.

In FIG. 4, S31 is divided into groups. This divides the agent 5 into groups 2 and 3, for example, as shown in FIG. S3
2 determines a representative agent for each group. For example, the representative agent of group 2 in FIG. 1 is determined as # 13.

In step S33, a frame (agent) is created. This creates the following frame as a frame (agent), for example. -Destination address: Representative agent # 13-Source address: Manager # 00-Agent data: Representative agent # 13 S34 transmits to the network.

The representative agent receives S35.
In S36, the representative agent determines the priority order with the alternative agent. In this case, for example, agents # 12 and # 11 are used as the alternative agents, and the priority and the monitoring time interval are determined as follows.

Agent priority order Monitoring time interval # 13 1 T1 # 12 2 T1 # 11 3 T1 S37 is stored in the monitoring table. This is stored as follows, which is shown in the monitoring table 7 of FIG. 7, which will be described later, based on the priority order determined in S36.

Agent Priority Monitoring Time Interval # 13 1 T1 # 12 2 T1 # 11 3 T1 S38 is sent by the representative agent to the alternative agent. This sends the contents of the monitoring table 7 set in S37 to the alternative agents (all agents other than the representative agent).

In step S39, the alternative agent saves it in the monitoring table. As described above, the manager 2 divides the agents connected to the network into groups, and notifies the representative agent of each group,
The representative agent determines and saves the alternative agent in the group, the priority order and the monitoring time interval, and sends it to the other agents in the group to save it in the monitoring table 7. As a result, all agents have the same monitoring table 7 for each group, the alternative agent substitutes when the representative agent described later stops operating, and the representative agent accepts and collects the abnormality information in the group when an abnormality occurs. It is possible to reduce the load between the networks and the load of the manager 2 by transmitting the data to the manager 2 by using the above method.

FIG. 5 shows a frame example (1) of the present invention. FIG. 5A shows an example of a frame (Echo). This is an example of the frame created in S1 of FIG. 2 described above, and the following items shown in the figure are set.

Destination address: ff (representing broadcast) Source address: # 00 (representing manager address) Control data: Echo (representing response request data) FIG. 5B shows a frame (EchoReply). ) An example is shown. This is an example of the frame created in S5 of FIG. 2 described above, and the following items shown in the figure are set.

Destination address: # 00 (representing the address of the manager 2) Source address: # 11 (representing the address of the agent) Control data: EchoReply (representing the response data) FIG. An example of a frame (SNMP request) is shown.
This is an example of the frame created in S9 of FIG. 2 described above, and the following items shown in the figure are set.

Destination address: # 11 (representing agent address) Source address: # 00 (representing manager address) Protocol data: GetRequestPDU (P
(Representing DU request data) FIG. 5D shows an example of a frame (SNMP response).
This is an example of the frame created in S13 of FIG. 2 described above, and the following items shown in the figure are set.

Destination address: # 00 (representing manager address) Source address: # 11 (representing agent address) Protocol data: GetResponse PDU
(Representing PDU response data) FIG. 5E shows an example of a frame (agent).
This is an example of the frame created in S24 of FIG. 3 described above, and the following items shown in the figure are set.

Destination address: # 13 (representing the address of the representative agent of group 2) Source address: # 00 (representing the address of the manager) Agent data: representative agent: # 13 Alternative agent: # 12 Priority: # 13, # 12 ... Monitoring time interval: T1 FIG. 6 shows an example of the agent management table of the present invention.
This agent management table 4 is provided with a manager 2 and all or necessary parts for each agent,
The following items shown in the figure are registered and managed.

-Agent address: # 11-Presence or absence of frame (Echo) response: Yes-Presence or absence of frame (SNMP) response: Yes-MIB community name: A-Access right: read Here, the MIB community name is Password (MIB community name) and access right (r when manager 2 accesses data from a representative agent etc.)
Only the reference is permitted in the case of ead, and the reference / update is permitted in the case of read / write). For example, the agent # 11 becomes effective only when the manager 2 makes an access request in which the MIG community name "A" and read are set in the frame.
Is capable of setting predetermined data (MIB data) in a frame and returning it to the manager 2 for reading by the manager.

FIG. 7 shows an example of the monitoring table of the present invention.
This monitoring table 7 is for all agents 5 in the group.
Have the same content and set the priority and monitoring time interval for the agents in the group. For example, it is set as shown below.

Agent Priority Order Monitoring Time Interval # 13 1 T1 # 12 2 T1 Here, the agent with the priority order “1” is the representative agent. Agents with a priority of "2" or later are substitute agents, which substitute for the representative agent when the representative agent stops operating (described later with reference to FIGS. 8 and 9).

Next, the procedure for substituting the representative agent by frame reception will be described in detail according to the order shown in the flowchart of FIG. FIG. 8 shows a flow chart of an alternative process by frame reception of the present invention.

In FIG. 8, in S41, it is determined whether a fixed time has elapsed. In the case of YES, the process proceeds to S42. NO
If, then wait. In S42, the representative agent creates a frame (Echo) for each agent and sends it to the alternative agent. In this, the representative agent transmits a frame in which the address of the alternative agent is set as the destination address in the frame (Echo) of FIG.

In step S43, it is determined whether or not the alternative agent has received within a fixed time. If YES, the process proceeds to S44. In the case of NO, since it is determined that the representative agent has stopped operating, the representative agent is switched to the representative agent in S60, and the management table is updated and transmitted to the agents of other members in the group in S61 (described later).

In S44, it is determined that the frame (Echo) is received within the fixed time in S43, so it is determined whether a response is required. In the case of YES, the process proceeds to S45. In the case of NO, the received frame is ignored (discarded) because it is determined that no response is required.

S45 is a frame (EchoRepl
Create and send y). This is as described above with reference to FIG.
In the frame (EchoReply) of the following: -Destination address: address of representative agent # 13-source address: address of alternative agent # 12-control data: EchoReply: response data.

In step S46, it is determined whether or not the representative agent has received a frame (EchoReply) within a fixed time. In the case of YES, the process proceeds to S47. In the case of NO, the process proceeds to S56, where information indicating that there is no response from the alternative agent (abnormality occurrence) is saved and the manager 2 is notified (described later).

In step S47, since it is determined in step S46 that it has been received within the fixed time, the frame (SNMP request) to each agent is created and transmitted. This is in the frame (SNMP request) of FIG. 5C described above: Destination address: alternative agent address # 12 Source address: representative agent address # 13 Protocol data: GetRequestPDU: P
Set and send DU request data.

In step S48, it is determined whether or not the alternative agent has received within a fixed time. In the case of YES, the process proceeds to S49. In the case of NO, since it is determined that the representative agent has stopped operating, the representative agent is switched to the representative agent in S60, and the management table is updated and transmitted to the agents of other members in the group in S61 (described later).

In S49, since it is determined in S48 that the frame (SNMP request) is received within the fixed time, it is determined whether a response is required. In the case of YES, the processing proceeds to S50. In the case of NO, the received frame is ignored (discarded) because it is determined that no response is required.

In step S50, a frame (SNMP response) is created and transmitted. This is in the frame (SNMP response) of FIG. 5D described above: -Destination address: Address of representative agent # 13-Source address: Address of alternative agent # 12-Protocol data: GetResponse PDU:
Set and send PDU response data.

In step S51, it is determined whether the representative agent has received a frame (SNMP response) within a fixed time. Y
In the case of ES, it is recognized as normal in S52, and the process proceeds to S53. In the case of NO, the process proceeds to S57, in which information indicating that there is no response from the substitute agent (abnormality occurrence) is saved and the manager 2 is notified (described later).

In S53, the agent management table is transmitted to the alternative agent. In S54, the agent management table transmitted in S53 is received and stored.

In S55, in the case of NO in S51, the SNMP response in the agent management table is changed to none in S57 (see FIG. 6), and in response to the notification to the manager in S58, the frame is set and transmitted. The manager 2 receiving this frame takes out the message, displays a message (for example, the occurrence of an error in the alternative agent) on the screen, and collects a log.

In S60, it is determined that the frame is not received within the fixed time in NO in S43 or NO in S48, so it is determined that the representative agent is not operating (determined as a failure), and the next priority is substituted. Switch the agent to the representative agent.

In step S61, the agent management table is updated and sent to the agents of other members in the group to be updated. As described above, the representative agent transmits the frame to all the alternative agents in the group at regular time intervals, and when the alternative agent does not receive the frame within the constant time, the alternative agent with the next priority in the monitoring table 7 becomes the representative agent. At the same time as switching, the behavior of the representative agent is set to be defective in the agent management table 4 (specifically, no response of frame (Echo), no response of frame (SNMP)) and notified to other agents in the group. After that, the new representative agent can be notified when an abnormality occurs, and the new representative agent can notify the manager 2 of the abnormality occurrence of the alternative agent. Also,
When the representative agent does not receive a frame from the alternative agent within a certain time, the operation of the corresponding alternative agent in the agent management table 4 is defective (specifically, no response of frame (Echo), frame (SNMP)).
It is possible to notify the other agents in the group by notifying that there is no response) and notify the manager 2 of the occurrence of the abnormality of the alternative agent. These allow mutual monitoring between the representative agency and the alternative agent.

Next, the alternative process by polling will be described in detail according to the order shown in the flowchart of FIG. FIG. 9 shows an alternative processing flowchart by polling according to the present invention.

In FIG. 9, a step S71 decides whether or not the alternative agent has passed a certain time. If yes,
Proceed to S72. If no, it stands by. S72 is
A frame (SNMP request) to each agent is created (see (c) in FIG. 5).

In step S73, the agent having a higher priority than itself is transmitted. In S74, the representative agent having a higher priority receives the frame transmitted in S73.

In step S75, a frame (SNMP response) is created (see (d) in FIG. 5). S76 transmits. S
77 determines whether or not it has been received within a certain time after being transmitted in S73. In the case of YES, it is determined in S78 whether it is recovered, in the case of YES, the process proceeds to S79, and in the case of NO, the received frame is ignored (discarded). On the other hand, in the case of NO in S77, since there is no frame response from the representative agent, the representative agent is switched to the alternative agent of the next priority in S84, and the process proceeds to S79.

In S79, the agent management table is updated. This is because in the case of YES in S78, it is found that the representative agent has been restored, and therefore the representative agent in the agent management table is operating (specifically, with a response of frame (Echo), frame (SNM).
(P) response), or in the case of YES in S77, it is determined that the representative agent has stopped operating (has failed), so the representative agent in the agent management table is not operating (specifically, in the frame). (Ec
no response of ho), no response of frame (SNMP))
Update or update.

In step S80, the data is transmitted to the agents of other members, and the contents of the agent management table 4 are made the same for all the agents 5 in the group. S81 is S78
If it is determined that the representative agent has been restored with YES, the representative agent up to now returns to the alternative agent.

In S82, the representative agent receives the contents of the agent management table transmitted in S80. In S83, when the representative agent is found to be restored in YES in S78, the restored representative agent returns to its original state in response to the former representative agent returning to the alternative agent.

As described above, when the alternative agent transmits a frame to a higher priority agent at regular time intervals and does not receive a response frame within a constant time, it switches to the representative agent and informs the agent management table 4 to that effect. It is set and sent to the agents of all other members in the group to notify it, the new representative agent switches to monitor the alternative agent, and notifies the manager 2 when an abnormality is detected or an abnormality notification is received. It is possible to automatically switch the representative agent. When the representative agent is restored, the current representative agent is switched to the original alternative agent and the representative agent is restored.

Next, the alternative processing by MIB polling will be described in detail according to the order shown in the flowchart of FIG. FIG. 10 shows a MIB polling flow chart of the present invention.

In FIG. 10, S91 creates a frame (polling). This creates a frame (polling) in which the following items are set, as shown in FIG.

Destination address: Agent address # 13 Source address: Manager address # 00 Trap data: MIB name (for example, "total number of transmitted data") Polling interval Threshold trap type S92 is transmitted .

In S93, the agent receives the frame. In S94, it is determined whether the frame received in S93 has passed the polling interval. If YES, S9
In step 5, a frame (MIB request) is created. This creates a frame (MIB request) in which the following items are set, as shown in FIG.

Destination address: Agent address # 13 (address of own agent) Source address: Manager address # 00 MIB data: MIB community name Access right MIB name S96 transmits.

S97 is received within the same node (received within the node of the same agent). S98 creates a frame (MIB response). This is shown in FIG.
As shown in (c) of 1, a frame (MIB response) in which the following items are set is created.

Destination address: manager address #
00 • Source address: Agent address # 13 • MIB data: MIB response data S99 is transmitted.

The step S100 receives. S101 is S1
It is determined whether the MIB value (MIB data) received at 00 is greater than a threshold value designated in advance. In the case of YES, since it has been determined as abnormal, the process proceeds to S102. If NO, the process returns to S94 and is repeated.

In step S102, a frame (trap) is created. This is as shown in FIG.
Create a frame (trap) with the following items set. -Destination address: Manager's address # 00-Sender's address: Agent's address # 13-Trap data: Trap community name Trap type (for example, "total transmission data count exceeds threshold value") S103 transmits .

The manager receives S104. S1
05 displays the content set in the frame received in S104 as a message on the screen.

In step S106, the log is collected. By the above, the manager 2 sends a frame (polling) to the agent 5, and the agent 5 creates a frame based on this and sends and receives in the same node, and the resulting MIB value exceeds the threshold value, for example. Create a frame (trap) to notify the manager 2 of the abnormal content,
It is possible to display the content of the error occurrence as a message on the screen and collect the log.

FIG. 11 shows an example of the frame of the present invention (part 2).
Is shown. FIG. 11A shows an example of a frame (polling). This is an example of the frame created in S91 of FIG. 10 described above, and the following items shown in the figure are set.

Destination address: Agent address # 13 Source address: Manager address # 00 Trap data: MIB name (for example, "total number of transmitted data") Polling interval Threshold trap type FIG. 11 (b) Shows an example of a frame (MIB request).
This is an example of the frame created in S95 of FIG. 10 described above, and the following items shown in the figure are set.

Destination address: Agent address # 13 Source address: Manager address # 00 MIB data: MIB community name (equivalent to password) Access right (for example, read (reference)) MIB name (access target 11C shows an example of a frame (MIB response).
This is an example of the frame created in S98 of FIG. 10 described above, and the following items shown in the figure are set.

Destination Address: Manager's address #
00 • Source address: Agent address # 13 • MIB data: MIB response data FIG. 11D shows an example of a frame (trap). This is an example of the frame created in S102 of FIG. 10 described above, and the following items shown in the figure are set.

-Destination address: Manager's address # 00-Source address: Agent's address # 13-Trap data: Trap community name Trap type (for example, "total transmission data count exceeded threshold")

[0095]

As described above, according to the present invention,
Since the agent 5 is provided with a polling function and the agent performs polling and notifies the manager 2 only when a state change or an abnormality occurs, load distribution of the network can be achieved.

[Brief description of the drawings]

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

FIG. 2 is a setting flowchart (No. 1) of the present invention.

FIG. 3 is a priority order determination flowchart by the manager of the present invention.

FIG. 4 is a priority order determination flowchart by the representative agent of the present invention.

FIG. 5 is a frame example (1) of the present invention.

FIG. 6 is an example of an agent management table of the present invention.

FIG. 7 is an example of a monitoring table of the present invention.

FIG. 8 is a flowchart of an alternative process by frame reception of the present invention.

FIG. 9 is a flowchart of an alternative process by polling according to the present invention.

FIG. 10 is a MIB polling flowchart of the present invention.

FIG. 11 is a frame example (2) of the present invention.

[Explanation of symbols]

 2: Manager 3: Monitoring information setting means 4: Agent management table 5: Agent 6: Polling means 7: Monitoring table

Claims (5)

[Claims] 1. In a network management system for managing agents connected to a network, agents connected to the network are divided into groups, and agents in the groups, representative agents, and priorities are set in frames to become representative agents. A manager provided with means for notifying, and means for setting the addresses and priorities of the agents in the group set in the notified frame in the monitoring table and transmitting them to other agents in the group for storage. And a representative agent that sends a frame error notification to the above manager when the frame is sent to the agent set in the monitoring table and there is no response or when the agent receives a frame error notification Distributed management system comprising the. 2. In a network management system for managing agents connected to a network, the agents connected to the network are divided into groups, and the agents within the group and the representative agents are set in a frame and notified to the representative agents. The manager with the means and the address of the agent in the group set in the frame notified above and the priority of the agent are set and set in the monitoring table and sent to other agents in the group and saved. Means, and when a frame is sent to the agent set in the monitoring table and there is no response, or when an abnormality notification is sent from the agent in a frame, a representative error notification is sent to the manager in a frame. Distributed management system comprising: the stringent, the. 3. When a frame is transmitted to an agent with a high priority by referring to the monitoring table and no response is received, the agent with the next priority is determined as the representative agent and all other agents in the group are selected. 3. The distributed management system according to claim 1 or 2, further comprising an agent having means for notifying the agent and the manager. 4. A group indicating that an agent with a higher priority returned to the representative agent when the frame was sent to the agent with a higher priority by referring to the monitoring table and no response was returned from the frame 2. An agent provided with means for notifying all other agents and managers within.
The distributed management system according to claim 3. 5. A frame in which an access request is set from a manager is received, and when the password in the frame and the authentication of the access right are OK, the data is set in the frame and transmitted to the own agent toward the network. A distributed management system comprising: an agent provided with means for notifying the manager of an abnormality when the difference between the received data and the designated data exceeds a predetermined value or is less than the predetermined value.