JP3293719B2 - Network management system - Google Patents

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to a network management system.
Systems, especially FDDI (Fiber Distributed Data Int)
erface) network built by
Network management system, and FDDI
(Fiber Distributed Data Interface)
Network that has been set up using SNMP (Simple Net Management
 Protocol)-SMT (Station ManagemenT) protocol change
Exchange device (proxy device: hereinafter, PROXY device or
Simply referred to as “PROXY”).
Network management system. 2. Description of the Related Art FIG.
Ethernet Wort as an FDDI network
Schematically shown, the SNMP manager (administrative
The node MG as the managed nodes A to D and the FDD
They are connected via an I-ring R. [0003] In such an FDDI network,
Therefore, as shown in FIG.
When managing the node A as the agent, the node MG
Is the management information to be collected for node A (hereinafter referred to as
MIB = Management InformatinBase = Information or simply MI
B) as an “MIB request” packet
Do the trust. [0004] On the other hand, the node A holds in its own station.
Instruction from the node MG about MIB information of each protocol
Edited by the "MIB response"
[0005] As described above, the reply is used as the SNMP protocol.
FIG. 39 shows a configuration diagram of an SNMP frame. [0006] On the other hand, if the node A
In this case, since the node MG has a different protocol,
Cannot be directly managed. For this reason, as shown in FIG.
The node C in the DDI ring R is set to the SNMP-SMT protocol.
This is a PROXY device that performs a color conversion.
Node A via the device. [0008] That is, as shown in FIG.
MIB information to be collected from node A to node C
SNMP as "MIB request" packet
Send by protocol. This is the IP (Interna
l Protocol) The frame shown in FIG.
Is transmitted. The node C receives the packet received from the node MG.
Analyze the request and determine that it is an MIB request for node A
Then, as shown also in FIG.
A “MIB request” is transmitted to the node A.
This is shown in FIG. 41 by specifying the MAC address of node A.
This is performed by transmitting the indicated frame. [0010] The node A receives the MIB specified by the node C.
Edit the information and send the “MIB response
"Return as a packet (with the MAC address of node C
The node C receives the frame.
“MIB response” packet to node MG
Reply (Specify IP address of node A and send frame
I will). Therefore, the PROXY device has the following functions.
Need to be. (1) Supports both SNMP and SMT protocols
That. (2) Node and SM connected to the same FDDI ring
Able to communicate by T protocol. (3) Communication with the SNMP manager and SNMP protocol
What you can trust. (4) Connected to the same FDDI ring
Node MAC (Media Acce
ss Control) address and instance correspondence information.
SNMP management of identification information by SNMP protocol
Be notified to (5) From the SNMP manager, install the node identifier
Request to collect MIB information of SNMP protocol with sense
For the node requested by the SNMP manager.
Be able to convert to T protocol and transmit. (6) Notified by SMT protocol from each node
SNMP proto with MIB instance with identification instance
To be able to convert it to an SNMP manager and notify the SNMP manager. An SNMP manager and a PROXY device
The relationship with the position is as follows. (1) FDDI ring is one PROXY device per ring
Manage with. (2) To communicate with each node via the PROXY device,
The manager needs the PROXY device to identify each node.
Collect the set identification instances. (3) The manager recognizes the nodes managed by the PROXY device.
With the MAC address notified with the separate instance
to manage. (4) The manager sends an MI to each node.
B information is requested, the information collected from the PROXY device
PROX is a frame in which an instance of code identification is set.
Send to Y device. SUMMARY OF THE INVENTION
In a network management system, SNMP management
SMT agent on FDDI network in advance
Alternatively, register the node management information of the PROXY device or the like.
Thus, the FDDI network was managed. However, it is not limited to the FDDI network.
Network configuration is always dynamic.
The latest node management information is managed whenever the network configuration changes.
If you do not register with the management manager,
I couldn't continue. Further, the PROXY device cannot be managed.
(Failure / high load), according to the registered information,
Same as above because network management was continued
However, the management of the network could not be continued. Therefore, the configuration of the FDDI network is changed.
Or PROXY equipment cannot be managed
State, the latest node management information is
Network management will be suspended until you register
Had a problem. Accordingly, the present invention provides an SNMP protocol or
PROX with SNMP-SMT conversion protocol function
FDDI ring via Y device (protocol converter)
Network that is managed by an SNMP manager
Network management system, the FDDI network
Periodically and automatically detect the above nodes and the status of the nodes
Enables understanding and is suitable for dynamic networks
The purpose is to provide a system. Means and Action for Solving the Problems[1] Automatic detection / selection of PROXY device: According to the present invention
In a network management system, an SNMP manager
Polled each node and returned
SNMP node is detected from the node management information of
Whether these SNMP nodes are PROXY devices
Is determined from the second node management information, and the PROXY device
SMT on the FDDI ring when there are multiple locations
A PROXY device as a medium for managing nodes
One unit according to the specified priority based on node management information
It is characterized in that the user selects only one. That is, first, the SNMP manager
For example, a ping (response check)
Poll) to perform polling.
Collect node management information. Reply this node management information
FDDI on the support media among the nodes that do
The node is a managed node, that is, an SNMP node. The SNMP manager manages the
The SNMP node that has become the SNMP-SMT conversion function
A second NO indicates whether the PROXY device is supported.
Confirm using the password management information. If the device is identified as a PROXY device,
Management PROX if there is only one PROXY device
Y device, but if there are multiple devices, the third device
According to a predetermined priority determined according to the node management information.
PROXY device that manages one specific PROXY device
Select as The management PROXY is automatically performed by the above processing.
Recognize the device and manage the device under the management PROXY device.
Various kinds of information can be collected for the network. [0025][2] Automatic switching of PROXY device (failure
At birth): In the present invention, the network management system of the above [1]
In the SNMP manager, at the time of the polling,
The SNMP agent information is transmitted a predetermined number of times at regular time intervals.
If no reply is received even after continuous retries,
The PROXY device is determined to be a failure, and the
Automatically switch to another PROXY device
Wear. That is, a timer is used and the timer
Retry is performed when an outage occurs.
If the set maximum number of retries is exceeded, the node
It is determined that the function as the ROXY device is not possible, and S
Determined by the NMP manager based on the third node management information
The managed PROXY devices according to their priorities
I do. With the above processing, the PROXY device is changed.
And to the managed node via the changed PROXY device.
To collect information. In the present invention, instead of performing a retry,
In addition, the SNMP manager communicates with the management PROXY device.
The PROXY device has received an asynchronous information notification from
When it is determined to be harmful, other
Automatically switch to protocol converter
You. [0029][3] Automatic switching of PROXY device (for load
): In the present invention, the network management system
The SNMP manager sends the SNMP
Number of times P agent information was retried at regular time intervals
The managed PROXY device when the cumulative value of
Is determined to be in a high-load state, and according to the above-described priorities.
Automatically switch to another protocol converter.
Wear. In the present invention, the number of retries is
In addition, the SNMP manager performs the management protocol conversion.
Management PROXY device based on the load monitoring information from the device.
When it is determined that the device is in a high load state,
Automatically switch to another protocol converter
Is also possible. [0031][4] Automatic switching of PROXY device
):In the present invention, the above network management system
In addition, the SNMP manager operates at a preset time.
To the current managed PROXY device to another PROXY device
Can be switched automatically. That is, the time is periodically taken out from the system.
To extract the corresponding PROXY device. PR extracted
If the OXY device is different from the currently managed PROXY device, the PR
Switch OXY device. [0033][5] Automatic recognition of node connection status on ring
Sense:In the present invention, the above network management system
The SNMP manager is set in each node in advance.
Number of upstream / downstream nodes and ports / port types
Collect and analyze management information on port connection status
Automatically recognizes the node connection state on the ring
Can be That is, SNMP that can be collected from each node,
The management information of the SMT protocol is as follows. -Node ad of the positional relationship (upstream / downstream) through which the token passes
Node number and port type of the node Port type of the port to which the port is connected [0035] Each port of the node based on these management information
It is possible to analyze the connection destination node of the port. Analysis of connection partner
Is for the port upstream from the node (the own MAC)
After completing all nodes, identify the connection partner of the downstream port
I do. An upstream port (a node having only one port)
Nodes are not included).
The nodes traversed upstream by the number of
You. However, the concentrator is the node itself.
The node that becomes the positional relationship upstream from the concentrator
Are connected. Therefore, the number of nodes equal to the number of ports
If there is a concentrator on the way to
Is the upstream node connected to the concentrator.
Need to be considered. The concentrator says “No
Nodes are connected by the number of connection ports in the node -1 "
Therefore, the node that has been further traced by the number of nodes is the connection partner
Becomes Since the downstream port is one port,
For the connection destination, set the own node as the connection destination from the FDDI ring
Set as the connection destination of the own node in the set node
An unconnected node is the connection destination. [0039][6] Automatic node addition (management target addition)
:In the present invention, the above network management system
And the SNMP managers are upstream and downstream of each node.
Newly added on the ring by managing nodes
To recognize managed nodes and automatically add them to managed nodes
Can be. That is, the SNMP manager manages each node.
To periodically collect management and return node management information
However, if it is not in the current management node, it is determined as a newly added node
And add it to the management target. [0041][7] Automatic addition of rings (addition of managed objects)
:In the present invention, the above network management system
And the SNMP manager checks the FDDI resource of each node.
New to nodes by managing the number of connections
Recognizes a ring that has been set and automatically manages that ring
Can be added to That is, the SNMP manager periodically
Node management information is collected, and F of the returned node management information is collected.
DDI ring number information is different from current managed FDDI ring number
In this case, it is determined that there is a newly added FDDI ring, and
Add to [0043][8] Change from PROXY device to SNMP
Dynamic switching (when all PROXY failures):In the present invention,
PR to switch in the network management system
If the OXY device is not on the FDDI ring,
A node having an SNMP agent function on the ring
It can be managed directly by the SNMP protocol. That is, all failures of the PROXY device have occurred.
In case, SNMP protocol function of FDDI network
Detects nodes that support
Switch to direct node management. [0045][9] Automatic disconnection from SNMP to PROXY
Replacement (when PROXY is restored):In the present invention, the above network
PROXY device on ring in network management system
Periodically monitor the failure recovery, and when recovery is detected, PR
It can be managed via an OXY device. That is, the recovery of the failed PROXY device is periodically performed.
And if this is detected, the FDDI
Is not monitored by the PROXY device and the SNM
When monitoring by the P protocol, the recovery PROXY device
This is to switch to monitoring by device. In the same manner as the automatic switching of the PROXY device described above,
Then, the monitoring is switched to monitoring by the restoration PROXY device. Hereinafter, a network management system according to the present invention will be described.
An embodiment of the system will be described in detail with reference to the drawings. [0049][1] Automatic detection and selection of PROXY device
Selection: Figures 1 to 6 Figure 1 shows an FDDI network with multiple PROXY devices.
In this example, the FDDI link is shown.
Group R has four nodes 1, 3 to 5 and one concentrator
The concentrator 6 is provided with a
Further, a node 2 is connected. Also, SNMP money
JA MG is connected via LAN communication path (Ethernet)
Connected to the node 5. In such an FDDI network,
The SNMP manager MG detects the PROXY device
The following processing is performed for selection. This is the flow of Fig. 2.
Chart, transmission / reception information shown in FIG. 3, and response shown in FIG.
Referring to the MIB information and the table configuration shown in FIG.
Will be described. There are two methods for this process.
Therefore, each will be described. (1) Failure detection due to no response reception <Step S1: FIG.
Detection> First, the SNMP manager MG
Send ping (response inquiry) to all nodes on R
Polling to detect the IP address. further,
For the node of the detected IP address, the information of each node
Table that sets MIB information for collecting information
(1) (see Fig. 5) "SNMP Agent Information"
Retrieves MIB and retrieves MIB by SNMP protocol
Send a get (collection) request for information. The MIB information is returned in response to this request.
Node with FDDI on the support media in the node
(SNMP node) is managed by the SNMP manager MG.
Elephant node. As a result, the response result of the MIB information
1 (see FIG. 3), the SNMP node
Node 1, Node 2, and Node excluding Node 4, which is a node
3. Node 5 is detected. The managed SNMP nodes 1 to 3,
5 is the IP address required for the SNMP protocol, S
Text that sets the node address of the NMP agent
Cable (2). <Same Step S2: Detection of PROXY Device
> Whether or not the node has the PROXY function
In order to determine the
Y function "Takes out MIB information and transmits it. Then, the response result 2 of the MIB information
Is the node where the PROXY device is a non-PROXY device
Node 5, node 2 and node 3 excluding node 5
Will be issued. PROXY nodes 1 to 3 that have been managed
Is the IP address of the PROXY device required for the SNMP protocol
Set the address in table (3). <Step S3: Selection of PROXY Device
> SNMP manager MG is SM on FDDI ring R
One PROXY device as a medium for managing T nodes
Select. SNMP manager MG is a PROXY device
Are the following two conditions. Condition 1: Connection ring is small (that is, PROXY device
To reduce the load on PROXY device nodes
Therefore, use a PROXY device with few FDDI rings.
Ahead). Condition 2: Directly connected to the ring (immediately
The PROXY device directly connected to the FDDI ring
PROXY device connected to the device and the concentrator
Is an FDDI ring that is not susceptible to failure of other nodes
Priority is given to the PROXY device directly connected to. Con
In the case of PROXY device connected to centrifuge
Cannot be managed if the concentrator node fails.
Will be). These conditions are applied to each PROXY device.
The following MIB information is shown in the table
Take it out from (1) and send it out. The response result is
Response results 3 and 4. MIB information: "Connection FDDI ring information (number of connections)" MIB information: "Node information (number and type of accommodation ports
And the response result 3 indicates that the node 3
Is also connected to the FDDI ring R1,
Not applicable. Also, from response result 4, node 2
Connected beyond the other node (concentrator 6)
Therefore, it does not correspond to the condition 2. From the above results, the SNMP manager M
G uses node 1 as a medium for managing the FDDI ring R.
Select as PROXY device. <Step S4: Management PROXY Node>
Confirmation> Next, after selecting the PROXY device,
(3) is sorted in order of priority and managed as shown in FIG.
Change the PROXY device flag. Then, the SNMP manager MG
Nodes managed by OXY node 1 and destination MAC addresses
Information that can determine correspondence between "instance and destination"
MAC address correspondence ”from table (1)
To send. According to the response result 5 of the MIB information,
The MAC address of the node managed by the PROXY device and the PR
Collecting the node identification instance of the OXY device
Can be. In FIG. 3, the response result 1 and the response
As can be seen by comparing the result 5 of the SNMP management,
MG is the information of the response result 5 (node 1
From node 4 for SNMP money
The node can be recognized as a node under the management of the ja. collection
Management information of each node (for each FDDI ring)
The MAC address and instance are set in Table (4).
Set. Hereinafter, the information of tables (3) and (4)
Specify the instance corresponding to node 4 from the PROX
By sending the SNMP protocol to the Y device
Thus, the MIB information of the node 4 can be obtained. As described above, the SNMP manager MG
It can automatically recognize the PROXY device and
The SMT node can be recognized via the device. Thereafter, the SNMP manager MG moves to FIG.
As shown, MIMO for the PROXY device (node 1)
B information GET is performed regularly, and the FDDI ring R
Network (all nodes on the ring R). [0069][2] Automatic switching of PROXY device (failure
(At birth): FIGS. 7 to 11 In the above embodiment [1], the SNMP manager MG
PROXY device as medium for managing DDI ring R
(Node 1) is automatically selected, and
All PROXY devices housed in the group R
(3) is set. After setting the management PROXY device,
The NMP manager MG passes through the management PROXY device,
Periodically from node 1 accommodated in FDDI ring R to M
Collects and manages IB information. In the embodiment [2], the management medium is selected.
Management of a failed PROXY device, and when a failure is detected
Select the management medium again.
Replacement processing. Hereinafter, this process is shown in FIG.
And the table configuration shown in FIG.
This will be described with reference to the time chart shown in FIG. <Step S5 in FIG. 7: MIB Information Receiving Unit
IM start> SNMP manager MG is a management PROXY device
Request for MIB information to node 1
When transmitting, timer value of data (1) (for example, 3 seconds)
To start the timer (1) and the MI of the table (5).
The number of retries of B information is initialized to “5” (see FIG. 8).
See). <Step S6: Retry of MIB Information
> Here, as shown by * 1 in FIG.
Response before the timeout of the revocation timer (1) occurs
When receiving, the PROXY device is normal (no failure)
Release of response non-reception determination timer (1)
And at the same time, set the number of retries in table (5) to “0”.
To clear. On the other hand, before receiving the response,
When the timeout of the non-reception determination timer (1) occurs
The retry count in the retry count setting table (5)
The number of executions is updated (+1), and the maximum number of data (2)
Compare with the number of tries. As shown by * 2 in FIG.
If the response can be collected once, the maximum retry
Since the number has not been reached, the PROXY node itself
Request to get MIB information and data
The timer (1) is started at the timing (1). Further, as indicated by * 3 in FIG.
Continue retrying when no data is received, and retry
If the number of times exceeds the maximum number of retries, the management PROX
The device Y (node 1) sees a failure state as shown in FIG.
Assuming, SNMP manager MG is the next PROXY node
Is performed. <Step S7: Management PROXY Node>
Switch> Management that SNMP manager MG recognized as a failure
The PROXY device (node 1) manages the table (3)
Change the PROXY flag to failure (= 2). This
Management PROXY device can be recognized as a failure. [0077] The management PROXY device on the FDDI ring
Table of processing results of embodiment [1]
PROXY device flag set in (3) is not managed
Set to PROXY (= 0) and higher in table (3)
Selected node 3. The modified PROXY device uses the node identification ID.
The “PROXY tube” in step S4
Process by the changed PROXY device
Address of the node to be connected and the node of the PROXY device
Collect the instances for identification and PR in table (4)
Change the node identification information of the OXY device. With the above processing, the PROXY device is changed.
And to the managed node via the changed PROXY device
Thus, a get request for MIB information can be made. (2) Failure detection by TRAP reception <Step S8: Failure TRAP reception> SNMP money
Ja MG has P as shown in FIG. 11 (and FIG. 40).
Asynchronous information notification transmitted from the ROXY device (TRA
P) can be received. The SNMP manager MG operates the PROXY device.
TRAP of "PROXY function disabled notification" from the device
In this case, the SNMP manager MG proceeds to step S7.
The “PROXY automatic switching process” is performed. [0082][3] Automatic switching of PROXY device (for load
12) to 17) The obstacles mentioned above when automatically switching PROXY devices
In addition to the detection, the load is detected as shown in FIG.
When it turns out that it is a load, the management PROXY device
There are two ways to switch. In this case, the following two methods are available.
real. These processes are shown in FIGS.
17 and the table configuration diagram shown in FIG. 16 and FIG.
This will be described with reference to the time chart shown in FIG. [Method 1] <Step S9 of FIG. 13: Start of load management timer> Data
The timer (2) is started by the timer value of (3). this
Initialize the accumulated number of retry times of hour data (4) to “0”
Set. The SNMP manager MG sends the MIB information
Retry when response is not received in collection process
U. When retry is performed as shown by * 1 in FIG.
The number of harm retries is cleared if there is a response.
To "0". In parallel with this, for the load shown in FIG.
Update data (4) to store the number of retries
go. This is not cleared even when a response is received. <S10: Load confirmation> Timer
When (2) times out, retry count for load
Maximum number of retries for data (4) and load
(In this example, “5”).
You. As a result, as shown in FIG.
(4) becomes 6 times and exceeds data (5), so high
Considered as load. The management PROXY (node
Since 1) cannot be managed, the steps in the embodiment [2] are not performed.
Management in the same way as “PROXY device switching processing” in step S7
Change the PROXY device. [Method 2] <Step S10 in FIG. 15: Confirmation of Load MIB>
Is the negative of the PROXY device to the PROXY device (node 1).
Extended MIB information for monitoring load-Set MIB information
You. Further, the threshold value of the table (6) shown in FIG.
"5" is set as information. The SNMP manager MG has a table
MIB information for collecting load monitoring from (6)
As shown in FIG. 14, the MIB information is periodically
Request a get from the managing PROXY device. When the response result is "5" or more
If so, it is considered that the load is high, and PROXY has become high load.
(Node 1) executes “PRO” in step S7 of the embodiment [2].
Switching the managed PROXY device in the same way
Change. [0092][4] Automatic switching of PROXY device
Alternate): FIGS. 18 to 20 In this embodiment, the SNMP manager MG
MIB information collection of nodes accommodated on the ring is PRO
Get MIB information via XY device (node 1)
And the flowchart shown in FIG.
Automatic timed switching step S12 of the PROXY device in accordance with
Execute First, the table (7) shown in FIG.
Make the settings for -FDDI ring number = 1-Change time 1 = 07:00-PROXY device name 1 = 1 (node 1)-Change time 2 = 17:00-PROXY device name 2 = 2 (node 2)-Change time 3 = 22 : PROXY device name 3 = 3 (node 3) Also, the PROXY device timed confirmation timer is set to (= 60).
Set. When the current time is 9:00, PR for each time
PRO corresponding to the change time of the OXY device setting table (7)
The XY device name 1 is taken out. Node 1 of the currently managed PROXY device and retrieval
Compared with the PROXY device (= node 1)
Since the PROXY device name corresponding to the hour is the same, the current management PRO
Let the XY device be node 1 (see FIG. 20). This is P
Check every ROXY device timed confirmation timer (60 minutes). During the above PROXY scheduled check, the current time is 1
At 7:00, a PROXY setting table for each hour
(7) No change from PROXY device name 2 corresponding to change time
Obtain C2. Compare with the currently managed PROXY device (node 1)
However, since it is different, the "PROXY device" of the embodiment [2] is different.
Automatic switching (when a failure occurs) ”
The device is switched to node 2 (see FIG. 20). When the current time reaches 22:00, the above
Change the management PROXY device to node 3 as shown in FIG.
0). [0099][5] Automatic recognition of node connection status on ring
Knowledge: Figures 21 to 27 This example illustrates nodes and ports on the FDDI ring.
It is intended to recognize the connection state of the client. The FDDI network shown in FIG.
1 compared to the FDDI network shown in FIG.
Node 2 is used as a concentrator on
The node 2 further includes nodes A,
The difference is that B and C are connected. Hereinafter, such an FDDI network will be described.
The operation is shown in the flowchart shown in FIG.
Table configuration diagram and response result diagram and diagram shown in FIG.
Flow chart of token shown in 25 and port shown in FIG. 26
And the flow of nodes on the primary ring shown in FIG.
This will be described with reference to the drawings. In such an FDDI network,
The SNMP manager MG connects to the FDDI ring R
Recognize the positional relationship (upstream / downstream) of the node being performed
Therefore, the MIB information is obtained from the table (1) shown in FIG.
Extract "upstream node" and MIB information "downstream node"
And send. Each node has the data as shown in FIG.
And the MIB from the SNMP manager MG
The data is returned in response to the get request (step of FIG. 22).
Step S13). Thus, the responses from all nodes
The result obtained is response result 1 shown in FIG.
Token flow on the ring from response result 1 of
However, it can be seen that it is as shown in FIG. That is, based on the MAC indicated by the symbol ■
Since the flow of tokens on the ring is determined, in this example 1
That is, it is along the next ring P. Of this token
The flow is set in table (8) for each node (the same
Step S14). Further, the connection state of the port indicated by □ is known.
Table (1), the MIB information “node
Number of ports ”, MIB information“ node port type / port
Port type of the connection destination of the port
And send it. That response
Are response results 2 and 3 in FIG.
S14). Port type is synonymous with "instance"
Is a word. From the response results 2 and 3, the port connection flow
FIG. 26 illustrates this. Node 2 is
Connected to another port because it has two
Is a concentrator with two openings
You. These are set in the table (9). Pos for each node
The port number is set in the table (10). The connection partner is recognized by the
After the end of all nodes, downstream from its own MAC
(Step S15). Here, identification of the connection partner for node 2
A processing example will be described. FIG. 27 shows the flow of nodes on the primary ring P.
This is shown. (A) From own port with MAC as reference position
Upstream port (SAS: Single Attachment Station = 1
Mouth node is not included) is the order from MAC to upstream
The nodes traversed upstream by the number of are set as connection partners. However,
When there is a concentrator having two ports (FIG. 21)
Node A, etc.) is the connection port in the node.
Port number-1 ”. That is, in the port, the“ order ”from the MAC
Is the upstream node A as the connection partner.
Then, the “number of orders = 2” from the MAC is the node B.
You. However, Node B has
Data, so it is actually connected under node A.
You. In the case of a concentrator,
Tracing upstream by the number of connected ports (number of connected ports -1) "
Node is the connection partner. Therefore, the port is
The node C of the stream is a connection partner. The port is connected to the node C by the concentrator.
Is not connected because node 3 is the upstream node of node C
The other party. (B) From own port with MAC as reference position
In the downstream port, node 2 is set as the connection partner.
Node 2 and node 2 is the connection partner
If there is no setting as, the connection partner. That is, the port is already connected to the node 1 by itself.
Node (node 2) is set as the connection partner,
Mode 1 is not set as the connection partner on the upstream side of the MAC.
No. Therefore, the node 1 is set as a connection partner. (C) Connection partners of all nodes as described above
An identification process is performed, and based on the results of FIGS.
The NMP manager MG manages the network as shown in FIG.
The connection mode can be recognized. [0116][6] Automatic node addition: FIGS. 28 to 30 In this embodiment, the SNMP manager MG
To manage the SMT node on the FDDI ring
Automatically manages newly added nodes on the ring when
The flowchart shown in FIG. 28 and the flowchart shown in FIG.
FDDI ring configuration diagram and node addition shown in FIG. 30
This is explained below with reference to the diagram showing the response results
You. <Step S16: Monitoring of Node> Node
Fig. 29 (1) showing the FDDI network before addition
And the SNMP manager MG is on the FDDI ring R
SMT nodes 1 to 3 (where nodes 1 and 2 are SNM
It also has the function of P node, and node 2 is managed PROXY
Device) as monitoring information.
MIB information of "code recognition" and "downstream node recognition"
Is sending to. The token flow at this time is indicated by an arrow
This is as indicated by the mark. The response from each SMT node to this
The result is as shown in FIG. 30 (1).
Are SMT nodes 1 to 3. Then, as shown in FIG.
4 is added, the response of the MIB information at this point is
The result is as shown in FIG. <Step S17: Monitoring of Adjacent Nodes>
Comparing the response results of 30 (1) and 30 (2),
The downstream node of node 2 and the upstream node of node 3 change
You can see that it is doing. From this result, between node 2 and node 3
It can be recognized that the node 4 has been added. On the basis of the recognition result, the SNMP management
MG has node knowledge for node 2 (PROXY device).
Request another instance. The instance of the identifier from node 2 to node 4
After obtaining the license, add it to the management target in Table (4).
You. [0124][7] Automatic ring addition: FIGS. 31 to 33 In this embodiment, the SNMP manager MG
Connected to a node on the FDDI ring via a device
When managing a ring, when a new ring is added
The management shown in FIG.
And the FDDI ring configuration shown in FIG. 32 and FIG.
Referring to the diagram showing the response result due to node addition,
This is described below. The functions and tokens of nodes 1 to 3
Is the same as that in FIG. 29 (1). <Step S18: Monitor PROXY Node
View> Show network before adding FDDI ring
In FIG. 32A, the SNMP manager MG
The PROXY device on the DDI ring R is monitored. This
FDDI ring information (connection
Number)] MIB information is transmitted periodically. The response from each SMT node is
The result is as shown in FIG.
(PROXY device) connection FDDI ring
Ing R only. Then, as shown in FIG.
2 and the FDDI ring R1 is added,
The response result of the MIB information is as shown in FIG.
become. <Step S19: Connection FDDI Ring
Monitor> This allows the SNMP manager MG to
Recognizes that the ring number information is different from the current managed FDDI ring number
Then, it is determined that there is a newly added FDDI ring, and the node 2
(PROXY device) New FDDI ring information
(Instance). Get information from node 2
After that, the FDDI ring R1 is set in the table (4) and managed.
Add to subject. [0129][8] Automatic switching from PROXY to SNMP
Replacement (when all PROXY failures occur): FIGS. 34 and 35 The SNMP manager sets the table as shown in FIG.
(2) The current SNMP interface that manages the FDDI ring
Agent function support node information is set as follows
Keep it. SNMP agent IP address = 133.161.59.221 SNMP agent MAC address = 00: 00: 0E: 00: 01 FDDI ring number = 1 The management PROXY device from the above embodiment [2].
When the device (node 1) fails, another PROXY device
After switching to (Node 3), the same applies to Node 3
Failure, and even on the same FDDI ring
It is assumed that a failure has occurred in all PROXY devices. <Step S20: SNMP Agent
Node recognition> Then, the SNMP manager
All PROXY devices that get information are faulty
First, the SNMP agent function support node is recognized.
Understand. First, according to the table (2), the “SNMP
Agent IP Address ”and“ SNMP Agent M ”
SNMP from node 5 (see FIG. 21)
Judge as an agent function support node. Then, the SNMP manager MG
Whether or not C5 supports SNMP-FDDI
For confirmation, the FDDI recommendation in Table (1) in FIG.
FDDI of "RFC1512 / RFC1285"
The related MIB information is transmitted directly to the node 5. When the response from the node 5 is received,
Node whose mode 5 supports SNMP-FDDI
Recognize that Hereinafter, the IP address of the node 5 is stored in the table (4).
Address and collect MIB information from node 5 directly.
Switch to get MIB information to contact node 5. Therefore, all PROXY devices were in trouble.
In this case, the node supporting SNMP-FDDI is
Switch to get MIB information directly to other nodes. [0138][9] Automatic disconnection from SNMP to PROXY
Replacement (when PROXY is restored): FIGS. 36 and 37 According to the above embodiment [8], the FDDI ring is
MIB information directly to P-FDDI support node 5
Get is done. <Step S21: PROXY recovery monitoring
> In parallel with this, the SNMP manager
PROX from table (1) shown in FIG.
The Y function related MIB information is taken out and the timer (1)
Depending on the period, get MIB information is periodically performed. While continuing the above processing, node 1
B information is returned. As a result, the PROXY device (no
It is considered that c) has been restored. <Step S22: PROXY recovery processing
> Here, the SNMP manager monitors the FDDI ring.
Check the viewing state and check the current SNMP-FDDI support node
5 is a direct MIB information get request to
Switch to Y device (node 1). Hereinafter, nodes accommodated in the FDDI ring
Of MIB information from the node 1 via the PROXY device of node 1
Performed by reason. As described above, the necessity according to the present invention is as follows.
According to the network management system, the SNMP manager
Polled each node and returned
SNMP node is detected from the network management information,
SNMP nodes convert SNMP-SMT protocol
Device (PROXY device) or not, and
Predetermined priority when there are multiple
So that only one device is selected according to
The P manager is a function of each node in the FDDI network.
It is not necessary to keep track of
Work management can be performed. Further, the SNMP manager sets the PROX
Y device, SNMP-FDDI support node, FDDI
The connection node and the number of FDDI rings can be automatically recognized.
Can easily respond to changes in network topology.
Automatic switching of PROXY device, PROX
Automatic switching from Y device management to SNMP-FDDI management,
Automatic switching from SNMP-FDDI management to PROXY device management
Switching to respond flexibly to changes in network conditions.
And stable management. The connection mode of the FDDI connection node is
FDDI network connection type
State can be expressed visually,
Provide easy-to-use management.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an FDDI network used for an embodiment [1] (automatic detection / selection of a PROXY device) of a network management system according to the present invention.

FIG. 2 is a flowchart showing a processing procedure of an embodiment [1] of the network management system according to the present invention.

FIG. 3 is a table showing transmission / reception information in the embodiment [1] of the network management system according to the present invention.

FIG. 4 is a table showing response information (MIB) to a management information collection request (get) in the embodiment [1] of the network management system according to the present invention.

FIG. 5 is a configuration diagram of a table used in the embodiment [1] of the network management system according to the present invention.

FIG. 6 is a block diagram showing an interrelationship between an SNMP manager and a PROXY device in the network management system according to the present invention.

FIG. 7 is a flowchart illustrating a processing procedure of an embodiment [2] of the network management system according to the present invention (automatic switching of a PROXY device (when a failure occurs)).

FIG. 8 is a configuration diagram of a table used in the embodiment [2] of the network management system according to the present invention.

FIG. 9 is a time chart for explaining the operation of the embodiment [2] of the network management system according to the present invention.

FIG. 10 is a block diagram showing a failure image of an SNMP manager and an agent in the embodiment [2] of the network management system according to the present invention.

FIG. 11 is a block diagram showing failure detection by TRAP reception in the embodiment [2] of the network management system according to the present invention.

FIG. 12 is an embodiment [3] of the network management system according to the present invention (automatic switching of PROXY devices (load distribution));
FIG. 4 is a flowchart schematically showing the processing procedure of FIG.

FIG. 13 is a flowchart illustrating a load detection method 1 in the embodiment [3] of the network management system according to the present invention.

FIG. 14 is a flowchart schematically showing management information collection processing of load detection method 2 in the embodiment [3] of the network management system according to the present invention.

FIG. 15 is a flowchart illustrating a load detection method 2 in the embodiment [3] of the network management system according to the present invention.

FIG. 16 is a configuration diagram of tables / data used in the embodiment [3] of the network management system according to the present invention.

FIG. 17 is a time chart for explaining the operation of the embodiment [3] of the network management system according to the present invention.

FIG. 18 shows an embodiment [4] of the network management system according to the present invention (automatic switching of PROXY devices (periodic switching)).
FIG. 4 is a flowchart schematically showing the processing procedure of FIG.

FIG. 19 is a configuration diagram of tables / data used in the embodiment [4] of the network management system according to the present invention.

FIG. 20 is a time chart for explaining the operation of the embodiment [4] of the network management system according to the present invention.

FIG. 21 is a configuration diagram of an FDDI network used for an embodiment [5] (automatic recognition of a node connection state on a ring) of a network management system according to the present invention.

FIG. 22 is a flowchart showing a processing procedure of the embodiment [5] of the network management system according to the present invention.

FIG. 23 is a configuration diagram of a table used in the embodiment [5] of the network management system according to the present invention.

FIG. 24 is a table showing response results at the time of management information collection in the embodiment [5] of the network management system according to the present invention.

FIG. 25 is a diagram showing a flow of tokens detected in the embodiment [5] of the network management system according to the present invention.

FIG. 26 is a diagram showing a flow of ports detected in the embodiment [5] of the network management system according to the present invention.

FIG. 27 is a diagram showing a flow of nodes on the primary ring detected in the embodiment [5] of the network management system according to the present invention.

FIG. 28 is a flowchart illustrating a processing procedure of an embodiment [6] (automatic node addition (management target addition)) of the network management system according to the present invention.

FIG. 29 is a block diagram for explaining the operation of the embodiment [6] of the network management system according to the present invention.

FIG. 30 is a table showing response results at the time of collecting management information in the embodiment [6] of the network management system according to the present invention.

FIG. 31 is a flowchart illustrating a processing procedure of an embodiment [7] of the network management system according to the present invention (automatic ring addition (management target addition)).

FIG. 32 is a block diagram for explaining the operation of the embodiment [7] of the network management system according to the present invention.

FIG. 33 is a table showing response results at the time of management information collection in the embodiment [7] of the network management system according to the present invention.

FIG. 34 shows an embodiment [8] of the network management system according to the present invention (automatic switching from PROXY to SNMP (P
It is a flowchart figure which shows the processing procedure of (at the time of ROXY all failure).

FIG. 35 is a configuration diagram of a table used in the embodiment [8] of the network management system according to the present invention.

FIG. 36 shows an embodiment of a network management system according to the present invention.

[9] (Automatic switching from SNMP to PROXY (P
FIG. 9 is a flowchart illustrating a processing procedure of (ROXY failure recovery)).

FIG. 37 shows an embodiment of a network management system according to the present invention.

It is a block diagram of the table used by [9].

FIG. 38 is a block diagram showing an SNMP management system using a conventionally known SNMP-SMT conversion PROXY device.

FIG. 39 is a configuration diagram of a conventionally known SNMP frame.

FIG. 40 is a diagram showing SNMP-SMT conversion by a conventionally known PROXY device.

FIG. 41 is a configuration diagram of a conventionally known SMT frame.

[Explanation of symbols]

 M SNMP manager 1-5 Node 6 Concentrator R, R1 FDDI ring P Primary ring S Secondary ring ■ MAC □ Port In the figure, the same reference numerals indicate the same or corresponding parts.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-7-334445 (JP, A) JP-A-7-226777 (JP, A) JP-A-6-309257 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) H04L 12/42

Claims (11)

(57) [Claims] 1. SNMP protocol or SNMP-SM
In a network management system in which an SNMP manager manages a plurality of nodes on an FDDI ring via a protocol conversion device having a T conversion protocol function, the SNMP manager polls each node and returns the first. SNMP from node management information
Nodes are detected, and whether or not these SNMP nodes are the protocol converters is determined from the second node management information. When a plurality of the protocol converters exist, the SMT nodes on the FDDI ring are managed. A network management system, wherein only one protocol conversion device as a medium is selected in accordance with a predetermined priority based on third node management information. 2. The network management system according to claim 1, wherein the SNMP manager performs the management when the SNMP agent information is not returned even if the SNMP agent information is continuously retried a predetermined number of times at a predetermined time interval during the polling. A network management system characterized in that a protocol conversion device is determined to be a failure and automatically switched to another protocol conversion device according to the priority. 3. The network management system according to claim 1, wherein the SNMP manager receives the asynchronous information notification from the management protocol conversion device and determines that the protocol conversion device is faulty. A network management system characterized by automatically switching to another protocol converter according to the order. 4. The network management system according to claim 1, wherein the SNMP manager is configured to execute the SNMP agent information when the cumulative value of the number of times of retry at predetermined time intervals exceeds the threshold value during the polling. A network management system characterized in that the management protocol converter is determined to be in a high load state and automatically switched to another protocol converter according to the priority. 5. The network management system according to claim 1, wherein said SNMP manager places said management protocol converter in a high load state based on load monitoring information from said management protocol converter. A network management system for automatically switching to another protocol converter according to the priority. 6. The network management system according to claim 1, wherein the SNMP manager automatically switches the currently managed PROXY device to another PROXY device at a preset time. And a network management system. 7. The network management system according to claim 1, wherein the SNMP manager comprises an upstream node and a downstream node set in advance in each node.
A network management system characterized by automatically recognizing a node connection state on the ring by collecting and analyzing information on the number of ports, port types, and port types to which the ports are connected. 8. The network management system according to claim 1, wherein said SNMP manager recognizes a newly added node on said ring by managing upstream and downstream nodes of each node. A network management system characterized by automatically adding to a management target. 9. The network management system according to claim 1, wherein said SNMP manager recognizes a ring newly added to a node by managing the number of FDDI rings connected to each node. A network management system characterized by automatically adding the ring to the management target. 10. The network management system according to claim 1, wherein a protocol conversion device to be switched does not exist on the FDDI ring.
A network management system characterized in that it is directly managed by a SNMP protocol with a node having an SNMP agent function on the ring. 11. The network management system according to claim 1, wherein the PROX on the ring is
A network management system characterized in that a failure recovery of a Y device is periodically monitored, and when recovery is detected, management is performed again via a PROXY device.