JP2007122439A - Communication apparatus and communication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To report the occurrence of a fault to an SNMP manager without allowing the network interface of a communication apparatus monitored by SNMP to be redundant. <P>SOLUTION: Fault monitoring programs 221, 321 detect communication faults in communication boards 20, 30. When the communication fault is detected in the self communication boards 20(30), SNMP agents 222, 322 determine whether an address imparted to the communication board 20(30) is usable or not, detect the connection state of the network and transmit an SNMP trap for reporting the occurrence of the fault to the SNMP manager 100 when the address is usable, change the address into the other one when the address is unusable and transmit a SNMP trap transmission request to the other SNMP agents 322, 222. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a communication device and a communication method for notifying a network management device of a detected failure.

  In a TCP / IP (Transmission Control Protocol / Internet Protocol) network, SNMP (Simple Network Management Protocol) is a protocol for monitoring and controlling communication devices connected to the network such as routers, PCs, and terminals via the network. is there.

  The SNMP is divided into SNMP manager and SNMP agent elements. The SNMP manager is a network management device in which management software corresponding to SNMP is installed, and the SNMP agent is a communication device to be managed. Say.

  The communication device includes a management information database called MIB (Management Information Base). The MIB stores network information collected by the SNMP agent. Specifically, it stores network information such as its own IP address, the number and size of received packets, the network usage rate, and the number of collisions. (For example, see Patent Document 1).

Network management by SNMP is performed when the SNMP manager refers to or changes the MIB information of the SNMP agent. Further, when an interface of the communication device is down (link down) due to some failure, or when the failure is recovered and the interface of the communication device is up (link up), the SNMP agent transmits an SNMP trap (packet). The SNMP manager can recognize and manage the event occurrence.
Japanese Patent Application Laid-Open No. 2004-280592

  By the way, in the prior art, when a failure occurs in a communication apparatus or the like having two communication boards shown in FIG. 8, each SNMP agent is started asynchronously and independently and sends an SNMP trap to the SNMP manager. Become.

  However, if the failure that has occurred is a network down due to a network cable disconnection in the route of one SNMP agent program or a failure of the network interface, the SNMP agent program cannot secure a route, so an SNMP trap is sent to the SNMP manager. There was a problem of not being able to notify. In order to cope with this problem, it has been necessary to take measures such as making the network interface of the communication device to be SNMP monitored redundant.

  The present invention has been made based on the above-described circumstances, and provides a communication device and a method thereof capable of notifying an SNMP manager of the occurrence of a failure without making a network interface of a communication device to be SNMP monitored redundant. The purpose is to provide.

  In order to solve the above-described problems, the present invention is a communication device that includes a plurality of communication units that are connected to a network management device via a network and each of which is assigned an address in advance, and any one address is A storage unit for storing information indicating that the communication unit is usable; a failure monitoring unit for detecting a failure in the communication unit; and when the failure monitoring unit detects a failure, an address assigned to the communication unit is A determination unit that determines whether or not the communication unit can be used based on the information; and if the address given to the communication unit is usable as a result of the determination, the network management device is notified of the occurrence of the failure. If the address given to the communication means is unusable as a result of the determination, other addresses can use the information stored in the storage means. Rewriting the information indicating that there is a communication apparatus characterized by comprising a, a failure notification means for notifying the occurrence of the fault to the network management device on the basis of the other addresses.

  According to the present invention, in the above-described invention, the failure notification unit detects a connection state of the network when the address given to the communication unit is usable as a result of the determination, and Is connected to the network management device, and when the network is disconnected, the address stored in the storage means is changed to another address. It is changed, and the occurrence of the failure is notified to the network management device based on the other address.

  According to the present invention, in the invention described above, the failure notification unit refers to the storage unit when the communication device is activated, and indicates an address to which information stored in the storage unit is given to the communication unit In the case other than information, the address given to the communication means is recorded in the storage means.

  The present invention is a communication method in a communication apparatus that is connected to a network management apparatus via a network and includes a plurality of communication means to which addresses are assigned in advance, and any one address can be used. On the basis of the information indicating whether or not an address assigned to the communication means is usable when a failure is detected; and a step of detecting a failure of the communication means; Determining if the address assigned to the communication means is usable as a result of the determination, notifying the occurrence of the failure to the network management device, and determining that the communication If the address given to the means is unusable, the information stored in the storage means can be used by another address Rewriting the information indicating a communication method characterized by comprising the step of notifying the occurrence of the fault to the network management device on the basis of the other addresses, the.

  According to the present invention, the communication device includes a storage unit that stores information indicating that any one address is usable, a failure monitoring unit that detects a failure of the communication unit, and the failure monitoring unit detects a failure. In this case, the determination means for determining whether the address assigned to the communication means can be used based on the information, and if the address assigned to the communication means as a result of the determination is usable, the network Notify the management device of the occurrence of a failure. If the address assigned to the communication means is unusable as a result of the determination, the information stored in the storage means indicates that another address can be used. It is configured to include failure notification means that rewrites information and notifies the network management device of the occurrence of a failure based on the other address. As a result, path control when a failure occurs can be executed internally, and the SNMP manager can be notified of the failure occurrence without making the network interface of the communication device to be monitored SNMP redundant. Can be improved. Further, it can be realized at a lower cost than the method of making the network interface of the communication device to be SNMP monitored redundant.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing an internal configuration of a communication apparatus according to the present embodiment. The communication device 5 of the present embodiment includes two communication boards 20 (30). The communication board 20 (30) includes a CPU 22 (32), a memory 23 (33), a disk 24 (34), and a PCI ( Peripheral Component Interconnect) / memory controller 25 (35) and network interface (NW IF) 21 (31) are mounted. Specifically, a board PC or the like is applied as the communication board 20 (30).

  Each of the CPUs 22 (32) has an SNMP agent program and a failure monitoring program to be described later. Of these, the failure monitoring program notifies the SNMP agent program of the occurrence of a failure when it detects the occurrence of the failure. It is a mechanism to do.

Each of the communication device boards 20 (30) is connected to the PCI bus 40 via the PCI / memory bus controller 25 (35) on the one hand, and the NWI / F 21 (31) and the router 2 (3) on the other hand. ) To the IP network 4.
In addition, a network management apparatus 1 in which an SNMP manager is installed is connected to the IP network 4.

2 to 4 are diagrams showing schematic steps of the communication method of the present embodiment.
As shown in FIG. 2, one SNMP agent program 222 (322) resides in the CPU 22 (32), and an attribute address that can be uniquely recognized is assigned to each SNMP agent program 222 (322). Specifically, the slot number of the slot in which the communication board 20 (30) of the communication device 5 is incorporated can be used as the attribute address.
The determination whether the attribute address of the communication board 20 (30) is the main attribute address is stored in the item of the main attribute address stored in the shared DB 240 (340) included in the communication board 20 (30). This is performed depending on whether or not the attribute address matches the attribute address of the own board. FIG. 2 shows an example in which the attribute address of the SNMP agent program 222 resident in the communication board 20 is a primary attribute address, and the attribute address of the SNMP agent program 322 resident in the communication board 30 is a secondary attribute address. .

  The shared DB 240 (340) is provided in each of the disks 24 (34) shown in FIG. 1, and stores information such as the above-described MIB in addition to the attribute address described above.

  In addition, the shared DB 240 (340) can be locked when the SNMP agent program 222 (322) updates or references data in order to maintain data consistency with respect to the asynchronously operating communication board 20 (30). It is configured as follows. Specifically, when the SNMP agent program 222 locks the shared DB 240, the information is notified to the SNMP agent program 322, and the SNMP agent program 322 accesses the shared DB 340 until a lock release notification is received from the SNMP agent program 222. It is configured so that it cannot be done. In addition, the shared DB 240 (340) can be locked when the SNMP agent program 222 (322) updates or references data in order to maintain data consistency with respect to the asynchronously operating communication board 20 (30). In this case, the lock and unlock process may be performed as follows, for example. That is, when the SNMP agent program 222 locks the shared DB 240, the shared DB 240 notifies the shared DB 340 to lock, and the shared DB 340 that has received the notification locks based on the notification. When the data update is completed, the SNMP agent program 222 notifies the shared DB 240 of unlocking, and the shared DB 240 that has received the notification notifies the shared DB 340 of unlocking.

  Further, in order to maintain the consistency of the shared DB 240 (340), when information is recorded in one of the shared DBs 240 (340) by the SNMP agent program 222 (322), the SNMP agent program 222 (in which information is recorded) 322) notifies the information recorded in the other SNMP agent program 322 (222), and the notified other SNMP agent program 322 (222) notifies the information notified to the shared DB 340 (240) after releasing the lock. It is configured to perform a recording process.

  In the state of FIG. 2, the failure monitoring program 221 of the communication board 20 means that a failure has occurred in the communication board 20 (for example, the CPU 22 does not detect the operation of the memory 23, the disk 24, and the PCI / memory controller 25). ) Is detected. As described above, since the attribute address of the SNMP agent program 222 of the communication board 20 is the main attribute address, an SNMP trap signal is transmitted from the SNMP agent program 222 to the SNMP manager 100 of the network management device 1.

  On the other hand, as shown in FIG. 3, when the failure monitoring program 321 of the communication board 30 detects the occurrence of a failure, the attribute address of the SNMP agent program 322 of the communication board 30 is a sub-attribute address, so the SNMP agent program 322 Performs inter-process communication with the SNMP agent program 222 of the communication board 20 of the main attribute address, and transmits an SNMP trap transmission request signal to the SNMP agent program 222. Then, the SNMP agent program 222 having the address of the main attribute that has received the SNMP trap transmission request transmits an SNMP trap signal to the SNMP manager 100 of the network management device 1.

  In addition, as shown in FIG. 4, when the SNMP agent program 222 of the address of the main attribute determines the connection state of the network interface and the link is down, communication cannot be performed using its own route. The SNMP agent transmission request signal is transmitted by performing inter-process communication with the SNMP agent program 322 having the attribute address, and the SNMP agent 322 having the sub-attribute address transmits the SNMP trap signal to the SNMP manager 100 of the network management device 1.

  As a result, even if the network goes down or the network interface fails, a communication path can be secured internally between the two SNMP agent programs 222 and 322, and the network interface of the communication device 5 to be monitored by SNMP can be changed. The occurrence of a failure can be notified to the network management device 1 (SNMP manager 100) without redundancy.

5 to 7 are flowcharts cited for explaining the operation of the communication apparatus of the present embodiment.
FIG. 5 shows the operation of the communication apparatus when the power is turned on, and FIGS. 6 and 7 show the operation of the communication apparatus 5 when the SNMP trap signal is transmitted.
Hereinafter, the operation of the communication apparatus of the present embodiment will be described in detail with reference to the flowcharts of FIGS. In the following description, it is assumed that the slot number (1, 2,... N) of the communication device 5 is applied to the attribute address.

As shown in FIG. 5, when the communication device 5 is powered on, each communication board 20 (30) loads and starts the SNMP agent program 222 (322) installed in the CPU 22 (32) (step). S61, S62).
The SNMP agent program 222 (322) that is started earlier among the started SNMP agent programs 222 (322) locks the shared DB 240 (340) (step S63), and is stored in the shared DB 240 (340). Referring to the attribute address, it is determined whether or not the self-communication board 20 (30) is the main attribute based on whether or not the main attribute address matches the attribute address of the self-communication board 20 (30) (step S64). When the main attribute address is “NULL”, the own communication board 20 (30) is assumed to be the main attribute address, and the attribute address of the own communication board 20 (30) is recorded in the main attribute address of the shared DB 240 (340). (Step S65). If the main attribute address matches the attribute address of the own board 20 (30), the shared DB 240 (340) is unlocked (step S67). When the main attribute address is the other communication board 30 (20), it is determined that the own communication board 20 (30) is the sub attribute (step S66), and the shared DB 240 (340) is unlocked (step S67).

  In the process shown in FIG. 5, the operations of both SNMP agent programs 222 (322) are described. However, as long as one of the shared DB 240 and the shared DB 340 is always locked, the shared DB 240 (340) The SNMP agent programs 222 and 322 do not become the main attribute at the same time because the other process is performed after one of the processes is performed.

  Next, as shown in FIG. 6, the SNMP agent program 322 (222) receives the information of failure detection from the failure monitoring program 221 (321), or the SNMP agent of the other communication board 30 (20). When the trap transmission request signal is received from the program 322 (222) (step S71), the SNMP agent program 222 (322) refers to the shared DB 240 (340), and the attribute address of the own communication board 20 (30) is the main attribute. It is determined whether it is an address (step S72). If it is determined that the address is the main attribute address, the link state of the network interface is further determined (step S73). Here, in the case of link-up, the SNMP agent program 222 (322) transmits an SNMP trap signal to the SNMP manager 100 of the network management device 1 (step S74). On the other hand, in the case of ring down, the SNMP agent program 222 (322) locks the shared DB 240 (340) to change the main attribute address (step S75), and updates the main attribute address of the shared DB 240 (340) (step S75). S76).

  Here, the link down means a state in which the router 2 (3) is down due to some failure, and the link up means a state in which the interface of the router 2 (3) is up after recovery from the failure.

  The update of the main attribute address is executed by incrementing the attribute address stored in the item of the main attribute address of the shared DB 240 (340), and the SNMP agent of the communication board of the next slot number is incremented by the increment. The program is the main attribute. When the value of the attribute address stored in the item of the main attribute address exceeds the number of slots of the communication board 20 (30) mounted, the main attribute address is reset to “1”.

  Finally, the SNMP agent program 222 (322) unlocks the shared DB 240 (340) (step S77), and transmits a trap transmission request signal to the SNMP agent program 222 (322) having the main attribute address (FIG. 8). Step S81). At this time, in order to prevent an infinite loop of the trap transmission request signal, a request from the same address is blocked by giving a transmission source address.

  On the other hand, after the process of step S77 or step S72, if the own communication board 20 (30) has a sub-attribute address, the communication board 20 (30) sends the SNMP agent program 222 (322) with the address of the main attribute. In response, an SNMP trap transmission request signal is transmitted (step S81).

  Subsequently, the SNMP agent program 222 (322) having the address of the main attribute transmits the SNMP trap signal to the SNMP manager 100 resident in the network management apparatus 1, and the communication board 20 (30 which transmitted the SNMP trap transmission request signal). ) Receives the response (transmission failure) (step S82: No), locks its own shared DB 240 (340) (step S83), and updates the main attribute address of the shared DB 240 (340) (step S84). ). Again, when the main attribute address exceeds the number (maximum value) of the communication boards 20 (30) mounted on the communication device 5, the main attribute address is reset to “1”.

  Finally, the shared DB 240 (340) is unlocked (step S85), and an SNMP trap transmission request is transmitted to the SNMP agent program 222 (322) of the main attribute address (step S81).

  As described above, the present embodiment internally notifies the SNMP manager of the occurrence of a failure without making the network interface of the communication device subject to SNMP monitoring redundant by internally executing the path control when the failure occurs. As a result, the SNMP manager reaches a higher reach of the SNMP trap, thereby improving the reliability. Also, by using the network interface of the other communication board of the monitored communication device, the more communication boards that are installed, the higher the cost due to the redundant installation of the network interface can be prevented. I can do it.

  In the above embodiment, when the address of the own communication board and the address stored in the shared DB 240 (340) match, it is determined as the main attribute, and when it does not match, it is determined as the sub attribute. The present invention is not limited to this embodiment, and all the address items are provided in the shared DB 240 (340), information indicating whether or not each address can be used is stored, and determination is performed with reference to the information. It may be.

  The storage unit of the present invention corresponds to the shared DB 240 (340), the failure monitoring unit corresponds to the failure monitoring program 221 (321), and the communication unit, the determination unit, and the failure notification unit include the SNMP agent program 222 ( 322).

It is a block diagram which shows the structure of the communication apparatus of this embodiment. It is the figure quoted in order to demonstrate the schematic process of the communication method of this embodiment. It is the figure quoted in order to demonstrate the schematic process of the communication method of this embodiment. It is the figure quoted in order to demonstrate the schematic process of the communication method of this embodiment. It is the flowchart quoted in order to demonstrate operation | movement of the communication apparatus of this embodiment. It is the flowchart quoted in order to demonstrate operation | movement of the communication apparatus of this embodiment. It is the flowchart quoted in order to demonstrate operation | movement of the communication apparatus of this embodiment. It is a figure which shows the schematic process of the communication method in a prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Network management apparatus (SMNP manager), 2 ... Router, 3 ... Router, 4 ... IP network, 5 ... Communication apparatus, 20 ... Communication board, 30 ... Communication board, 22, 32 ... CPU, 23, 33 ... Memory, 24, 34 ... disk, 25, 35 ... PCI / memory controller, 21, 31 ... network interface (NW IF), 100 ... SMNP manager, 221, 321 ... fault monitoring program, 222, 322 ... SNMP agent program, 240, 340 ... Shared DB

Claims (4)

A communication device comprising a plurality of communication means connected to a network management device via a network, each of which is assigned an address in advance,
Storage means for storing information indicating that any one address is usable;
Fault monitoring means for detecting a fault in the communication means;
When the failure monitoring unit detects a failure, a determination unit that determines whether the address assigned to the communication unit is usable based on the information;
If the address assigned to the communication means is usable as a result of the determination, the network management apparatus is notified of the occurrence of the failure, and the address assigned to the communication means is used as a result of the determination. If it is not possible, the information stored in the storage means is rewritten with information indicating that another address can be used, and the occurrence of the failure is notified to the network management device based on the other address. Fault notification means to
A communication apparatus comprising: The failure notification means includes
As a result of the determination, when the address assigned to the communication means is usable, the connection state of the network is detected, and when the network is connected, the network management device is informed of the failure. If the network is in a disconnected state, the address stored in the storage means is changed to another address, and the network management device is notified to the network management device based on the other address. The communication apparatus according to claim 1, wherein the occurrence of a failure is notified. The failure notification means includes
When the communication device is activated, the storage unit is referred to, and when the information stored in the storage unit is other than the information indicating the address provided to the communication unit, the address assigned to the communication unit is stored in the storage unit. The communication apparatus according to claim 1, wherein the communication apparatus records the information. A communication method connected to a network management device via a network and having a plurality of communication means to which addresses are assigned in advance.
Storing information indicating that any one of the addresses is usable in a storage means;
Detecting a failure of the communication means;
If a failure is detected, determining based on the information whether the address assigned to the communication means is usable; and
As a result of the determination, if the address assigned to the communication means is usable, notifying the network management device of the occurrence of the failure;
As a result of the determination, when the address assigned to the communication means is unusable, the information stored in the storage means is rewritten with information indicating that another address can be used. Notifying the network management device of the occurrence of the failure based on an address;
A communication method comprising:

Images