JP6503780B2 - Network monitoring system and network monitoring method - Google Patents

Description

  The present invention relates to a network monitoring system and a network monitoring method, and more particularly to redundancy of periodic monitoring of a network monitoring system and a network monitoring method.

  In recent years, the functions of network devices are diverse, and the contents to be regularly monitored, such as the number of network devices, the sites managed by the network devices, and the types of failures in the sites, are expanding. Network monitoring devices that monitor the network devices are required to perform maintenance regularly, continuously, and safely.

  Therefore, the network monitoring system was considered to prepare a plurality of network monitoring devices and monitor the network devices, but the problem that the load on the network devices increases if the plurality of network devices are accessed simultaneously is there. Therefore, although it is necessary to perform periodic monitoring from the network monitoring device to the network device, it is necessary to perform the monitoring from a specific one network monitoring device, but if a specific network monitoring device and the network device are linked and monitored, There is a problem that it can not be relieved if there is a problem with the network monitoring device. In addition, there is a problem that allocation of periodical monitoring according to the recovery method and processing capacity in the case of failure of the network monitoring device becomes complicated when periodical network devices are periodically monitored from a specific network monitoring device without linkage. .

  As a technique related to the above, in Patent Document 1, even when any of a plurality of monitoring servers is stopped, monitoring is performed without interrupting the monitoring of the network element that the monitoring server that has stopped the monitoring process is monitoring targets. The method to continue is described. This is a technology in which a plurality of monitoring servers monitoring the monitoring target receive allocation information for extracting the monitoring target network element from the plurality of network elements, and monitor the monitoring target network element according to the allocation information. It is. When the management server connected to the plurality of monitoring servers detects that the monitoring processing by any of the plurality of monitoring servers has stopped, the monitoring processing is performed on the monitoring servers other than the monitoring server that has stopped the monitoring processing. The monitoring server that has stopped transmitting transmits allocation information for extracting a network element including a network element to be monitored as a monitoring target.

  Further, in Patent Document 2, in order to provide a network monitoring type determining apparatus capable of performing appropriate network monitoring, the apparatus includes a plurality of monitored nodes monitored by a plurality of monitoring nodes through the line 1, and each monitored node A technique is disclosed comprising transmitting means for transmitting the own monitored node number to each monitoring node. Further, the relay number calculation means calculates the number of relays of the monitored node to be relayed from each monitoring node to each monitored node for each monitoring node based on the own monitored node number from the transmitting means, and determines the monitoring mode. A technique is disclosed that determines the monitoring mode so as to share the monitored node with the monitoring node whose relay count is the smallest by comparing the relay counts of the same monitored node among the monitoring nodes.

Unexamined-Japanese-Patent No. 2010-087834 Japanese Patent Application Publication No. 07-087088

  However, since patent document 1 monitors the monitoring object according to the predetermined allocation conditions, when there exists a difference in the processing capacity of a monitoring apparatus, the subject that the progress of processing of a monitoring apparatus comes out bias was there.

  Moreover, since the patent document 2 determines a monitoring form using the network monitoring form determining apparatus which has a means to calculate the number of times of relay, there existed a subject that a network monitoring form determining apparatus was needed.

  SUMMARY OF THE INVENTION The object of the present invention is to address this problem, and to monitor a network periodically, even if a failure occurs in the network monitoring device, the network and the network device without using a dedicated device. It is to make the device redundant.

  In the present invention, in order to solve the above problems, in a network monitoring system provided with a plurality of network monitoring devices monitoring a plurality of network devices, the network monitoring device is controlled to change the network devices to be monitored every unit time. It is characterized by

  Further, in the present invention, in a network monitoring method of a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices, the network monitoring device controls to change the network devices to be monitored every unit time. It is characterized by having.

  According to the present invention, it is possible to make a network monitoring device redundant in order to continuously monitor periodically even if a failure occurs in the network monitoring device, the network and the network device without using a dedicated device. It becomes.

FIG. 1 is a block diagram showing the configuration of a network monitoring system according to an embodiment of the present invention. It is a sequence chart explaining operation of a network surveillance system in an embodiment of the invention. It is a figure which shows the example of the periodic monitoring content which should be implemented with respect to the object network apparatus in embodiment of this invention, and its apparatus. It is a figure which shows the information of the network monitoring apparatus in embodiment of this invention. It is a figure which shows the state of the network monitoring apparatus in embodiment of this invention. It is a figure which shows the list | wrist of the network apparatus which the own network monitoring apparatus in the embodiment of this invention could not monitor. It is a figure which shows which network monitoring apparatus should process the periodic monitoring item which became 0 in the remaining time in embodiment of this invention. FIG. 7 is a diagram for performing a rescue that is created when the network device in the embodiment of the present invention can access from the own network monitoring device. It is a flowchart which shows the operation | movement in embodiment of this invention. It is a block diagram showing composition of one network surveillance system in the present invention.

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

  The configuration of the embodiment of the present invention will be described using FIG. FIG. 1 is a block diagram showing the configuration of a network monitoring system according to the embodiment.

  In FIG. 1, a network monitoring system 100 is configured of a plurality of network monitoring devices and a plurality of network devices.

  The network 1 is a network in which network monitoring devices are mutually monitorably connected. That is, the first network monitoring device 2, the second network monitoring device 3, and the n-th network monitoring device 4 are monitored. Reference numeral 5 denotes a network for monitoring network devices from the network monitoring device, and a network for monitoring the first network device 8 from each of the network monitoring devices. 6 is a network for monitoring the network device from the network monitoring device, and is a network for monitoring the second network device 9 from each network monitoring device. Reference numeral 7 denotes a network for monitoring network devices from the network monitoring device, and a network for monitoring the m-th network device 10 from each of the network monitoring devices.

  Thus, each network monitoring device is connected to be able to monitor all network devices.

  Next, the configuration of the present embodiment will be described in detail with reference to FIGS.

  FIG. 2 shows the network monitoring system of this embodiment as a sequence chart. In FIG. 2, the same components as those in FIG. 1 are denoted by the same reference numerals.

  The maintenance person 21 issues an instruction 28 to the network monitoring apparatus in order to perform periodic monitoring. The instruction 28 instructs the first network monitoring device 2, the second network monitoring device 3, ..., the n-th network monitoring device 4 to periodically monitor the networks 5, 6, and 7. It shows the operation. Here, a list of network devices to be regularly monitored shown in FIG. 3 to be described later, a list of all network monitoring devices shown in FIG. 4, and information on the start point t0 of periodic monitoring are expanded. Information other than the information of the start point t0 may be set in advance.

  The start point t0 designates the time after the maintenance person 21 gives the instruction 28. Further, in the present embodiment, time synchronization is performed to perform processing at the same time between the network monitoring devices. For example, NTP (Network Time Protocol) may be used for time synchronization, but any method may be used as long as time synchronization can be achieved.

  The monitoring operation 29 shows an operation of mutually monitoring life and death among the network monitoring devices 2, 3,. Monitoring is periodically performed within time t which is the minimum unit of the monitoring cycle, and when there is a change in the life and death of each network monitoring device, the network monitoring device viability information of FIG. 5 (2) Update from alive to alive or stop to alive.

  FIG. 5 is a list showing the status of the network monitoring device. This FIG. 5 may be generated each time monitoring, or may be generated in advance, and may be changed in accordance with the survivability state obtained by monitoring. This information is determined every time t and used to determine a network monitoring device to perform periodic monitoring. The alive monitoring may be, for example, simultaneous distribution of survival information by broadcasting of a UDP (User Datagram Protocol), but it is sufficient if the alive can be monitored by any method.

  Also, in the monitoring operation 29, a list of device information that can not be monitored by the network monitoring device shown in FIG. 6 described later is advertised. This advertisement may be used to determine the alive status of the network monitoring device. In the present specification, advertisement has the same meaning as broadcast notification. When the network monitoring device can not monitor, for example, it is a case where a certain network monitoring device is disconnected from the network device to be monitored.

  The monitoring operation 30 shows an operation of determining a network device to be monitored at time t0 and performing periodic monitoring. The details of this internal operation are shown in the flowchart showing the operation of the embodiment of FIG. 9 described later. The monitoring operation 31 indicates an operation of determining a network device to be monitored at time t1 and performing periodic monitoring. The details of this internal operation are shown in the flowchart showing the operation of the embodiment of FIG. 9 described later. The monitoring operation 32 shows an operation of determining a network device to be monitored at time tx and performing periodic monitoring. The details of this internal operation are shown in the flowchart showing the operation of the embodiment of FIG. 9 described later.

  FIG. 3 shows an example of a target network device and contents of periodic monitoring to be performed on the device. All network monitoring devices share the same information in FIG.

  (1) of FIG. 3 is a number indicating the fixed and allocated process. The numerical value is an integer of 1 or more. Here, it is assumed that processing contents i exist.

  (2) of FIG. 3 shows network device information to be subjected to periodical monitoring. For example, it is an IP (Internet Protocol) address of a network device.

  (3) of FIG. 3 shows the contents of treatment to be regularly monitored. For example, collection of failure information, execution of a specific command, collection of normal information such as a packet discard rate, collection of a configuration change state, and the like. A plurality of items of periodic monitoring can be set for one network device, such as N_3_c and N_3_d.

  (4) of FIG. 3 is a processing weight. Set an integer greater than 1 and less than the network monitoring device. The processing weight is a numerical value indicating how many times the processing unit of the monitoring item is 1 when it is one. For example, if the processing to be performed in one monitoring item is OK but the processing is not performed twice unless the same processing is performed twice, processing weight 2 is set.

  (5) of FIG. 3 is a collection period. It shows by what time t which is the minimum time unit.

  (6) of FIG. 3 is the remaining time. Every time t elapses, 1 t is subtracted. If the remaining time becomes 0, periodic monitoring processing is performed on the target network device in accordance with the periodic monitoring processing content of (3). When the maintenance person 21 requests the instruction 28, the initial time is set to a time within a collection cycle of 1t to (5) so as to be dispersed in time. The process in which the remaining time has become 0 is regularly monitored, and after completion of the periodic monitoring, the collection cycle is added again.

  FIG. 4 shows information of the network monitoring device. All network monitoring devices have the same information.

  (1) in FIG. 4 is a number indicating the number of the network monitoring device allocated fixedly. The numerical value is an integer of 1 or more. Here, it is assumed that the n-th unit exists.

  (2) of FIG. 4 is network monitoring device information. For example, it is an IP address for connecting to a network monitoring apparatus.

  (3) of FIG. 4 shows the processing capacity of each network monitoring device. The numerical value is an integer of 1 or more.

  FIG. 5 is a list showing the status of the network monitoring device. This FIG. 5 may be generated each time monitoring, or may be generated in advance, and may be selected in accordance with the survival status.

  (1) of FIG. 5 is a number indicating the number of the fixedly allocated network monitoring device, which is the same as (1) of FIG. 4 for linking with (1) of FIG.

  (2) of FIG. 5 is the survivability information of the network monitoring device. Life and death monitoring information is updated by the processing of the monitoring operation 29 of FIG.

  (3) of FIG. 5 is the range number of the weighted index. This value represents a range indicating which network monitoring device should perform the process when an index number shown in (2) of FIG. 7 described later is assigned. This index range number is not assigned when the network monitoring device is stopped. The head of the index range number, that is, the youngest index number given that there are a plurality of network monitoring devices that perform the processing, 0 ≦ index number <processing capacity addition value of (3) in FIG. The maximum value of the weighted index range in the column ≦ index number <The processing capacity value of FIG. 4 (3) corresponding to the number of the network monitoring device to be set.

  In addition, if it is known whether all network monitoring devices are viable, it is possible to calculate the total processing capacity of the surviving network monitoring devices. This value is ww in this embodiment. ww is an integer of 1 or more. When all network monitoring devices are alive, the value of ww is (W_1 + W2 + W3 + ... W_n). This value is used in FIG.

  That is, in the present embodiment, a number is allocated to the target device by an appropriate calculation formula every unit time, and the monitoring device to which the number is assigned monitors the target device per unit time. The assigned number is the index number, and the assignment of the monitoring device is the index range.

  FIG. 6 is a list of network devices that could not be monitored by the network monitoring device assigned for monitoring. If the network monitoring device detects a network device that can not be monitored, it advertises to other network monitoring devices in the monitoring operation 29 of FIG. This FIG. 6 is processed by aggregating and processing information from other network monitoring devices, and is used to create FIG.

  (1) of FIG. 6 shows information of the network monitoring device. For example, it is its own IP address.

  (2) in FIG. 6 shows information on network devices that the network monitoring device assigned for monitoring can not access. For example, it is the IP address of the network device that could not be accessed.

  FIG. 7 is a list showing on which network monitoring device the periodic monitoring item generated at every cycle t and for which the remaining time of (6) in FIG. 3 is 0 should be processed. That is, it is a list that designates which network monitoring device is to be allocated for monitoring the network device.

  (1) of FIG. 7 is a list of processes in which the remaining time of (6) of FIG. 3 is zero.

  (2) of FIG. 7 is an index number. A numerical value obtained by adding the processing weight of FIG. 3 (4) of the previous item to the index number of the previous item is a remainder value divided by the processing capacity ww of all the network monitoring devices described above in FIG. The index number of the first item is a surplus value obtained by dividing k (tx) by the processing capacity ww. This extra value is expressed using mod. Here, tx is time elapsed (an integer greater than or equal to 0), and k () is an integer greater than or equal to 0 that outputs a distributed numerical value that is nonlinear. This is because the network monitoring device monitors different network devices as much as possible, and if this process is performed using a linear value, it is possible to have a pattern in which the same network monitoring device monitors the same network device. It is conceivable that k () is, for example, the x value multiplied by a prime greater than the processing power ww, but any calculation formula may be used as long as dispersed non-linear values can be obtained. Starting from the network monitoring device of the index number designated first, the network device to be monitored is changed every unit time.

  A value that is linear here, for example, the same value as tx, can also be given, but in that case, the network device monitored by the network monitoring device may become stationary. For example, if the collection cycle is 4 t, four network monitoring devices, and the processing capabilities of the network monitoring devices are equal, the target network monitoring devices listed in the value range shown in (3) of FIG. Become. In the above case, be sure to set the collection cycle to t. This is to provide a remedy when monitoring between a specific network monitoring device and a specific network device is not possible.

  In (3) of FIG. 7, target network monitoring device information calculated from the index range shown in (3) of FIG. 5 is set to the index number set in (2) of FIG. A network monitoring device having the same information as the network monitoring device information periodically monitors the network device shown in (1) of FIG. At this time, when the network apparatus which can not be accessed by the network monitoring apparatus of FIG. 6 is successfully accessed by accessing the network apparatus, FIG. 8 is updated.

  FIG. 8 is a list for performing rescue, which is created when the network monitoring device updated and allocated every cycle t can access the network device shown in FIG.

  FIG. 8 (1) is a list of network monitoring device information obtained in FIG. 6 (1). This list includes one or more network monitoring devices.

  (2) of FIG. 8 shows network device information that can not be periodically monitored that could not be monitored in the list of network monitoring devices obtained in (2) of FIG. However, network devices that can not be monitored from the allocated network monitoring device do not register.

  (3) of FIG. 8 shows the periodic monitoring operation of (1) of FIG. 7 (a process in which the remaining time of FIG. 3 has become 0). However, if the target network monitoring device information in (3) of FIG. 7 does not include the network information created in (1) of FIG. 8, registration is not performed. In the list of one network monitoring device information, a plurality of numbers may be set and there may be a plurality of lines.

  (4) of FIG. 8 is an index number. The index number is set in the same manner as (2) in FIG. 7 for each list of devices in (1) in FIG. The initial value is a remainder value obtained by dividing the same k (tx) as in (2) of FIG. 7 by ww, and the index number returns to the initial value when the contents of the list of network monitoring device information change. Here, the ww value is the total value of (3) processing capacities of the surviving network monitoring device information in FIG. 4, but the list value of the network monitoring device information is further subtracted.

  (5) in FIG. 8 is target network monitoring device information obtained from the target device index number. Here, without using the weighted index range of (3) of FIG. 5, network monitoring in which the network monitoring device information list of (1) of FIG. 8 is added to the survivability information of (2) of FIG. Use weighted index range information recomputed as the device is down.

  The operation of the embodiment of the present application will be described using FIGS. 2, 3, 5, and 7 to 9.

  FIG. 9 is a flowchart showing the operation of the monitoring operations 30 to 32 of FIG. 2 as the operation of the embodiment.

  In FIG. 2, when periodic monitoring is started at the start point t0, first, in the item of remaining time in FIG. 3, an item which is subtracted by 1 t and becomes 0 is detected (S101). List the items that will be 0 in the remaining time, arrange the items in order, and set other items, Fig. 7 In other words, a list showing which network monitoring device should process the periodic monitoring items whose remaining time is 0 Create (S102).

  The range of the weighted index of (3) of FIG. 5 is set from the existence state of the network monitoring device of (2) of FIG. 5 (S103). This FIG. 5 may be generated each time the monitoring is performed, or may be generated in advance and selected according to the survival status.

  Based on the item which becomes 0 in the remaining time of FIG. 7, the processing weight of (4) of FIG. 3, and time tx, the index number of (2) of FIG. 7 is assigned (S104). From the content set in FIG. 7 (2) and the weighted index range set in (3) of FIG. 5, the target network monitoring device information of (3) of FIG. 7 is set (S 105). If the target network monitoring device information in (3) of FIG. 7 is the own network monitoring device, periodic monitoring is performed on the target network device (S106). Here, with regard to the network device advertised as being unable to be regularly monitored shown in FIG. 6, periodic monitoring is not performed and only life and death are monitored. This is because, as described in FIG. 8, another network monitoring device may monitor the corresponding network device.

  Based on the information (information of FIG. 6) of the network devices that can not be regularly monitored from the network monitoring devices obtained in the monitoring operation 29 of FIG. 2, the network monitoring device information of FIG. To do (S107). However, if it can not be monitored from its own network monitoring device, it will not be placed on the list.

  Based on the network apparatus information that can not be regularly monitored in FIG. 8 and the information on the item for which the remaining time in FIG. 7 has become 0, the information on the item for which the remaining time in (3) of FIG. 8 becomes 0 is set (S108). However, when (3) of FIG. 7 is included in the list of network monitoring device information of (1) of FIG. 8, it is not put on the list. At this time, there may be a plurality of set lines in (1) and (2) of FIG.

  The index number of (4) of FIG. 8 is set from the information of the item in which the remaining time of (3) of FIG. 8 is 0 (S109).

  The target network monitoring device of (5) of FIG. 8 is set from the information of (4) of FIG. 8 (S110).

  When the target network monitoring device in (5) of FIG. 8 is the own network monitoring device, periodic monitoring is performed on the target network device (S111).

  For the next cycle, the time of the collection cycle of (5) of FIG. 3 is added to the item of (6) of FIG. 3 where the remaining time is 0 (S112).

  As described above, in the present embodiment, the network monitoring device's processing capability and the weight of the monitoring content of the network device are checked at predetermined intervals, and the network monitoring device monitors each network monitoring device according to the check result. The device can be changed (decided). Therefore, even if there is a difference in processing capacity among a plurality of network monitoring devices, it is possible to alleviate the processing bias toward a specific network monitoring device. Furthermore, if there is a network device that can not be monitored from a certain network monitoring device, have another network monitoring device monitor this network device, and if monitoring is possible, exclude the network monitoring device that could not be monitored. The network devices to be monitored for each network monitoring device can be re-determined.

  FIG. 10 is a block diagram showing the configuration of one network monitoring system.

  Although the embodiment of the present invention is configured of a plurality of networks, in FIG. 10, it is configured of one network 204.

  The network monitoring system 200 has one network configuration, and includes a first network monitoring device 201, a second network monitoring device 202,..., An n-th network monitoring device. Further, it has a first network device 205, a second network device 206,..., An m-th network device 207. The network 204 is connected to monitor a plurality of network devices from a plurality of network monitoring devices.

  In this case, since there are no network devices that can monitor one network monitoring device but can not monitor another network monitoring device, the information in FIGS. 6 and 8 is not necessary.

  Further, in the present embodiment, when updating the table of FIG. 8, the network device that can not be regularly monitored is the network device that is to be monitored in FIG. 7, but the periodic monitoring content of FIG. If there are few network devices that are not to be monitored, there is a possibility that relief will not be performed. In that case, alive monitoring may be performed on each network device, and a network device whose connection has been confirmed may be the target of FIG. Further, in the present embodiment, when alive monitoring is not normally performed among the network monitoring devices but periodic monitoring can be performed on the network devices, periodic monitoring is performed simultaneously from a plurality of locations from the network monitoring device to the network monitoring device. In order to avoid this event, check the execution log of the network device, and if another network monitoring device requests periodic monitoring at the same time zone, take measures to prevent double periodic monitoring, such as not performing the corresponding periodic monitoring. You may introduce it.

  As described above, according to the present embodiment, a number is allocated to a target device by an appropriate calculation formula every unit time, and a network monitoring device to which the number is assigned is a target network device per unit time Monitor. Therefore, even if a failure occurs in the network monitoring device, the network, and the network device without using a dedicated device, it is possible to make the network monitoring device redundant in order to continuously monitor periodically.

  The present invention is not limited to the above-described embodiment, and various changes and modifications can be made without departing from the scope of the present invention.

  INDUSTRIAL APPLICABILITY The present invention is applicable to the field of network monitoring systems that periodically and continuously monitor the status of network devices such as line fault status, processing load status, and power consumption.

1 Network 2 Network Monitoring Device S_1
3 Network monitoring device S_2
4 Network monitoring device S_n
5 Network 6 Network 7 Network 8 Network Device N_1
9 Network Device N_2
10 Network Device N_m
21 Maintenance person 28 Instruction 29 Monitoring operation 30 Monitoring operation 31 Monitoring operation 32 Monitoring operation 100 Network monitoring system 200 Network monitoring system 201 Network monitoring device S_1
202 Network monitoring device S_2
203 Network monitoring device S_n
204 Network 205 Network Device N_1
206 Network Device N_2
207 Network Device N_m

Claims (9)

In a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices,
The network monitoring device controls to change the network device to be monitored every unit time.
A network monitoring system characterized in that the network monitoring device is controlled by the network monitoring device specified first to start changing the network device to be monitored to another network device per unit time. . In a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices,
The network monitoring device controls to change the network device to be monitored every unit time.
The network monitoring system, wherein the time taken for monitoring per unit time is unit time. In a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices,
The network monitoring device controls to change the network device to be monitored every unit time.
A network monitoring system, wherein the network monitoring apparatus weights and monitors the same network apparatus by monitoring the same network apparatus a plurality of times. In a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices,
The network monitoring device controls to change the network device to be monitored every unit time.
The network monitoring system, when changing the network devices to be monitored every unit time, selects the network devices to be monitored in a non-linear order. In a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices,
The network monitoring device controls to change the network device to be monitored every unit time.
When the network monitoring device controls the other network monitoring device to monitor when the non-monitoring network device is present, and the network monitoring device with the non-monitoring network device monitors next A network monitoring system for excluding and monitoring the unmonitorable network device.   The network monitoring system according to claim 1, wherein the time required for monitoring per unit time is set as a unit time.   The network monitoring system according to any one of claims 1, 2 and 6, wherein the network monitoring device performs weight monitoring by monitoring the same network device a plurality of times.   8. The network monitoring device according to claim 1, wherein the network monitoring device selects the network device to be monitored in a non-linear order when changing the network device to be monitored every unit time. Network monitoring system according to one of them. In a network monitoring method of a network monitoring system comprising a plurality of network monitoring devices monitoring a plurality of network devices,
The network monitoring device has a step of controlling to change the network device to be monitored every unit time.
The network monitoring method comprises the step of controlling the other network monitoring device to monitor if there is the network device which can not be monitored.

Images