JP4946804B2 - Network failure location identification system, terminal, relay device, network failure location identification method and program - Google Patents

Description

  The present invention relates to a technique for identifying a location where a failure has occurred in a communication network, and more particularly to a technology for identifying a location where a failure has occurred without degrading communication quality.

  As networks have become widespread, and among them, the use of rich content such as video and audio is increasing, the amount of data exchanged over the network is only increasing. In such a network, it goes without saying that techniques for detecting abnormalities in communication devices, lines, terminals, etc. are becoming increasingly important. In particular, in communication using rich content, it is required to detect an abnormality without degrading the quality of currently performed communication.

  A method commonly used for detecting an abnormality is a method called health check. In the health check, a dedicated packet (hereinafter referred to as health packet) is transmitted along with the packet of the current communication, and a reply from the computer or network device (hereinafter referred to as check target device) to be checked is sent. This is a method of receiving and confirming normal operation of the device to be checked.

  It is possible to detect an abnormality of the network or device from an external monitoring device by the health check. In addition, it is possible to specify in which part of the check target device an abnormality has occurred, depending on the type of command transmitted in the health packet. For example, the operation of layer 3 can be confirmed by checking the Ping command. By checking the port, the operation of layer 4 can be confirmed. The cause of the abnormality can be specified such as checking the protocol such as HTTP and confirming the operation of the application.

  The shorter the health check is performed, the more quickly the abnormality of the device to be checked can be identified and dealt with. However, since the health packet is added to the packet of the communication that is currently being performed, if the health check is performed at a short interval, the load on the communication device or the line increases. For this reason, it is necessary to distribute the load related to the health packet by an appropriate method.

  For example, the following techniques are disclosed as a method for distributing the load related to the detection of a network abnormality. Patent Document 1 discloses a technique for suppressing an increase in the load on a monitoring device by performing relay-type distributed health check using one of the computers on the network as a monitoring master. Patent Document 2 discloses a technology for reducing a communication load by aggregating a health check with a base device and communicating only the result between the bases in order to suppress an increase in communication between the bases due to the health check. Has been.

  Patent Document 3 discloses that the transmission cycle of health packets can be changed, normality / abnormality is determined for each target node system, the result is saved, and a transmission path is selected, so that a line abnormality can be detected even in a multistage relay system. A technique is disclosed that enables automatic switching by detecting this. Patent Document 4 discloses a technique for measuring a communication quality for each system and detecting a communication quality deterioration section. Patent Document 5 discloses a technique for detecting “fluctuation” from a moving average of delay times and thereby detecting an optimum communication path.

JP 2000-010949 A JP 2006-094180 A JP 2003-018217 A JP 2003-309602 A JP 2004-159146 A

  In the network, when an abnormality occurs in any of the relay devices while communication via a plurality of relay devices is performed, the communication is stopped or the communication quality is deteriorated. Therefore, it is necessary to quickly identify the location where an abnormality has occurred by an appropriate method. Therefore, health checks are performed at short intervals for all relay devices.

  As disclosed in Patent Documents 1 and 2, there is a technology for performing a health check by a relay method using a specific computer on a network for monitoring. However, they cannot monitor a communication path between relay apparatuses that is dynamically determined when communication is performed. Further, although these technologies reduce traffic between bases related to health check, they do not reduce the load related to health check for each relay device.

  Incidentally, Patent Document 3 only describes a method of switching transmission paths by health check, and Patent Documents 4 to 5 describe a method of detecting a decrease in communication quality, and these techniques can be combined with Patent Documents 1 and 2. It does not lead to the solution of the above problem.

  An object of the present invention is to provide a network fault location identification system, a terminal, and a terminal capable of specifying a fault occurrence location in a relay device and a communication path without applying an excessive load on the relay device in communication via a large number of relay devices. An object of the present invention is to provide a relay device, a network fault location identifying method, and a program.

  In order to achieve the above object, a network fault location identifying system according to the present invention includes at least a first application that operates on a first terminal and a second application that operates on a second terminal. A network for communicating by forming a channel via one or more relay devices and lines, and a system for identifying a fault location on the network, wherein the first terminal and the second terminal are connected to the channel. The first communication means for transmitting and receiving the first health check packet at the first interval and the time difference between the transmission time and the reception time of the first health check packet are detected, and fluctuation is detected by a statistical technique from the detected time difference. First fluctuation measuring means for detecting the fluctuation, first health check interval readjustment means for adjusting the first interval widely when fluctuation is detected, fluctuation, Transmission means for notifying the relay apparatus that the network has been detected, wherein the relay apparatus transmits / receives a second health check packet to / from the line at a second interval, and the transmission means fluctuates. And detecting the time difference between the second health check packet transmission time and the reception time and detecting the time difference between the second health check packet and the second health check interval readjustment means. And a second fluctuation measuring means for detecting fluctuation from the calculated time difference by a statistical method.

  In order to achieve the above object, a terminal according to the present invention includes at least one or more pieces between a first application operating on a first terminal and a second application operating on a second terminal. A network that performs communication by forming a channel via a relay device and a line, and is a terminal that forms a first terminal or a second terminal, and transmits and receives a health check packet to the channel at regular intervals Detection method, a fluctuation measurement means that detects the time difference between the transmission time and the reception time of the health check packet, and detects fluctuation from the detected time difference by a statistical method, and a health that adjusts the interval widely when fluctuation is detected. It is characterized by having a check interval readjustment means and a transmission means for notifying the relay apparatus that a fluctuation has been detected.

  To achieve the above object, at least one relay apparatus according to the present invention is provided between a first application that operates on a first terminal and a second application that operates on a second terminal. A relay unit and a network that forms a channel passing through a line and performs communication, the relay unit through which the channel passes, and a communication unit that transmits and receives health check packets to the line at a predetermined interval; Health check interval readjustment means for narrowing the interval when notified from the terminal or the second terminal that fluctuation has been detected, and the time difference between the transmission time and reception time of the health check packet is detected and detected. And a fluctuation measuring means for detecting fluctuation from the time difference by a statistical method.

  In order to achieve the above object, a network fault location identifying method according to the present invention includes at least a first application that operates on a first terminal and a second application that operates on a second terminal. A method for identifying a fault location on a network in a network in which communication is performed by forming a channel via one or more relay devices and lines, wherein the first terminal or the second terminal is connected to the channel. A first normal health check step for transmitting and receiving a first health check packet at a first interval, and a second normal health check step for the relay device to transmit and receive a second health check packet at a second interval to the line Then, the first terminal or the second terminal detects the time difference between the transmission time and the reception time of the first health check packet, and integrates from the detected time difference. A first fluctuation measuring step for detecting fluctuations by an artificial technique, and a first terminal or a second terminal for adjusting the first interval widely when fluctuations are detected in the first fluctuation measuring step. 1 health check interval readjustment step, a transmission step in which the first terminal or the second terminal notifies the relay device that fluctuation has been detected, and the relay device has detected fluctuation in the transmission step And the second health check interval readjustment step for adjusting the second interval narrowly in response to the second health check interval readjustment step, the relay device detects the time difference between the transmission time and the reception time of the second health check packet, and from the detected time difference And a second fluctuation measuring step for detecting fluctuation by a statistical method.

  In order to achieve the above object, a network failure location specifying program according to the present invention includes at least a first application that operates on a first terminal and a second application that operates on a second terminal. In a network that communicates by forming a channel via one or more relay devices and lines, a health check packet is sent to and received from the computer forming the first terminal or the second terminal at regular intervals. The process to detect the time difference between the transmission time and the reception time of the health check packet, detect the fluctuation from the detected time difference using a statistical method, and the health check to adjust the interval widely when fluctuation is detected It is characterized in that an interval readjustment process and a process of notifying a relay device that a fluctuation has been detected are executed.

  In order to achieve the above object, another network failure location specifying program according to the present invention is provided between a first application that operates on a first terminal and a second application that operates on a second terminal. A network in which communication is performed by forming a channel via at least one relay device and a line, and a process of transmitting and receiving health check packets to the computer forming the relay device at regular intervals; When it is notified from the first terminal or the second terminal that fluctuation has been detected, a process of adjusting the interval narrowly, and a time difference between the transmission time and the reception time of the health check packet is detected, and from the detected time difference And a process of detecting fluctuations by a statistical method.

  As described above, the present invention normally narrows the health check interval for the channel and widens the health check interval for the line. On the other hand, if a failure occurs, the health check interval for the channel is widened to the line. Since the interval between the health checks is narrowed, it is possible to perform a health check that makes it easy to identify the location where a failure has occurred in a range that does not deteriorate communication quality during normal times and does not become overloaded when a failure occurs. This makes it possible to identify the failure location of the relay device and the communication path without overloading the relay device in communication via a large number of relay devices. An identification system, a terminal, a relay device, a network failure location identification method, and a program can be provided.

  FIG. 1 is a block diagram showing a configuration example of a network 1 according to the embodiment of the present invention. In the network 1, a terminal 101a and a terminal 101b are connected via a relay device 102a and a relay device 102b. In the present invention, the number of terminals and relay devices and the number of lines connecting them are not particularly limited as long as they are plural. Here, in order to simplify the explanation and make it easier to understand the present invention, two examples are shown as minimum units. The relay device and the terminal are collectively referred to as a communication device.

  A plurality of applications (AP) 103a1 and 103a2 are operating on the terminal 101a. Also at the terminal 101b, the APs 103b1 and 103b2 are operating. Each of APs 103a1-2 and 103b1-2 may be one or plural. The terminal 101a and the relay apparatus 102a are connected by a plurality of lines 104a and 104b, and the relay apparatus 102a and the relay apparatus 102b are connected by a plurality of lines 104c and 104d. The relay apparatus 102b and the terminal 101b is connected by a plurality of lines 104e and 104f.

  FIG. 2 is a block diagram showing a channel 220 for connecting APs to each other in the network 1 disclosed in FIG. In FIG. 2, a channel 220 for establishing a one-to-one connection according to a known protocol such as TCP, HTTP, UDP, or the like is constructed between the AP 103a1 and the AP 103b1. Note that description of the AP 103a2, AP 103b2, the line 104b, the line 104d, and the line 104f is omitted.

  The channel 220 is formed by a route of terminal 101a-line 104a-relay apparatus 102a-line 104c-relay apparatus 102b-line 104e-terminal 101b, and connects between AP 103a1 and AP 103b1. Since the channel 220 connects APs that perform end-to-end communication, the channel 220 does not depend on the number of lines and communication devices on the communication path forming the channel 220.

  FIG. 3 is a block diagram illustrating the configuration of the communication device 300 disclosed in FIGS. 1 and 2. Note that the terminals 101a and 101b and the relay devices 102a and 102b illustrated in FIGS. 1 and 2 are collectively referred to as a communication device 300. The communication device 300 includes an application or relay unit 310 (hereinafter referred to as AP or relay unit 310), a communication unit 320, a failure detection unit 330, and a storage unit 301. The storage unit 301 stores communication information 410, channel information 420, and line information 430, which will be described later, and data can be read and written from both the communication unit 320 and the failure detection unit 330.

  The communication device 300 can operate as both a terminal and a relay device. For convenience, the communication device 300 that operates as a terminal may be referred to as a terminal 300a, and the communication device 300 that operates as a relay device may be referred to as a relay device 300b. However, the communication device of the present invention may include a device that operates only as a terminal and a device that operates only as a relay device.

  In the AP or relay unit 310, the AP 311 operates when the communication device 300 is the terminal 300a, and the relay unit 312 operates when the communication device 300 is the relay device 300b. The AP 311 communicates with the channel 321 of the communication unit 320, and the relay unit 312 communicates with the line 322. With this configuration, the AP 311 can communicate with the AP 311 of another communication apparatus using the channel 321, and the relay unit 312 uses the line 322 to be a terminal in which the channel 321 is mounted on the other relay apparatus 300 b and the line. Can communicate with 300a.

  The communication unit 320 includes a channel 321 and a line 322. There can be multiple of each. The channel 321 communicates with the AP 311 and the channel 321 is connected to the line 322 to communicate with other communication devices. It is not the channel 321 but the line 322 that actually communicates.

  The line 322 is connected to another communication device and performs communication related to data transmission / reception of the channel 321 on the line. When the communication device is the relay device 300b, there is a case where there is no channel 321 on the communication device and only the line 322 exists. In that case, communication from the channel constituted by another communication device is relayed on the line. . The channel 321 and the line 322 are respectively connected to the failure detection load adjustment unit of the failure detection unit 330, and can transmit and receive health packets. At that time, the communication unit 320 reads the communication information 410, the channel information 420, and the line information 430 from the storage unit 301, and performs communication based on the contents set by the information.

  The failure detection unit 330 includes a fluctuation measurement unit 331, a health check interval readjustment unit 332 (hereinafter referred to as health check interval readjustment unit 332), an interval change transmission unit 333 (hereinafter referred to as interval change transmission unit) to other devices. 333), and fault detection load adjusting means 334. Each operation will be described later.

  FIG. 4 is a block diagram showing a configuration of data stored and transmitted by the communication apparatus 300 shown in FIG. Data stored in the storage unit 301 of the communication device 300 includes communication information 410, channel information 420, and line information 430. In addition, the communication device 300 performs health checks by transmitting health packets 440 at predetermined intervals. Each data structure will be described below.

  The communication information 410 includes a communication ID 411 given for each communication performed in the network 1, a maximum interval 412 and a minimum interval 413 that hold the maximum value and minimum value of the health check interval in communication, and the current communication. It includes an execution interval 414 that is an interval for performing a health check, and a priority 415 that represents the priority of the current communication in the entire system.

  The channel information 420 includes a channel ID 421 given for each channel formed in the network 1, a communication ID 411 corresponding to the channel ID 421, an owned line 422 including a line ID 431 (described later) of the line forming the channel, and the channel , A post-change interval 423 at the time of quality degradation, which is an interval for performing a health check when communication quality degradation is detected.

  The line information 430 includes a line ID 431 given for each line forming the network 1, a communication ID 411 corresponding to the line ID 431, an accommodation channel 432 including a channel ID 421 included in the line, and deterioration in communication quality in the line. It includes a post-change interval 433 at the time of quality degradation, which is an interval for performing a health check when it is detected.

  The health packet 440 communicates to the other communication device 300 that the command 441 that is a command to the communication device 300 that is the transmission destination of the packet, the channel ID 421 and the line ID 431 that indicate the health check target, and that the deterioration of communication quality has been detected. Quality change information 442 to be transmitted, and a transmission time 443 that is a time at which the packet is transmitted. The command 441 describes the start, change, stop, etc. of the health check. If the health check target is a channel, a channel ID 421 is described, and if it is a line, a line ID 431 is described. The transmission time 443 is a reference time for detecting “fluctuation” described later.

  The fluctuation measurement unit 331 monitors the difference between the transmission time 443 and the reception time of the health packet 440 sent via the failure detection load adjustment unit 334, and communication quality is deteriorated when fluctuation more than a certain level occurs in the difference. Is determined to have occurred. Alternatively, when the health check packet does not reach the transmission destination within a certain time, it is determined that communication quality deterioration has occurred. Note that a method for detecting the occurrence of fluctuations by using a statistical method based on the difference between the transmission time 443 and the reception time in this manner is already known from the above-mentioned Patent Document 5 and the like. Since this is not the main point of the present invention, a detailed description is omitted.

  When it is determined that the quality degradation has occurred in the channel, the fluctuation measuring unit 331 notifies the health check interval readjustment unit 332 between the lines and channels of the communication quality degradation.

  Further, when the channel quality change information 442 is turned on in the health packet 440 received by the communication unit 320 from another communication device, the communication unit 320 notifies the health check interval readjustment unit 332 to perform communication. An instruction is given to change the health check interval of the line 322 accommodated in the apparatus.

  The storage unit 301 holds channel information 420 and line information 430 as data, thereby holding information on the channel 321 and the line 322. The channel information 420 includes the channel-owned line 422, and the line information 430 includes the accommodated channel 432. Therefore, the channel information 420 and the line information 430 hold the mutual relationship.

  The health check interval readjustment unit 332 operates upon receiving notification of channel quality deterioration from the fluctuation measurement unit 331. The health check interval readjustment unit 332 reads the after-change interval 423 when the quality of the channel information 420 is deteriorated and the after-change interval 433 when the quality of the channel information 430 is deteriorated, and the communication ID 411 of the channel information 420 when the quality of the channel is deteriorated. The communication maximum interval 412 and the minimum interval 413 and the communication maximum interval 412 and the minimum interval 413 indicated by the communication ID 411 of the line information 430 are changed.

  At this time, the priority 415 indicated in the communication information 410 is also changed, the channel is subtracted as −1, and the line is added as +1. Each priority 415 has an upper limit value and a lower limit value, and is not changed to a value exceeding the upper limit value or a value lower than the lower limit value. With the above operation, the maximum interval 412 and the minimum interval 413 and the priority 415 of the health check interval between the lines and channels in the communication apparatus 300 are changed. When changed, control is transferred to the interval change transmission means 333.

  The interval change transmission unit 333 operates according to the health check change request from the health check interval readjustment unit 332. By turning on the quality change status of the health packet 440 related to the channel 321, the change of the health check interval of the line related to the channel is notified to the other communication device 300. Thereafter, the control is transferred to the failure detection load adjusting unit 334, and the health check interval corresponding to the load of the communication device is determined.

  The failure detection load adjusting unit 334 is configured to set the maximum interval 412 and the minimum interval set for each communication so as not to exceed the load determined by the communication device based on the communication information 410 included in the channel information 420 and the line information 430. An execution interval 414 for performing an actual health check is determined based on the priority 415 within the range of values indicated by 413. Based on the determined execution interval 414, the channel 321 and the line 322 perform health checks in the communication unit 320. In the determination of the execution interval 414, any means can be applied as long as the means can be adjusted so as not to overload the communication device.

  FIG. 5 is a flowchart showing the health check operation executed by the terminal 300a shown in FIG. First, the operation is started (step S500), and the channel 220 is constructed (step S501). In the terminal 300a that forms both ends of the channel 220, the health check interval readjustment unit 332 performs setting so that the health check of the channel 220 is performed with a high priority and a narrow interval (step S502). Then, in the terminal 300a, the fluctuation measuring means 331 measures the difference between the transmission time 443 and the reception time, and whether or not fluctuation more than a certain level has occurred or received from another communication apparatus 300 (terminal 300a or relay apparatus 300b). It is determined whether or not the quality change information 442 is included in the health packet 440 (step S503).

  When it is determined that fluctuation has occurred or when the health packet 440 including the quality change information 442 is received, the health check interval readjustment unit 332 performs a health check on the channel 220 with a low priority and a wide interval. (Step S504). At the same time, the interval change transmission means 333 transmits a health packet 440 including quality change information 442 to change the health check interval to the other communication devices 300 (terminal 300a or relay device 300b) configuring the channel 220. (Step S505).

  When it is confirmed that the location where the failure has occurred is identified by the check performed by the relay device 300b that is the transmission source of the health packet 440, that is, it is confirmed that the health packet 440 transmitted in step S606 described later has been received (step S506). Returning to step S502, the health check at the normal time is resumed.

  FIG. 6 is a flowchart showing the health check operation executed by the relay device 300b shown in FIG. First, the operation is started (step S600), and the channel 220 is constructed (step S601). In the relay device 300b that relays the communication of the channel 220, the health check interval readjustment unit 332 sets the health check of the channel 220 to be performed with a low priority and a wide interval (step S602).

  Then, similarly to the terminal 300a, the fluctuation measuring means 331 measures the difference between the transmission time 443 and the reception time, and whether or not fluctuation more than a certain level has occurred, or another communication apparatus 300 (terminal 300a or relay apparatus 300b). It is determined whether or not the quality change information 442 is included in the health packet 440 received from (step S603).

  When it is determined that fluctuation has occurred or when the health packet 440 including the quality change information 442 is received, the health check interval readjustment unit 332 performs the health check of the channel 220 with a high priority and a narrow interval in the relay device 300b. (Step S604). At that time, the failure detection load adjustment unit 334 determines the execution interval 414 within the range of values indicated by the maximum interval 412 and the minimum interval 413, so that the communication apparatus 300 that is the target of the operation that performs the actual health check is performed. The load can be adjusted so as not to be higher than the set load.

  A health check is performed at the priority and interval set as described above, and a location where a failure has occurred is identified (step S605). At this time, if a deterioration in communication quality is detected at both ends of the communication apparatus 300 that forms both ends of a certain line, it can be regarded as a physical failure of the line. Further, if it is determined that the quality is deteriorated from the communication device 300 connected to both ends of a certain relay device, it can be considered that a failure has occurred in the central relay device.

  When the location where the failure has occurred is identified in step S605, the health packet 440 including the quality change information 442 is transmitted so as to restore the health check interval (step S606), and the process returns to step S602 to resume the normal health check. To do.

  In the operations illustrated in the flowcharts of FIGS. 5 to 6, it is described that the quality change information 442 is transmitted to the relay device 300 b when the terminal 300 a detects a fluctuation. However, since the health check is always performed on both the terminal 300a and the relay device 300b even at a low priority and a wide interval, the operation state (steps S503 and S603) when no failure has occurred. However, there is a possibility that the quality change information 442 is transmitted to the terminal 300a by detecting the fluctuation on the relay device 300b side. When the terminal 300a forming one end of the channel 220 detects a fluctuation and the quality change information 442 is transmitted, the terminal 300a forming the other end also receives the quality change information 442 and sets the health check interval. change.

  FIG. 7 is a table showing health check intervals that are actually set in the operations shown in FIGS. 5 to 6 performed in the network 1 shown in FIG. The unit of the health check interval is expressed in milliseconds (ms).

  Here, the channel A is connected between the AP 103a1 operating on the terminal 101a and the AP 103b1 operating on the terminal 101b. Channel A is formed by a route of terminal 101a-line 104a-relay apparatus 102a-line 104c-relay apparatus 102b-line 104e-terminal 101b.

  Further, the channel B is connected between the AP 103a2 operating on the terminal 101a and the AP 103b2 operating on the terminal 101b. Channel B is formed by a route of terminal 101a-line 104b-relay apparatus 102a-line 104d-relay apparatus 102b-line 104f-terminal 101b.

  In a state in which no fluctuation is detected (no failure has occurred), the terminal 101a and the terminal 101b are set with a health check interval of 100 ms for the channel A and 200 ms for the channel B. A health check interval of 10,000 ms is set for all the lines connected to 101b. Also in relay apparatuses 102a and 102b, a health check interval of 10000 ms is set for all lines connected to these relay apparatuses.

  When the terminal 101a or the terminal 101b detects a fluctuation for channel A (when a failure occurs), a post-change interval 423 at the time of quality degradation is applied to channel A, and a health check interval of 10000 ms is set. Then, the post-change interval 433 at the time of quality degradation is applied to the lines 104a, c, e, which are the lines forming the channel A, and a health check interval of 100 ms is set. In principle, the health check interval is not changed for channels B and lines 104b, d, and f forming channel B.

  However, for the line 104d, it is determined that the relay apparatuses 102a and 102b are overloaded by narrowing the health check interval for the channel A. Therefore, the failure detection load adjusting unit 334 of the relay apparatuses 102a and 102b determines the health for the line 104d. The check interval is further extended to 20000 ms. As the priority 415 in the channel A communication increases, the priority 415 in the channel B communication decreases, thereby causing such a change in the health check interval for the channel B having a lower priority. be able to.

  As described above, in the state where no failure has occurred, the load related to the health check is reduced in the relay device. As a result, even if the relay apparatus accommodates a line other than that shown in FIG. 1, it is possible to prevent a load from affecting the communication quality.

  When the occurrence of a failure is detected by detecting the fluctuation of the channel, the load relating to the health check of the channel is reduced, and a fine health check is performed on the line. This facilitates the identification of the location where the failure has occurred. In addition, the health check interval for the line can be set within a range in which the communication device does not become overloaded.

  Although the present invention has been described with the specific embodiments shown in the drawings, the present invention is not limited to the embodiments shown in the drawings, and is known so far as long as the effects of the present invention are achieved. It goes without saying that any configuration can be adopted.

  The present invention can be applied to a system that performs communication via a large number of relay devices. In particular, the present invention is suitable for a system that uses rich content such as video and audio, such as a system such as an IP phone or a WEB conference. Further, the present invention is not limited to end-to-end communication, and can be applied to, for example, a client-server communication system.

It is a block diagram shown about the structural example of the network which concerns on embodiment of this invention. FIG. 2 is a block diagram illustrating a channel for connecting APs to each other in the network disclosed in FIG. 1. It is a block diagram shown about the structure of the communication apparatus disclosed by FIG. 1 and FIG. FIG. 4 is a block diagram illustrating a configuration of data stored and transmitted by the communication device illustrated in FIG. 3. 4 is a flowchart illustrating a health check operation performed by the terminal illustrated in FIG. 3. 4 is a flowchart illustrating a health check operation performed by the relay device illustrated in FIG. 3. 7 is a table showing intervals of health checks that are actually set in the operations shown in FIGS. 5 to 6 performed in the network shown in FIG. 1.

Explanation of symbols

1 Network 101a, 101b, 300a Terminal 102a, 102b, 300b Relay device 103a1, 103a2, 103b1, 103b2, 311 AP (application)
104a, 104b, 104c, 104d, 104e, 104f, 322 Line 300 Communication device 220, 321 Channel 301 Storage means 310 AP or relay means 312 Relay means 320 Communication means 330 Fault detection means 331 Fluctuation measurement means 332 Health check interval readjustment means 333 Interval change transmission means (transmission means)
334 Fault detection load adjustment means 410 Communication information 415 Priority 420 Channel information 430 Line information 440 Health packet (health check packet)

Claims (21)

Communication is performed by forming a channel via at least one relay device and a line between the first application operating on the first terminal and the second application operating on the second terminal. A network for identifying a failure point on the network,
A first communication means for transmitting and receiving a first health check packet at a first interval between the first terminal and the second terminal, and a transmission time and a reception time of the first health check packet; A first fluctuation measuring means for detecting fluctuation by a statistical method from the detected time difference, and a first health for widely adjusting the first interval when the fluctuation is detected. Check interval readjustment means, and transmission means for notifying the relay device that the fluctuation is detected,
When the relay apparatus is notified that a second health check packet is transmitted / received to / from the line at a second interval and that a fluctuation is detected by the transmission unit, the second interval A second health check interval readjustment means for narrowly adjusting the time, a time difference between the transmission time and the reception time of the second health check packet is detected, and fluctuation is detected by a statistical method from the detected time difference A network fault location identifying system comprising two fluctuation measuring means.   2. The network fault location identification according to claim 1, wherein the transmission unit causes the first communication unit to transmit information including information related to the detection of the fluctuation in the first health check packet. system. The first terminal, the second terminal, and the relay device each have a failure detection load adjustment unit that adjusts the first interval and the second interval to fall within a specified range,
The first health check interval readjustment unit and the second health check interval readjustment unit adjust the first interval and the second interval by changing the designated range. The network fault location identifying system according to claim 2.   The first health check interval readjustment unit and the second health check interval readjustment unit change the first interval and the second interval, respectively, and simultaneously change the priority for each communication. The network fault location identifying system according to claim 3, wherein:   5. The network fault location specifying system according to claim 4, wherein the fault detection load adjusting means adjusts the first interval or the second interval widely for the communication with lower priority. . Communication is performed by forming a channel via at least one relay device and a line between the first application operating on the first terminal and the second application operating on the second terminal. A terminal that forms the first terminal or the second terminal, and
Communication means for transmitting and receiving health check packets to the channel at regular intervals;
Fluctuation measuring means for detecting a time difference between the transmission time and the reception time of the health check packet, and detecting fluctuations by a statistical method from the detected time difference;
Health check interval readjustment means for widely adjusting the interval when the fluctuation is detected;
And a transmission means for notifying the relay apparatus that the fluctuation has been detected.   The terminal according to claim 6, wherein the transmission unit causes the communication unit to transmit information including that the fluctuation is detected in the health check packet. It has a failure detection load adjusting means for adjusting the interval so that it falls within a specified range,
The terminal according to claim 7, wherein the health check interval readjustment unit adjusts the interval by changing the designated range.   9. The terminal according to claim 8, wherein the health check interval readjustment unit changes the priority for each communication simultaneously with changing the interval. Communication is performed by forming a channel via at least one relay device and a line between the first application operating on the first terminal and the second application operating on the second terminal. A relay device through which the channel passes,
Communication means for sending and receiving health check packets to the line at regular intervals;
Health check interval readjustment means for narrowly adjusting the interval when notified of fluctuation from the first terminal or the second terminal;
A relay apparatus, comprising: a fluctuation measuring unit that detects a time difference between a transmission time and a reception time of the health check packet and detects a fluctuation from the detected time difference by a statistical method.   The health check interval readjustment means adjusts the interval narrowly when information relating to the detection of the fluctuation is detected in the health check packet received from the first terminal or the second terminal. The relay device according to claim 10, wherein the relay device is characterized. It has a failure detection load adjusting means for adjusting the interval so that it falls within a specified range,
12. The relay apparatus according to claim 11, wherein the health check interval readjustment unit adjusts the interval by changing the designated range.   The relay apparatus according to claim 12, wherein the health check interval readjustment unit changes the priority for each communication at the same time as changing the interval.   The relay apparatus according to claim 13, wherein the failure detection load adjustment unit adjusts the interval widely for the communication with the lower priority. Communication is performed by forming a channel via at least one relay device and a line between the first application operating on the first terminal and the second application operating on the second terminal. A method for identifying a fault location on the network,
A first normal health check step in which the first terminal or the second terminal transmits and receives a first health check packet to the channel at a first interval;
A second normal health check step in which the relay device transmits and receives a second health check packet to the line at a second interval;
The first terminal or the second terminal detects a time difference between the transmission time and the reception time of the first health check packet, and detects fluctuation by a statistical method from the detected time difference. Fluctuation measurement process,
A first health check interval readjustment step in which the first terminal or the second terminal adjusts the first interval widely when fluctuation is detected in the first fluctuation measurement step;
A transmission step in which the first terminal or the second terminal notifies the relay apparatus that the fluctuation has been detected;
A second health check interval readjustment step in which the relay device adjusts the second interval narrowly in response to fluctuation detected in the transmission step;
The relay apparatus includes a second fluctuation measuring step of detecting a time difference between a transmission time and a reception time of the second health check packet and detecting fluctuation from the detected time difference by a statistical method. A characteristic network fault location identification method.   16. The network failure location specifying method according to claim 15, wherein the transmission step includes transmitting information to the relay apparatus including information on the fluctuation detected in the first health check packet.   The first health check interval readjustment step and the second health check interval readjustment step each change a designated range in which the first interval and the second interval can be adjusted. The network fault location identifying method according to claim 16.   The first health check interval readjustment step or the second health check interval readjustment step changes the first interval and the second interval, and simultaneously changes the priority for each communication. The network fault location identifying method according to claim 17, wherein:   The first health check interval readjustment step or the second health check interval readjustment step adjusts the first interval or the second interval widely for the lower priority communication. The network fault location identifying method according to claim 18, characterized in that it is characterized in that: Communication is performed by forming a channel via at least one relay device and a line between the first application operating on the first terminal and the second application operating on the second terminal. In the network to perform, the computer forming the first terminal or the second terminal,
Processing to send and receive health check packets to the channel at regular intervals;
Detecting a time difference between the transmission time and the reception time of the health check packet, and detecting fluctuations by a statistical method from the detected time difference;
A health check interval readjustment process for widely adjusting the interval when the fluctuation is detected;
A network fault location specifying program that executes processing for notifying the relay device that the fluctuation has been detected. Communication is performed by forming a channel via at least one relay device and a line between the first application operating on the first terminal and the second application operating on the second terminal. In the network to perform, to the computer forming the relay device,
Processing to send and receive health check packets to the line at regular intervals;
A process of narrowing the interval when notified of fluctuation from the first terminal or the second terminal;
A network fault location specifying program that detects a time difference between a transmission time and a reception time of the health check packet and executes a process of detecting fluctuations by a statistical method from the detected time difference.

Images