Silent failure detection device and silent failure detection program

Abstract

To accurately detect a silent failure in a communication network.SOLUTION: The silent failure detection device includes a PM information acquisition unit and a silent failure detection unit. The PM information acquisition unit periodically acquires performance monitor information indicating a communication state from each of a plurality of network devices constituting a communication network. The silent failure detection unit determines values of a plurality of failure determination parameters based on the performance monitor information acquired by the acquisition unit, and determines whether or not a silent failure has occurred in the communication network based on a failure determination score calculated from the values of the plurality of failure determination parameters.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an apparatus, a method, and a software program for detecting silent faults occurring in a communication network. [Background technology]

[0002] In many cases, communication networks are equipped with a function that detects a failure and outputs an error message. In such cases, the location of the failure can be identified based on the error message, allowing the administrator of the communication network to deal with the failure at an early stage.

[0003] However, not all failures are detected, and sometimes no error message is output even when a failure has occurred. In the following description, such failures may be referred to as "silent failures." Note that silent failures include not only cases where a failure actually occurs, but also "signs of an impending failure."

[0004] In this situation, a fault detection device has been proposed that detects faults occurring in a communication network at low cost and early stage when the impact of the fault is relatively small (for example, Patent Document 1). Related techniques are also described in Patent Documents 2 to 5. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-072723 [Patent Document 2] Japanese Patent Application Laid-Open No. 2009-017393 [Patent Document 3] US Patent Application Publication No. 2018 / 0227208 [Patent Document 4] International Publication No. 2023 / 084599 [Patent Document 5] Japanese Patent Application Laid-Open No. 2003-244146 Summary of the Invention [Problem to be solved by the invention]

[0006] As mentioned above, techniques for detecting silent faults have been proposed. However, currently, silent faults often affect communication. Therefore, a method for detecting silent faults more accurately is required.

[0007] An object of one aspect of the present invention is to provide an apparatus and method for accurately detecting silent failures in a communication network. [Means for solving the problem]

[0008] A silent failure detection device according to one aspect of the present invention includes an acquisition unit that periodically acquires performance monitor information indicating a communication state from each of a plurality of network devices that constitute a communication network, and a detection unit that detects a silent failure occurring in the communication network based on the performance monitor information acquired by the acquisition unit. The detection unit determines values ​​of a plurality of fault determination parameters based on the performance monitor information, and determines whether a silent failure has occurred in the communication network based on a fault determination score calculated from the values ​​of the plurality of fault determination parameters. [Effects of the Invention]

[0009] According to the above-described aspect, silent failures can be detected accurately in a communication network. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of a communication network in which a silent fault detection device according to an embodiment of the present invention is used. [Figure 2] FIG. 1 illustrates an example of a configuration of a network device. [Figure 3]FIG. 2 is a diagram illustrating an example of a functional configuration of a silent failure detection device. [Figure 4] 10A and 10B are diagrams illustrating an example of a counter difference value calculated by a difference calculation unit. [Figure 5] 10 is a flowchart illustrating an example of a process for detecting a silent failure. [Figure 6] FIG. 10 illustrates an example of a difference information management table. [Figure 7] 10 is a flowchart illustrating an example of a release process. [Figure 8] FIG. 6 is a diagram illustrating an example of a use case for verifying the procedure of the flowchart illustrated in FIG. 5. [Figure 9] FIG. 1 illustrates an example of a hardware configuration of a silent failure detection device. DETAILED DESCRIPTION OF THE INVENTION

[0011] FIG. 1 shows an example of a communication network in which a silent fault detection device according to an embodiment of the present invention is used. The communication network 1 according to the embodiment of the present invention includes a network element (NE) 2 at each node. The network element 2 transmits an optical signal at the physical layer. The optical signal may be a WDM (Wavelength Division Multiplexed) signal. The network element 2 can also transmit and receive packets. The packets may be, for example, IP packets.

[0012] A network management system (NE-OPS: Network Element Operation System) 3 monitors the state of the communication network 1 and controls the operation of each network device 2. At this time, the network management system 3 may collect performance monitor information from each network device 2. In this case, the performance monitor information represents the communication state detected or measured in each network device 2.

[0013] The silent failure detection device 10 is implemented in the network management system 3. The silent failure detection device 10 periodically acquires performance monitor information from each network device 2. Then, based on the acquired performance monitor information, the silent failure detection device 10 determines whether or not a silent failure has occurred in each network device 2.

[0014] Fig. 2 shows an example of the configuration of the network device 2. The network device 2 includes a receive buffer 21, an FCS processing unit 22, a packet processing unit 23, a transmit buffer 24, an ingress side discard counter 25, an FCS error counter 26, and an egress side discard counter 27. The network device 2 may include other functions or circuits not shown in Fig. 2. In the example shown in Fig. 2, the network device 2 includes one input port and one output port, but may include multiple input ports and multiple output ports.

[0015] A packet arriving at the network device 2 via the communication network 1 is written to the receive buffer 21. At this time, the header of the received packet is checked. For example, the destination address and source address set in the header of the received packet are checked. If the destination address and source address of the received packet match, this means that the packet sent by the network device 2 has returned to that network device 2. In other words, it is determined that a loop error has occurred.

[0016] An FCS (Frame Check Sequence) processing unit 22 detects an FCS error by using the frame check sequence of the received packet.

[0017] The packet processing unit 23 processes the received packet based on the header or overhead of the received packet. Then, the packet to be transmitted to another network device 2 is written to the transmission buffer 24. The packets written to the transmission buffer 24 are output to the communication network 1 in order.

[0018] The ingress-side discard counter 25 counts the number of packets discarded in the receive buffer 21. The received packets written to the receive buffer 21 are read at a predetermined rate and processed by the packet processing unit 23. Therefore, if the packet reception rate exceeds a predetermined threshold, an overflow occurs in the receive buffer 21, and some of the received packets are discarded. In addition, if the above-mentioned loop error is detected, the packets are also discarded in the receive buffer 21.

[0019] The FCS error counter 26 counts the number of FCS errors detected by the FCS processing unit 22. In this embodiment, it is assumed that received packets in which an FCS error is detected are discarded. In this case, the FCS error counter 26 counts the number of packets discarded due to an FCS error. The egress-side discard counter 27 counts the number of packets discarded in the transmission buffer 24. Packets written to the transmission buffer 24 are read at a predetermined rate and output to the communication network 1. Therefore, when a larger number of packets than expected are transmitted, an overflow occurs in the transmission buffer 24, and some of the transmitted packets are discarded.

[0020] In this way, the network device 2 detects or measures the communication status of the network device 2 using multiple counters (25, 26, 27). The count values ​​of each counter are collected by the silent failure detection device 10 as performance monitor information. As an example, upon receiving a polling signal from the silent failure detection device 10, the network device 2 transmits the count values ​​of each counter to the silent failure detection device 10. The count values ​​of the above-mentioned three counters are examples of performance monitor information. The performance monitor information may also include other parameters related to the communication status of the network device 2.

[0021] Fig. 3 shows an example of the functional configuration of the silent fault detection device 10. The silent fault detection device 10 includes a PM information acquisition unit 11, a difference calculation unit 12, a silent fault detection unit 13, and a detection result output unit 14. Note that the silent fault detection device 10 may further include other functions not shown in Fig. 3.

[0022] The PM information acquisition unit 11 acquires performance monitor information from each network device 2. Specifically, the PM information acquisition unit 11 periodically transmits a polling signal to the network device 2. In response to this, the network device 2 that receives the polling signal transmits the count values ​​of each counter (25, 26, 27) as performance monitor information to the silent failure detection device 10. As a result, the PM information acquisition unit 11 periodically acquires the count values ​​of each network device 2.

[0023] The interval at which the PM information acquisition unit 11 acquires the performance monitor information is not particularly limited, but may be, for example, 15 minutes. Also, when the number of packets discarded in each network device 2 is small, the PM information acquisition unit 11 may acquire the performance monitor information at long intervals (for example, one hour).

[0024] In the following description, the count value by the ingress side discard counter 25 may be referred to as the "ingress side discard count value." The count value by the FCS error counter 26 may be referred to as the "FCS error count value." The count value by the egress side discard counter 27 may be referred to as the "egress side discard count value."

[0025] The difference calculation unit 12 calculates the difference between the count value acquired at the immediately preceding sampling time and the newly acquired count value for the performance monitor information periodically acquired by the PM information acquisition unit 11. That is, counter difference values ​​are calculated for the ingress discard count value, the FCS error count value, and the egress discard count value. In other words, the difference calculation unit 12 detects changes in the performance monitor information (the ingress discard count value, the FCS error count value, and the egress discard count value).

[0026] Fig. 4 shows an example of a counter difference value calculated by the difference calculation unit 12. In this embodiment, the PM information acquisition unit 11 acquires performance monitor information from each network device 2 at 15-minute intervals. Fig. 4 also shows a counter difference value for one network device 2.

[0027] In the case shown in FIG. 4, for example, the difference in the Ingress discard count value during the sampling period "2024 / 05 / 30 / 00:30-00:45" is "35." This counter difference value represents the difference between the Ingress discard count value acquired at 0:30 on May 30, 2024, and the Ingress discard count value acquired at 0:45 on May 30, 2024. In other words, this counter difference value represents that the number of discarded packets counted by the Ingress discard counter 25 during the 15 minutes from 0:30 to 0:45 on May 30, 2024, is 35. Therefore, this counter difference value represents that 35 received packets were discarded in the receive buffer 21 during this sampling period.

[0028] Furthermore, the difference in the FCS error count value during the sampling period "2024 / 05 / 30 / 00:30-00:45" is "5." This counter difference value indicates that the number of FCS errors counted by FCS error counter 26 during the 15 minutes from 0:30 to 0:45 on May 30, 2024 is five. In other words, this counter difference value indicates that five received packets were discarded by FCS processing unit 22 during this sampling period due to FCS errors.

[0029] The counter difference value calculated by the difference calculation unit 12 is notified to the silent failure detection unit 13. At this time, the difference calculation unit 12 may notify the silent failure detection unit 13 of all the calculated counter difference values. However, when each network device 2 in the communication network 1 is operating normally, it is expected that the number of detected errors will be small. Also, when the transmission rate of each network device 2 is lower than a threshold level, it is expected that the number of discarded packets will be small. In other words, during normal operation, it is expected that each counter difference value will be zero. Therefore, the difference calculation unit 12 may notify the silent failure detection unit 13 of the calculated counter difference value only when the calculated counter difference value is not zero. With this configuration, it is possible to reduce the memory capacity for storing counter difference values.

[0030] The silent failure detection unit 13 detects a silent failure that occurs in the communication network 1 based on the counter difference value notified by the difference calculation unit 12. Here, the counter difference value is calculated based on performance monitor information detected in each network device 2. Therefore, the silent failure detection unit 13 can detect a silent failure that occurs in each network device 2. Alternatively, the silent failure detection unit 13 can determine whether or not a silent failure has occurred in each network device 2.

[0031] The detection result output unit 14 outputs the detection result by the silent failure detection unit 13. That is, when the silent failure detection unit 13 detects a network device 2 in which a silent failure is occurring, the detection result output unit 14 outputs a notification indicating that a silent failure has occurred in that network device 2. This notification is displayed, for example, on a computer of the administrator of the communication network 1.

[0032] Note that the silent failure detection unit 13 actually detects a network device 2 suspected of having a silent failure. That is, the above-mentioned notification indicates a network device 2 suspected of having a silent failure. Therefore, in the following description, a notification generated when a network device 2 suspected of having a silent failure is detected may be referred to as a "suspected failure notification." Then, when a silent failure is detected in the communication network 1, the administrator of the communication network 1 will prioritize investigating the network device 2 indicated by the suspected failure notification. This can shorten the time required to recover from a failure.

[0033] For example, when a network device 2 experiences an increase in discarded packets received from an opposite device, or an increase in FCS errors in packets received from the opposite device, it is assumed that the problem is a software error in the opposite device. Alternatively, it is assumed that there is a problem in the optical fiber between the opposite device and the network device 2. Therefore, when the ingress discard count value or FCS error count value of a certain network device 2 increases, it is effective to check the opposite device or the optical fiber between the opposite device and the network device 2.

[0034] Furthermore, when the number of discarded packets to be transmitted or forwarded increases in a certain network device 2, a software error is suspected in that network device 2. Therefore, when the egress discard count value of a certain network device 2 increases, it is effective to investigate that network device 2.

[0035] In this manner, in the embodiment of the present invention, the silent failure detection device 10 implemented in the network management system 3 collects performance monitor information from each network device 2, allowing the administrator of the communication network 1 to isolate the location where the silent failure is occurring, thereby shortening the time required to recover from the failure.

[0036] 5 is a flowchart showing an example of a process for detecting a silent failure. The PM information acquisition unit 11 acquires performance monitor information from the network device 2 at a predetermined cycle (for example, every 15 minutes). The difference calculation unit 12 calculates difference information for each counter every time the PM information acquisition unit 11 acquires performance monitor information, and notifies the silent failure detection unit 13 of the calculated difference information. The difference information corresponds to a change in the count value (i.e., a counter difference value indicating the difference between the immediately preceding count value and a new count value). The process of this flowchart is repeatedly executed at a predetermined cycle (for example, every 15 minutes). As an example, the process of this flowchart is executed in synchronization with the timing at which the PM information acquisition unit 11 acquires performance monitor information. The process of this flowchart is executed for each network device 2.

[0037] In S1, the silent failure detection unit 13 checks whether or not difference information has been notified from the difference calculation unit 12. In this embodiment, the difference calculation unit 12 outputs difference information when the counter difference value is not zero (i.e., when the counter value has changed in the network device 2). When difference information has been notified from the difference calculation unit 12, the processing of the silent failure detection unit 13 proceeds to S2. On the other hand, when difference information has not been notified from the difference calculation unit 12, the processing of the silent failure detection unit 13 proceeds to S20, which will be described later.

[0038] In S2, the silent failure detection unit 13 determines whether the port / link of the network device 2 is normal. The method for determining whether the port / link of the network device 2 is normal is not particularly limited, and any known method can be used. For example, whether the port / link is normal may be determined by acquiring port information or alarm information in response to a polling signal for acquiring performance monitor information. Alternatively, whether the port / link is normal may be determined by acquiring port information or alarm information from the network management system 3. If the port / link is normal, the processing of the silent failure detection unit 13 proceeds to S3. On the other hand, if the port / link is not normal, it is clear that a fault has occurred in the network device 2, and the silent failure detection unit 13 terminates the processing.

[0039] In S3, the silent failure detection unit 13 stores the difference information notified by the difference calculation unit 12 in a difference information management table. As shown in FIG. 6, the difference information management table manages a predetermined number of the latest difference values ​​for each count item (ingress discard count value, FCS error count value, egress discard count value). In this embodiment, the difference values ​​are stored in association with the sampling period when the difference was not zero. For example, for the ingress discard count value, difference values ​​are recorded for each of five consecutive sampling periods between 11:00 and 12:15 on May 31, 2024. This information represents a state in which received packets are continuously discarded in the receive buffer 21 of the network device 2.

[0040] In S4, the silent failure detection unit 13 checks whether a suspected failure notification has been output. A suspected failure notification is output when it is determined in the processing of this flowchart that a silent failure is suspected to have occurred. If a suspected failure notification has not been output, the processing of the silent failure detection unit 13 proceeds to S5. On the other hand, if a suspected failure notification has already been output, there is no need to execute processing to detect a silent failure, and so the processing of the silent failure detection unit 13 ends.

[0041] In S5, the silent failure detection unit 13 refers to the difference information management table and determines whether difference information is being generated continuously. Here, difference information is generated when the values ​​of the counters (25 to 27) of the network device 2 change. These counters (25 to 27) are counted up when a packet is discarded in the network device 2. Therefore, in S5, it is determined whether packet discarding is continuing in the network device 2. Here, "continuously" means not sudden. For example, if difference information is recorded in the difference information management table at a frequency of once a day or more, the silent failure detection unit 13 may determine that difference information is being generated continuously. If difference information is being generated continuously, the processing of the silent failure detection unit 13 proceeds to S6. On the other hand, if difference information is not being generated continuously, it is considered that the difference information was generated due to a sudden cause other than a silent failure, and the processing of the silent failure detection unit 13 ends. For example, if the time difference between the date and time of the sampling period of a latest information record and the date and time of the sampling period of the latest information record immediately before it is one week or more in the differential information management table, it is determined that differential information is not being generated continuously.

[0042] In S6, the silent failure detection unit 13 refers to the differential information management table and detects the frequency at which differential information is generated. The frequency may be calculated based on the time difference between the date and time of the sampling period of the oldest information record and the date and time of the sampling period of the latest information record in the differential information management table. In the case shown in FIG. 6, differential information is recorded at 15-minute intervals for the ingress-side discard count value. That is, packet discards occur frequently in the receive buffer 21. In contrast, differential information is recorded at approximately 1-hour intervals for the FCS error count value. That is, packet discards due to FCS errors occur rarely.

[0043] In S7, the silent failure detection unit 13 refers to the difference information management table and detects a difference value. The difference value represents the number of packets discarded within one sampling period. The difference value may also be an average value. In the case shown in FIG. 6, the difference value of the ingress discard count value is calculated by averaging the difference values ​​of the five records to obtain "30." For the FCS error count value, the average difference value obtained is "2."

[0044] In S8, the silent failure detection unit 13 determines the weights of the failure determination parameters. In this embodiment, a discard frequency parameter, a discard number parameter, and a simultaneous occurrence parameter are used as the failure determination parameters.

[0045] The discard frequency parameter is specified in S6. In this embodiment, the weight of the discard frequency parameter is between 5 and 10. Specifically, when the discard frequency is high, the weight is 10, and when the discard frequency is low, the weight is 5. For example, when the interval at which packet discards occur is shorter than 20 minutes, the weight of the discard frequency parameter is 10, and when the interval at which packet discards occur is longer than 1 hour, the weight of the discard frequency parameter is 5.

[0046] The discard number parameter is calculated in S7. In this embodiment, the weight of the discard number parameter is between 1 and 10. Specifically, when the number of discarded packets is large, the weight is 10, and when the number of discarded packets is small, the weight is 1. For example, when the average number of discarded packets within a sampling period exceeds 100, the weight of the discard number parameter is 10, and when the average number of discarded packets within a sampling period is less than 5, the weight of the discard number parameter is 1.

[0047] In this embodiment, the weight of the concurrency parameter is 1, 5, or 10. Specifically, when packet loss is detected by any one of the three counters (25 to 27) shown in FIG. 2, the weight is 1. When packet loss is detected simultaneously by any two of the counters (25 to 27), the weight is 5. When packet loss is detected simultaneously by all of the counters (25 to 27), the weight is 10.

[0048] In S9, the silent failure detection unit 13 calculates a failure score based on the weight of each failure determination parameter determined in S8. In this embodiment, the failure score is calculated by adding up the weights of the three failure determination parameters described above.

[0049] In S10 to S11, the silent failure detection unit 13 compares the failure score calculated in S9 with a predetermined threshold. In this embodiment, the threshold is 10. If the failure score is greater than the threshold, the silent failure detection unit 13 determines that a silent failure may have occurred and generates a suspected failure notification. The suspected failure notification includes information identifying the network device 2 in which a failure score greater than the threshold has been detected. The detection result output unit 14 then outputs the suspected failure notification. The suspected failure notification is displayed, for example, on the computer of the administrator of the communication network 1. After this, in S12, the silent failure detection unit 13 resets a release counter. The release counter will be described later.

[0050] When the difference information is not notified from the difference calculation unit 12 (S1: No), the silent failure detection unit 13 executes a cancellation process in S20.

[0051] Fig. 7 is a flowchart showing an example of the cancellation process. The cancellation process corresponds to S20 in the flowchart shown in Fig. 5. That is, the cancellation process is executed when difference information is not notified from the difference calculation unit 12.

[0052] After outputting the failure suspicion notification, the silent failure detection unit 13 monitors whether or not further packet discarding occurs in the network device 2. At this time, if a period of time continues during which no packet discarding is detected, the count value of the release counter is counted up. Then, when the count value of the release counter becomes larger than a predetermined threshold, the silent failure detection unit 13 determines that packet discarding related to the silent failure has stopped occurring, and outputs a failure suspicion cancellation notification. Specifically, this is as follows.

[0053] In S21, the silent failure detection unit 13 checks whether a suspected failure notification has been output. As described above, a suspected failure notification is output when it is determined in the processing of the flowchart shown in FIG. 5 that a silent failure is suspected to have occurred. If a suspected failure notification has been output, the silent failure detection unit 13 counts up the release counter in S22. Note that the release counter is reset in S12 of the flowchart shown in FIG. 5, as described above. In other words, the release counter is reset to zero when the suspected failure notification is output.

[0054] In S23, the silent failure detection unit 13 compares the count value of the release counter with a predetermined threshold. If the count value of the release counter is greater than the predetermined threshold, the silent failure detection unit 13 outputs a failure suspicion release notice in S24. Note that the failure suspicion release notice is displayed, for example, on the computer of the administrator of the communication network 1, in the same way as the failure suspicion notice.

[0055] In this way, the silent failure detection unit 13 determines whether a silent failure has occurred in the network device 2 or in the vicinity of the network device 2, based on packet discards that occur in the network device 2. At this time, the fault determination parameters are weighted based on the location where packet discards occur within the network device 2, the frequency of packet discards, and the number of packet discards, and whether a silent failure has occurred is determined based on the sum of the weighted fault determination parameters. Therefore, by appropriately setting the weights of each fault determination parameter, the accuracy of determining whether a fault is suspected is improved. Furthermore, if no packet discards occur for a predetermined period of time, the suspected fault notification is canceled, making it possible to check the situation in real time and create a history.

[0056] Next, various use cases that may occur in the network device 2 are applied to the procedure of the flowchart shown in Figure 5. As a result, for each use case, it is determined whether or not there is a suspicion of a silent failure. The use cases considered below are shown in Figure 8.

[0057] <Case 1> In case 1, the optical fiber connected to the receiving port of the network device 2 is deteriorated. Alternatively, the optical fiber connector is not properly inserted into the receiving port of the network device 2. This causes a decrease in the quality of the received optical signal, and an FCS error may be detected in the network device 2. In this case, the FCS error counter 26 counts the number of packets discarded due to an FCS error.

[0058] When a packet is discarded in the network device 2 due to an FCS error, difference information corresponding to the FCS error counter 26 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." Here, it is assumed that the port / link of the network device 2 is normal (S2: Yes). It is also assumed that a suspected failure notification has not yet been output (S4: Yes). Furthermore, if the optical fiber is deteriorated or the optical fiber connector is not properly inserted, the quality of the received optical signal will remain low and FCS errors will continue to occur, so the determination in S5 is "Yes."

[0059] In addition to case 1, in other cases described below (i.e., cases 2, 3, 4a to 4e, 5a, 5b, and 6), the ports / links of network device 2 are normal (S2: Yes), and a suspected failure notification has not yet been output (S4: Yes).

[0060] Since FCS errors occur continuously, the frequency of packet discards increases. Therefore, the weight of the discard frequency parameter is "10." The number of discarded packets depends on the communication volume. When the communication volume is low, the weight of the discard number parameter is "1," and when the communication volume is high, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter is "1 to 10" depending on the communication volume. Note that in this embodiment, it is assumed that no packet discards occur in the receive buffer 21 and no packet discards occur in the transmit buffer 24. In other words, the weight of the concurrency parameter is "1."

[0061] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "12" when the communication volume is low and "21" when the communication volume is high. In other words, the fault determination score is "12 to 21." Here, the threshold value used in S10 is "10." Therefore, Case 1 is determined to be a suspected silent failure.

[0062] The silent failure detection device 10 outputs a failure suspicion notification. This failure suspicion notification includes information identifying the network device 2 whose failure determination score has exceeded the threshold, and information indicating that an FCS error has occurred.

[0063] <Case 2> In case 2, the optical fiber connection is incorrect, and a packet sent from network device 2 returns to that network device 2. In this case, the destination address of the received packet matches the address of the network device 2, so a loop error is detected and the received packet is discarded in the receive buffer 21. At this time, the ingress side discard counter 25 counts the number of packets discarded due to the loop error.

[0064] When a packet is discarded in the network device 2 due to a loop error, difference information corresponding to the ingress side discard counter 25 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." Furthermore, since the loop error continues to occur until the misconnection of the optical fiber is resolved, the determination in S5 is "Yes."

[0065] Since loop errors occur continuously, the frequency of packet discards increases. Therefore, the weight of the discard frequency parameter is "10." The number of discarded packets depends on the communication volume. When the communication volume is low, the weight of the discard number parameter is "1," and when the communication volume is high, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter ranges from "1 to 10" depending on the communication volume. In this embodiment, it is assumed that packet discards due to FCS errors and packet discards in the transmit buffer 24 do not occur. In other words, the weight of the simultaneous occurrence parameter is "1."

[0066] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "12" when the communication volume is low and "21" when the communication volume is high. In other words, the fault determination score is "12 to 21." Here, the threshold value used in S10 is "10." Therefore, Case 1 is determined to be a suspected silent failure.

[0067] The silent failure detection device 10 outputs a failure suspicion notification. This failure suspicion notification includes information for identifying the network device 2 whose failure determination score has exceeded the threshold, and information indicating that a packet discard has occurred in the receive buffer 21.

[0068] <Case 3> In Case 3, cosmic rays hitting the Earth cause an electronic circuit in network device 2 to malfunction, causing a specific bit in a packet processed by that electronic circuit to be fixed to 0 or 1. Depending on which bit is fixed and to what value it is fixed, an FCS error may occur in network device 2 receiving this packet. In this case, FCS error counter 26 counts the number of packets discarded due to an FCS error.

[0069] When a packet is discarded in the network device 2 due to an FCS error, difference information corresponding to the FCS error counter 26 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." In addition, in case 2, FCS errors will continue to occur until the failed electronic circuit is replaced, so the determination in S5 is "Yes."

[0070] However, even if a specific bit in a packet is fixed to a specific value due to a fault in an electronic circuit, an error may not occur. For example, if a bit whose original value is "1" is fixed to "1" due to a fault, no error occurs. Therefore, the frequency and number of discarded packets depend on the position of the bit affected by the fault.

[0071] Therefore, the weight of the discard frequency parameter is assumed to be "5 to 10." The weight of the discard number parameter is assumed to be "1 to 4." In this embodiment, it is assumed that packet discards do not occur in the receive buffer 21 and the transmit buffer 24. In other words, the weight of the concurrency parameter is "1."

[0072] In this case, the fault determination score, which represents the sum of the weights of the three discard determination parameters, is between 7 and 15. Here, the threshold value used in S10 is 10. Therefore, Case 3 may be determined to be a suspected silent fault, depending on the position of the bit affected by the fault.

[0073] When the fault determination score exceeds 10, the silent fault detection device 10 outputs a fault suspicion notification. This fault suspicion notification includes information identifying the network device 2 whose fault determination score exceeds the threshold, and information indicating that an FCS error has been detected.

[0074] <Case 4a> In case 4a, instead of a failure, the reception rate of the network device 2 temporarily exceeds the threshold. That is, burst reception occurs. In this case, the reception buffer 21 of the network device 2 may overflow. That is, packets are discarded in the reception buffer 21, and the ingress-side discard counter 25 counts the number of discarded packets.

[0075] When a packet is discarded in the receiving buffer 21 of the network device 2, difference information corresponding to the ingress side discard counter 25 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes".

[0076] However, if the transmission rate of the opposite device decreases and the reception rate at the network device 2 falls below the threshold, packets will no longer be discarded in the reception buffer 21. In other words, the situation in which received packets are discarded will not continue. In this case, the determination in S5 is "No," and the fault determination score is not calculated. Therefore, in case 4a, it is determined that a silent failure has not occurred.

[0077] <Case 4b> In case 4b, unlike case 4a, burst reception occurs repeatedly in the network device 2. That is, every time burst reception occurs, packets are discarded in the reception buffer 21.

[0078] When a packet is discarded in the receive buffer 21 of the network device 2, difference information corresponding to the ingress-side discard counter 25 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." In addition, because burst reception occurs repeatedly, the determination in S5 is also "Yes."

[0079] The weight of the discard frequency parameter is assumed to be "5 to 10." The number of discarded packets depends on the data volume of each burst communication. When the data volume of the burst communication is small, the weight of the discard number parameter is "1," and when the data volume of the burst communication is large, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter is "1 to 10" depending on the data volume of the burst communication. Note that in this embodiment, it is assumed that packet discard due to FCS errors and packet discard in the transmit buffer 24 do not occur. In other words, the weight of the concurrency parameter is "1."

[0080] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "7" when the discard frequency is low and the communication volume is small, and is "21" when the discard frequency is high and the communication volume is large. In other words, the fault determination score is "7 to 21." Here, the threshold value used in S10 is "10." Therefore, case 4b may be determined to be a suspected silent failure depending on the frequency of burst communication and the data volume of each burst communication.

[0081] When the fault determination score exceeds 10, the silent fault detection device 10 outputs a fault suspicion notification. This fault suspicion notification includes information identifying the network device 2 whose fault determination score exceeds the threshold, and information indicating that a packet discard has occurred in the receive buffer 21.

[0082] <Case 4c> In case 4c, instead of a failure, the transmission rate of the network device 2 temporarily exceeds the threshold. This causes temporary congestion within the network device 2. In this case, the transmission buffer 24 may overflow. That is, packets are discarded in the transmission buffer 24, and the egress-side discard counter 27 counts the number of discarded packets.

[0083] When a packet is discarded in the transmission buffer 24 of the network device 2, difference information corresponding to the Egress side discard counter 27 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes".

[0084] However, when the transmission rate described above falls below the threshold, packets are no longer discarded in the transmission buffer 24. In other words, the situation in which transmission packets are discarded does not continue. In this case, the determination in S5 is "No," and the fault determination score is not calculated. Therefore, in case 4c, it is determined that a silent failure has not occurred.

[0085] <Case 4d> In case 4d, unlike case 4c, a situation occurs repeatedly in which the transmission rate temporarily exceeds the threshold in the network device 2. That is, packets are discarded in the transmission buffer 24 every time a burst transmission occurs.

[0086] When a packet is discarded in the transmission buffer 24 of the network device 2, difference information corresponding to the Egress side discard counter 27 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." In addition, because burst transmission occurs repeatedly, the determination in S5 is also "Yes."

[0087] The weight of the discard frequency parameter is assumed to be "5 to 10." The discard number parameter depends on the data volume of each burst communication. When the data volume of the burst communication is small, the weight of the discard number parameter is "1," and when the data volume of the burst communication is large, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter is "1 to 10" depending on the data volume of the burst communication. Note that in this embodiment, it is assumed that packet discards in the receive buffer 21 and packet discards due to FCS errors do not occur. In other words, the weight of the concurrency parameter is "1."

[0088] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "7" when the discard frequency is low and the communication volume is small, and is "21" when the discard frequency is high and the communication volume is large. In other words, the fault determination score is "7 to 21." Here, the threshold value used in S10 is "10." Therefore, case 4d may be determined to be a suspected silent failure depending on the frequency of burst communication and the data volume of each burst communication.

[0089] When the fault determination score exceeds 10, the silent fault detection device 10 outputs a fault suspicion notification. This fault suspicion notification includes information identifying the network device 2 whose fault determination score exceeds the threshold, and information indicating that a packet discard has occurred in the transmission buffer 24.

[0090] <Case 4e> In this embodiment, the network device 2 includes a table in which path information for forwarding a received packet to a destination is set. In this case, the network device 2 refers to the table using the path information (e.g., a VLAN ID that identifies a virtual LAN) set in the header of the received packet, and forwards the packet to the destination node.

[0091] In case 4e, incorrect path information is set in the header of the packet sent from the opposite device. In this case, the network device 2 cannot forward the received packet, and the received packet is discarded in the receive buffer 21.

[0092] When a packet is discarded in the receive buffer 21 of the network device 2, difference information corresponding to the ingress side discard counter 25 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." Furthermore, since the incorrect path information continues until the settings of the opposing device are corrected, the determination in S5 is also "Yes."

[0093] The weight of the discard frequency parameter is assumed to be "5 to 10." The number of discarded packets depends on the communication volume. When the communication volume is low, the weight of the discard number parameter is "1," and when the communication volume is high, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter is "1 to 10" depending on the communication volume. In this embodiment, it is assumed that packet discards due to FCS errors and packet discards in the transmission buffer 24 do not occur. In other words, the weight of the concurrency parameter is "1."

[0094] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "7" when the discard frequency is low and the communication volume is small, and is "21" when the discard frequency is high and the communication volume is large. In other words, the fault determination score is "7 to 21." Here, the threshold value used in S10 is "10." Therefore, case 4e may be determined to be a suspected silent failure depending on the discard frequency and communication volume.

[0095] When the fault determination score exceeds 10, the silent fault detection device 10 outputs a fault suspicion notification. This fault suspicion notification includes information identifying the network device 2 whose fault determination score exceeds the threshold, and information indicating that a packet discard has occurred in the receive buffer 21.

[0096] <Case 5a> In case 5a, the forwarding destination of the received packet is not registered in the network device 2. For example, case 5a can occur when the header of the received packet is erroneously rewritten due to a bug in software implemented in the network device 2. If the forwarding destination of the received packet is not registered in the network device 2, the received packet is discarded in the receive buffer 21 or the transmit buffer 24.

[0097] When a packet is discarded in the receive buffer 21 or the transmit buffer 24 of the network device 2, difference information corresponding to the ingress side discard counter 25 or the egress side discard counter 27 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." Furthermore, since the state in which the forwarding destination is not registered will continue until the software is updated, the determination in S5 is also "Yes."

[0098] The weight of the discard frequency parameter is assumed to be "5 to 10." The discard number parameter depends on the communication volume. When the communication volume is low, the weight of the discard number parameter is "1," and when the communication volume is high, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter is "1 to 10" depending on the communication volume. In this embodiment, it is assumed that packet discard occurs in only one of the receive buffer 21 or the transmit buffer 24. In other words, the weight of the concurrency parameter is "1."

[0099] Then, the fault determination score, which represents the total weight of the three discard determination parameters, is "7" when the discard frequency is low and the communication volume is small, and is "21" when the discard frequency is high and the communication volume is large. In other words, the fault determination score is "7 to 21." Here, the threshold value used in S10 is "10." Therefore, case 5a may be determined to be a suspected silent failure depending on the discard frequency and communication volume.

[0100] When the fault determination score exceeds 10, the silent fault detection device 10 outputs a fault suspicion notification. This fault suspicion notification includes information identifying the network device 2 whose fault determination score exceeds the threshold, and information indicating that a packet discard has occurred in the receive buffer 21 or the transmit buffer 24.

[0101] <Case 5b> In case 5b, packet flooding is performed when the network device 2 receives an unknown packet. That is, when the destination of the received packet is not registered in the packet forwarding table, multicast / broadcast forwarding is performed. However, multicast / broadcast forwarding generates a large number of packets to be transmitted, which makes the transmission buffer 24 prone to overflow.

[0102] When a packet is discarded in the transmission buffer 24 of the network device 2, difference information corresponding to the Egress side discard counter 27 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." Furthermore, since the multicast / broadcast transfer described above can be repeatedly executed until the cause of the generation of the unknown packet is eliminated, the determination in S5 is also "Yes."

[0103] The weight of the discard frequency parameter is assumed to be "5 to 10." If the number of discarded packets is not assumed to be large, the weight of the discard count parameter is assumed to be "1 to 4." In this embodiment, it is assumed that packet discards in the receive buffer 21 and packet discards due to FCS errors do not occur. In other words, the weight of the concurrency parameter is "1."

[0104] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "7 to 15." Here, the threshold value used in S10 is "10." Therefore, case 5b may be determined to be a suspected silent fault.

[0105] When the fault determination score exceeds 10, the silent fault detection device 10 outputs a fault suspicion notification. This fault suspicion notification includes information identifying the network device 2 whose fault determination score exceeds the threshold, and information indicating that a packet discard has occurred in the transmission buffer 24.

[0106] <Case 6> In case 6, packets are discarded at multiple locations within the network device 2. For example, some of the received packets are discarded due to overflow of the receive buffer 21, and further, some of the packets read from the receive buffer 21 are discarded due to an FCS error.

[0107] When a packet is discarded in the receive buffer 21 of the network device 2, difference information corresponding to the ingress-side discard counter 25 is generated and provided to the silent failure detection unit 13. In addition, when a packet is discarded due to an FCS error, difference information corresponding to the FCS error counter 26 is generated and provided to the silent failure detection unit 13. Therefore, the determination in S1 is "Yes." It is also assumed that the determination in S5 is "Yes."

[0108] The weight of the discard frequency parameter is assumed to be "5 to 10." The number of discarded packets depends on the communication volume. When the communication volume is low, the weight of the discard number parameter is "1," and when the communication volume is high, the weight of the discard number parameter is "10." In other words, the weight of the discard number parameter is "1 to 10" depending on the communication volume. Furthermore, in case 6, packet discards occur at two locations within the network device 2, so the weight of the concurrency parameter is "5."

[0109] In this case, the fault determination score, which represents the total weight of the three discard determination parameters, is "11 to 25." Here, the threshold value used in S10 is "10." Therefore, case 6 is determined to be a suspected silent fault.

[0110] The silent failure detection device 10 outputs a failure suspicion notification. This failure suspicion notification includes information for identifying the network device 2 whose failure determination score has exceeded the threshold, and information indicating the location where packet discarding has occurred.

[0111] In this way, according to the embodiment of the present invention, a network device 2 suspected of having a silent failure is identified. Therefore, the administrator of the communication network 1 can use the network management system 3 to isolate the location where the silent failure is occurring. For example, when the failure determination score of the network device 2X exceeds a threshold, it is determined that a silent failure has occurred in the network device 2X or between the network device 2X and its opposing device. In addition, the embodiment of the present invention has the following effects.

[0112] The silent failure detection device 10 monitors the state of the network device 2 taking into consideration the continuity of packet discards, the simultaneous occurrence of packet discards, and the number of discarded packets, and therefore has a high degree of accuracy in detecting silent failures. For example, it can detect cases where packet discards continue even though the number of discarded packets is small.

[0113] By monitoring packet discards due to FCS errors, it is possible to detect deterioration of optical fiber and improper insertion of optical fiber connectors.

[0114] It is possible to detect silent failures based on the traffic flow rate in each network device 2. However, in this case, if the paths between the network devices 2 become complex, the processing load for estimating the traffic volume increases, and the size of the software program required for this increases. In contrast, the silent failure detection device 10 detects silent failures based on the value of the discard counter in each network device 2, so the size of the software program required for this is small, and the amount of processing is also small. Furthermore, the cost of resources (memory and CPU) required to detect silent failures is low.

[0115] When detecting silent failures based on the traffic flow rate in each network device 2, an estimated value of the traffic flow rate is used, which may result in low accuracy in determining silent failures. In contrast, the silent failure detection device 10 detects silent failures based on the number of packets actually discarded in the network device 2. This results in high accuracy in determining silent failures.

[0116] In the above example, the fault determination score is calculated from three parameters (discard frequency parameter, discard number parameter, and concurrent occurrence parameter), but the embodiment of the present invention is not limited to this method. For example, the fault determination score may be calculated from any two of the discard frequency parameter, discard number parameter, and concurrent occurrence parameter. Alternatively, the fault determination score may be calculated from four or more parameters.

[0117] <Hardware configuration> 9 shows an example of the hardware configuration of the silent failure detection device 10 (or the network management system 3). The silent failure detection device 10 is realized by a computer system 100 including a processor 101, a memory 102, a storage device 103, an input / output device 104, a recording medium reader 105, and a communication interface 106.

[0118] The processor 101 executes a silent failure detection program stored in the storage device 103. When the processor 101 executes the silent failure detection program, the functions of the PM information acquisition unit 11, difference calculation unit 12, silent failure detection unit 13, and detection result output unit 14 shown in FIG. 3 are provided. The memory 102 is used as a work area for the processor 101. The storage device 103 stores the silent failure detection program and other programs. The difference information management table shown in FIG. 6 is stored in the memory 102 or the storage device 103.

[0119] The input / output device 104 may include input devices such as a keyboard, a mouse, a touch panel, and a microphone. The input / output device 104 may also include output devices such as a display device and a speaker. The recording medium reader 105 can acquire data and information recorded on the recording medium 110. The recording medium 110 is a removable recording medium that can be attached to and detached from the computer system 100. The recording medium 110 may be realized, for example, by a semiconductor memory, a medium that records signals optically, or a medium that records signals magnetically. The silent fault detection program may be provided to the computer system 100 from the recording medium 110. The communication interface 106 provides a function for connecting to a network. When the silent fault detection program is stored in the program server 120, the computer system 100 may acquire the silent fault detection program from the program server 120. [Explanation of symbols]

[0120] 1. Communication Network 2. Network Devices 3 Network Management Systems 10 Silent fault detection device 11 PM information acquisition department 12 Difference calculation part 13 Silent fault detection unit 14 Detection result output section 21 Receive Buffer 24 send buffers 25 Ingress discard counter 26 FCS error counter 27 Egress discard counter

Claims

1. an acquisition unit that periodically acquires performance monitor information representing a communication state from each of a plurality of network devices that constitute a communication network; a detection unit that detects a silent failure occurring in the communication network based on the performance monitor information acquired by the acquisition unit, The detection unit determining values ​​of a plurality of fault determination parameters based on the performance monitor information; A fault determination score is calculated from the values ​​of the plurality of fault determination parameters, and it is determined whether or not a silent fault has occurred in the communication network. A silent fault detection device characterized by:

2. The performance monitor information acquired by the acquisition unit from a first network device of the plurality of network devices includes first counter information representing the number of packets discarded in a receive buffer of the first network device, second counter information representing the number of packets discarded due to an error in the first network device, and third counter information representing the number of packets discarded in a transmit buffer of the first network device.

2. The silent fault detection device according to claim 1.

3. the performance monitor information acquired by the acquisition unit from a first network device among the plurality of network devices includes counter information indicating the number of packets discarded in the first network device; the plurality of failure determination parameters include a first parameter representing a frequency at which packets are discarded in the first network device, and a second parameter representing the number of packets discarded in the first network device; The detection unit the value of the first parameter is increased as the frequency of packets being discarded in the first network device increases; The greater the number of packets discarded in the first network device, the greater the value of the second parameter. calculating the disability determination score by adding the value of the first parameter and the value of the second parameter; When the fault determination score is greater than a predetermined threshold, it is determined that a silent fault has occurred in the first network device or between the first network device and an opposing device of the first network device.

2. The silent fault detection device according to claim 1.

4. the plurality of fault determination parameters further include a third parameter representing the number of locations where packet discarding occurs in the first network device; The detection unit the value of the third parameter is increased as the number of locations where packet discarding occurs in the first network device increases; calculating the disability determination score by adding up the value of the first parameter, the value of the second parameter, and the value of the third parameter; When the fault determination score is greater than a predetermined threshold, it is determined that a silent fault has occurred in the first network device or between the first network device and an opposing device of the first network device.

4. The silent fault detection device according to claim 3.

5. a difference calculation unit that calculates a difference between a number indicated by the counter information most recently acquired by the acquisition unit and a number indicated by the counter information newly acquired by the acquisition unit; The detection unit determines whether a silent failure has occurred in the communication network when the difference is not zero.

4. The silent fault detection device according to claim 3.

6. periodically acquiring performance monitor information representing a communication state from each of a plurality of network devices constituting the communication network; determining values ​​of a plurality of fault determination parameters based on the acquired performance monitor information; A fault determination score is calculated from the values ​​of the plurality of fault determination parameters, and it is determined whether or not a silent fault has occurred in the communication network. A silent fault detection program that causes a computer to execute a process.

Images