JP2000101583A - Network monitor support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect and analyze a transmission fault on a network in real time at a physical layer level during a network operation. SOLUTION: Near the respective terminating resistors 10a and 10b of a bus 10, fault monitor support devices 11a and 11b are connected. At the time of detecting a packet transmitted through the bus 10, one fault monitor support device 11b transmits position detection pulses for detecting the packet origin position to the other fault monitor support device 11a. The other fault monitor support device 11a fetches a bit string from the bus 10, detects the various kinds of diagnostic parameters based on the signal waveform and specifies a fault factor from the combination of the diagnostic parameters. Also, the packet origin position is detected based on the time from packet reception to position detection pulse reception and it is outputted together with the fault factor. Further, a cable test is executed corresponding to the value of the diagnostic parameter and the test result is outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a network monitoring support technology.

[0002]

2. Description of the Related Art As the network connection of PCs in a company progresses, the intranet and the Internet become large-scale, and the connection form of the network network shifts from a bus type using a single cable to a star type using a HUB or the like. I'm starting. Along with this, various managers that can handle multiple segment management and network management tools such as software analyzers that analyze network transmission failures based on whether the logical configuration of transmission data conforms to the network specifications are also used for network management. It is supposed to be.

[0003] By the way, even if a transmission failure at the physical layer level occurs on the network, there is a possibility that a trouble such as a system failure may occur. Not detectable. Therefore, at present, engineers who are familiar with the network standards are required to
A transmission failure at the physical layer level is detected and diagnosed by flowing a test signal through a network line and detecting it with a cable tester or the like, or (b) monitoring the waveform of a signal flowing through the network line with an oscilloscope or the like.

[0004]

However, the conventional methods (a) and (b) for detecting and diagnosing transmission faults at the physical layer level have the following drawbacks. In order to perform the line test (a) using the cable tester or the like, it is necessary to supply a test signal to a network line, and therefore, the operation of the network must be temporarily stopped. That is, according to the line test (a) using the cable tester or the like, it is difficult to grasp the state of the operating network in real time, and its execution hinders efficient operation of the network. On the other hand, to monitor the signal waveform on the network line (b) and discover the intermittent transmission failure that tends to occur on the network without overlooking it, a technician must lengthen the waveform of the signal flowing on the network line. It must be monitored over time. further,
Engineers are required to have a high level of knowledge and experience to discriminate these transmission faults from signal waveforms and identify their causes when a plurality of transmission faults occur in a superimposed manner. . That is, according to the monitoring of the signal waveform on the network line (b), the work load imposed on the engineer becomes extremely large.

Accordingly, an object of the present invention is to provide a network monitoring support device capable of detecting and analyzing a transmission failure on a network at a physical layer level in real time while the network is operating. In this way, it is intended to reduce the work load on the system administrator, improve the reliability of the network, and increase the efficiency of operation.

[0006]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention relates to a network monitoring support apparatus for detecting a network transmission failure, wherein each of the network transmission failures represents the transmission failure. A storage unit storing correspondence information in which transmission failure information and a combination of diagnostic parameter values representing an abnormal pattern included in the waveform of a transmission signal at the time of the transmission failure are stored; and A pulse detecting means for sequentially capturing a transmission signal transmitted through the signal transmission medium;
From the waveform of the transmission signal captured by the pulse detection means,
A combination of a diagnostic parameter detecting means for sequentially detecting a plurality of diagnostic parameter values representing an abnormal pattern included in the waveform, and a diagnostic parameter value detected by the diagnostic parameter detecting means from correspondence information stored in the storage means; A failure diagnosing unit that extracts transmission failure information associated with the network and diagnoses a transmission failure of the network based on the transmission failure information; and an output unit that outputs a diagnosis result by the failure diagnosis unit. A network monitoring support device is provided.

According to the network monitoring support apparatus of the present invention, an abnormal pattern included in a waveform of a transmission signal transmitted through a signal transmission medium of an operating network is detected in real time, and a combination of the abnormal patterns is detected. Since the transmission failure is automatically diagnosed immediately based on the above, the work load on the system administrator can be reduced and the reliability of the network can be improved.

In the network monitoring support apparatus, the diagnostic parameter value may include a network utilization rate smaller than a predetermined network utilization lower limit value.
And when at least one of the number of gap violations larger than a predetermined gap violation number upper limit is detected, the signal transmission of the network is performed at a timing at which the end of the transmission signal captured by the pulse detection unit is captured. If test execution means for executing a test of the medium is provided, and the failure diagnosis means further diagnoses the state of the signal transmission medium of the network based on a test result by the test execution means, the operation of the network is improved. Since the test of the signal transmission medium of the network can be executed at the optimal timing without any hindrance, the reliability of the network can be improved without hindering efficient operation of the network. Also, based on the test results,
Since the state of the signal transmission medium of the network is automatically diagnosed, the work load on the system administrator is reduced.

Further, such a network monitoring support device is provided with history holding means for holding history information of diagnostic parameter values sequentially detected by the diagnostic parameter detecting means, and the output means is further provided by the history holding means. By outputting the retained history information in chronological order,
The system administrator can discover intermittent transmission failures that are likely to occur on the network without overwhelming the monitoring burden of long-time continuous monitoring.

[0010]

Embodiments of the present invention will be described below with reference to the accompanying drawings.

First, the basic configuration of the network system according to the present embodiment will be described.

As shown in FIG. 1, this network system is constructed by connecting network resources 11a to 11h such as personal computers and printers with a bus cable 10. Among these network resources 11a to 11h, the one connected closest to the terminating resistors 10a and 10b at both ends of the bus cable 10 is as follows.
The network monitoring support devices 11a and 11b detect and diagnose network transmission failures at the physical layer level.

One network fault monitoring support device 11
b (hereinafter referred to as slave 11b), as shown in FIG. 2 (a), a packet detection circuit 20 for detecting a packet transmitted on the bus cable 10, and immediately after detecting the packet, a position detection pulse A pulse transmission circuit 21 for transmitting to the network fault monitoring support device 11a is provided.

The other network fault monitoring support device 11
As shown in FIG. 2 (b), a pulse / packet detecting circuit 22 which sequentially takes in a bit string with high impedance from the bus cable 10 and outputs it with low impedance, A diagnostic parameter detector 23 for detecting in real time a diagnostic parameter representing an abnormal pattern included in the waveform of the transmission signal captured by the circuit 22, a position detector 29 for detecting the position of the packet transmission source, and a cable for executing a cable test A test execution unit 24, a control circuit 25 for controlling the cable test execution unit 24 based on the diagnostic parameters detected by the diagnostic parameter detection unit 23, and a diagnostic parameter detection unit 23
A failure diagnostic unit 27 for diagnosing a failure factor based on the detected diagnostic parameter, etc., a memory (not shown) storing various information (described later) necessary for the diagnostic processing of the failure factor, etc., and outputs an analysis result of the diagnostic parameter and the like. An output device 26 is provided. The actual master 11a includes a signal processing device 28a (pulse / packet detection circuit 22, diagnostic parameter detection unit 23, cable test execution unit 24, control circuit 25) directly connected to the bus cable 10, and a personal computer for data analysis. It is configured by connecting the computer 28b (the fault diagnosis unit 27, the storage unit, and the output device 26) with the RS232C.

Now, the diagnostic parameter detecting unit 21
As shown in FIG. 5, the diagnostic parameter values (gap violation number, positive polarity number, collision number, network utilization rate,
It has the following diagnostic parameter detection circuits 30 to 39 for detecting a collision occupancy, a CRC error, a source address of a LOW level packet and a CRC check error, and a packet length in real time. The gap violation number detection circuit 30 counts a packet interval time (see FIG. 4) between a packet and the next packet, and counts the count value as a reference time (a standard value 9.6 μm defined by IEEE 802.3).
A case smaller than s) is detected as a gap violation, and the number of detections per unit time is counted as a gap violation number. Further, the positive polarity number detection circuit 31 outputs noise other than a transmission signal from a personal computer or the like, that is, a predetermined voltage level (for example, +
A positive polarity signal (250 mV) or more (see FIG. 5) is detected, and the number of detections per unit time is counted as a positive polarity number. The collision number detection circuit 32 detects collision data (see FIG. 6) exceeding a predetermined voltage level (for example, +1.6 V) by superimposing a plurality of packets, and uses the number of detections per unit time as the number of collisions. Count. In addition, the network utilization detection circuit 3
3 counts the packet time length A and the packet interval time B until the next packet, respectively, and calculates the ratio {B / (A + B) of the packet time length A to the sum of the packet time length A and the packet interval time B. )} × 100 and outputs it as the network utilization rate (see FIG. 8). Also,
The collision occupancy detection circuit 34 calculates the time length A of the packet.
And the time length C of the collision data included in the packet are counted, and the ratio of the collision data time length C to the packet time length A (C / A) × 10
Calculate 0 and output it as collision occupancy
(See FIG. 7). The CRC check circuit 35 performs a CRC calculation on the bit string of the packet, and outputs a result of the CRC check based on the CRC calculation. Also, L
The OW level packet detection circuit 36 detects a packet below the proper packet voltage level,
When the packet is amplified to the voltage level of the appropriate packet at 0, the address register 37 detects the source address of the packet, outputs the detection result, and outputs the detection result.
The C check circuit 38 performs a CRC calculation on the bit string of the packet and outputs a result of the CRC check based on the CRC calculation. The packet length detection circuit 3
No. 9 counts the number of bits from the delimiter of the packet to the CRC as the packet length.

When the position detection unit 29 detects the end of the transmission signal captured by the pulse / packet detection circuit 22, the transmission time from the time of detection to the detection of a position detection pulse (hereinafter, packet source detection ) Is counted at 100M clock. In practice, the position detection unit 29 is not provided independently, and the transmission time detection circuit of the cable diagnosis execution unit 24 described later is used as the position detection unit 29 under the control of the control circuit 25. I have.

The control circuit 25 periodically obtains the diagnostic parameter value and the packet transmission source detection time from the diagnostic parameter detecting section 21 and the position detecting section 29, and sequentially transfers them to the fault diagnosing section 27. Further, as shown below, a cable test execution command or a cable length measurement execution command is given to the cable test execution unit 24, the test result is obtained from the cable test execution unit 24, and the obtained result is sequentially transferred to the fault diagnosis unit 27. . (1) When the network utilization is smaller than the predetermined lower limit of network utilization, (2) When the collision occupancy is higher than the predetermined upper limit of collision occupancy, (2) The predetermined number of collisions If the number of collisions is larger than the upper limit value, or (3) the case where the positive polarity number is larger than the predetermined positive polarity upper limit value corresponds to one of the cases, the cable test execution unit 24 is instructed to execute the cable test. And output
After a predetermined time has passed since the cable test execution command was given, the cable test execution unit 24 (voltage level detection circuit 93 and transmission time detection circuit 92 described later) transmits a cable test result (voltage level and transmission time (The count value) and transfers it to the fault diagnosis unit 27. If the cable is short-circuited near the connection point of the master 11a, or if the terminating resistor 10a on the master 11a side of the bus cable 10 is disconnected, data cannot be collected by the cable test executed at this time. For
The control circuit 25 forcibly terminates the cable test and gives a cable length measurement execution command to the cable test execution unit 24 (because it is difficult to distinguish between a test pulse described later and a reflected wave). A transmission time count value described later) and the like are acquired, and the acquired values are transferred to the fault diagnosis unit 27.

Then, the cable test execution unit 24
As shown in the figure, a pulse transmission circuit 91 for sending a test pulse to the bus cable 10 in response to a cable test execution command from the control circuit 25, and a preamble to the bus cable 10 in response to a cable length measurement execution command from the control circuit 25 The round-trip transmission time until the test pulse or the like sent from the sending packet sending circuit 94 and the pulse sending circuit 91 returns is 10
A transmission time detection circuit 92 that counts at 0M clock, and a voltage level detection circuit 93 that detects the voltage level of the test pulse reflected and returned at a defective portion (cable break, short circuit, etc.) of the bus cable 10 are provided.

When the cable test execution command is transferred from the control circuit 25, the pulse transmission circuit 91 first generates a test pulse having a sufficient pulse width (for example, 100 ns) which does not attenuate due to the opening of a short circuit at a distant point.
The transmission signal is transmitted to the bus cable 10 at the timing when the end of the transmission signal captured by the pulse / packet detection circuit 22 is detected. If the test pulse and its reflected wave are superimposed due to a short circuit or an open circuit at a short distance point, and the pulse / packet detection circuit 22 cannot properly detect the reflected wave, the pulse transmission circuit 91 performs this test. Instead of a test pulse, a test pulse having a pulse width (for example, 50 ns) that does not overlap with a reflected wave due to a short or open at a short distance point is newly transmitted to the bus cable 10. Then, the voltage level detection circuit 93
Each time the reflected wave of the test pulse sent from the pulse transmission circuit 91 is properly detected by the pulse / packet detection circuit 22,
Four different thresholds for the voltage level of the reflected wave
(250mV, -250mV, -500mV, -1.0
V). The transmission time detection circuit 92 counts the round trip transmission time from when the test pulse is transmitted from the pulse transmission circuit 91 to when the pulse / packet detection circuit 22 detects the head of the reflected pulse of the test pulse. I do. When the bus cable 10 is normal, the reflection of the test pulse does not return and the transmission time detecting circuit 9 does not return.
2 overflows, the control circuit 25
Detects a counter overflow instead of the round trip transmission time counter value.

On the other hand, when the cable length measurement execution command is transferred from the control circuit 25, the packet transmission circuit 94
The 16-byte preamble is transmitted to the bus cable 10. When the preamble is transmitted from the packet transmission circuit 94, the transmission time detection circuit 92 determines the start of the position detection pulse from the slave 11 b from the end of the preamble transmission from the packet transmission circuit 94 to the pulse / packet detection circuit 22. The transmission time up to the detection is counted. If the cable length measurement fails, the counter of the transmission time detection circuit 92 overflows, so that the control circuit 25 detects the counter overflow instead of the round trip transmission time counter value.

The memory stores diagnostic parameter thresholds (the network utilization upper limit value, the collision occupancy upper limit value, the positive polarity upper limit value, and the collision number upper limit value) for the failure diagnosing unit 27 to determine whether a warning is necessary. Value), output device 2
In addition to the warning message output from the storage unit 6, the correspondence information in which the case of the transmission failure of the network and the combination of the diagnostic parameters at the time of occurrence of the failure are stored. Specifically, as shown in FIG. 10A, the correspondence information 100 includes, for each transmission failure case of the network, the case name 100a and the signal to be detected when the case occurs. Abnormalities (CRC error, positive polarity number, alignment error (described later), short packet (described later), long packet (described later), collision, gap violation) are stored in association with each other. In FIG. 10 (b),
The waveform patterns of the packets predicted to have been transmitted through the bus cable 10 when the respective transmission failure cases occurred are shown. The transmission failure examples shown here are:
This is merely an example, and it is possible to add information on a new failure case to the corresponding information as needed.

The fault diagnosing unit 27 includes a control circuit 25
Every time diagnostic parameters are periodically transferred from
As history information, a network utilization rate, a collision occupancy rate, a positive polarity number, a gap violation number, and a collision number are stored in a memory, and a data group included in the history information is arranged from the output device 26 in time series. Output graph. Further, as described below, the output device 26 outputs a warning message according to the diagnostic parameter value and the packet source detection time periodically transferred from the control circuit 25. First, when receiving the data indicating the source address of the LOW level packet and the presence / absence of the CRC check error, the failure diagnosis unit 27 determines from the output device 26 the source address of the LOW level packet and the presence / absence of the CRC check error. Output. In addition, the network usage rate, collision occupancy rate, positive polarity number, and collision number are stored in the respective diagnostic parameter thresholds stored in the memory (network usage rate upper limit value, collision occupancy rate upper limit value, positive polarity number upper limit value, If any of the diagnostic parameters exceeds the diagnostic parameter threshold value as compared with the upper limit of the number of collisions, a warning message read from the memory is output from the output device 26 together with the name and value of the diagnostic parameter. . Furthermore, comparing the number of gap violations with the upper limit of the number of gap violations stored in the memory, if the number of gap violations exceeds the upper limit of the number of gap violations, the packet source and the master that caused the gap violations. the distance L ₁ between the 11a calculates the following equation (1), an output device the value L ₁ 2
6 is output.

L ₁ = L ₀ −t ₁ / (2 × t ₀ ) (1) where L ₀ is the bus cable 10 measured in advance.
, T ₀ is the time required for the pulse to transmit through the bus cable 1 m, and t ₁ is the time for detecting the pulse source.

Further, the failure diagnosis section 27 analyzes the diagnostic parameter values to diagnose a network transmission failure as described below, and outputs the diagnosis result to the output device 26.
Output from First, the packet length is analyzed to determine whether the packet has a normal length (64 bytes to 1518 bytes). Specifically, the packet length is 64B
It is determined that an alignment error has occurred if the packet length is less than ye, a short packet if the packet length exceeds 1518 bytes, and a long packet if the packet length is not a multiple of eight. Then, referring to the correspondence information stored in the memory, the analysis result of the packet length (short packet, long packet, alignment error)
Then, a case name 100a corresponding to a combination of the presence / absence of a CRC error, the presence / absence of a positive polarity number, the presence / absence of a collision number, and the presence / absence of a gap violation number is searched. Then, the warning message read from the memory is output from the output device 26 together with the case name 100a. Further, the distance L ₁ between the packet source and the master 11a that caused to output the warning message is calculated by Equation (1) described above, the value L ₁ also be output from the output device 26.

When the cable test result is transferred from the control circuit 25, the fault diagnosis unit 27 diagnoses the state of the cable based on the test result and outputs the diagnosis result. Specifically, when the result of the cable test is a counter overflow, it is determined that the state of the bus cable 10 is normal, and the determination result is output to the output device 2.
6; otherwise, based on the voltage level of the test pulse, it is diagnosed whether the abnormality occurring in the cable is impedance mismatch or short circuit,
Causes outputs the diagnosis result and a message indicating the state of the bus cable 10 is abnormal from the output unit 26, calculates the following equation the distance L ₂ from the master 11a to defective portions of the bus cable 10 (2), the The output device 26 outputs a value L ₂ and a message indicating that the state of the bus cable 10 is abnormal.

L ₂ = t _c / (2 × t ₀ ) (2) where t ₀ is the time required for the pulse to transmit through the bus cable 1 m, and t _c is the transmission time of the cable test execution unit 24. This is a count value of the time detection circuit 92.

When the cable length measurement result is transferred from the control circuit 25 instead of the cable test result, the failure diagnosis unit 27 diagnoses the cable condition based on the measurement result and outputs the diagnosis result. I do. Specifically, if the result of the cable length measurement is a counter overflow, it is determined that the bus cable test has failed, and the result of the determination is output from the output device 26. Distance L _{2 to} slave 11b
Is calculated by equation (2), and the value and a message indicating that the state of the bus cable 10 is normal are output to the output device 26.
Output from

[0028]

According to the network fault monitoring support apparatus of the present invention, a transmission fault on the network can be detected and automatically analyzed at the physical layer level in real time while the network is operating. Therefore, it is possible to reduce the work load on the system administrator, improve the reliability of the network, and increase the efficiency of operation.

[Brief description of the drawings]

FIG. 1 is a diagram showing a basic configuration of a network system according to an embodiment of the present invention.

FIG. 2A is a configuration diagram of a slave of the network monitoring support device according to an embodiment of the present invention, and FIG.
It is a lineblock diagram of a master of a network monitoring support device concerning one embodiment of the present invention.

FIG. 3 is a configuration diagram of a diagnostic parameter detector of FIG. 2;

FIG. 4 is a diagram for explaining a packet interval time detected by the gap violation number detection circuit of FIG. 2;

FIG. 5 is a diagram for describing a positive polarity signal detected by a positive polarity number detection circuit in FIG. 2;

FIG. 6 is a diagram for explaining collision data detected by a collision number detection circuit in FIG. 2;

FIG. 7 is a diagram for explaining a collision occupancy detected by the collision occupancy detection circuit of FIG. 2;

FIG. 8 is a diagram for explaining a network utilization rate detected by the network utilization rate detection circuit of FIG. 2;

FIG. 9 is a configuration diagram of a cable test execution unit of FIG. 2;

10A is a diagram logically showing a data structure of correspondence information stored in a memory, and FIG. 10B is a diagram showing a bus cable when a failure case registered in the correspondence information occurs. FIG. 7 is a diagram illustrating a predicted waveform of a packet being transmitted.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Bus cable 10a, 10b ... Bus cable 11a, 11b ... Network fault monitoring support apparatus 20 ... Packet detection circuit 21 ... Pulse transmission circuit 22 ... Pulse detection circuit 23 ... Diagnostic parameter detection part 24 ... Cable test execution part 25 ... Control circuit Reference Signs List 26 output device 27 fault diagnosis unit 30 gap violation number detection circuit 31 positive polarity number detection circuit 32 collision number detection circuit 33 network usage detection circuit 34 collision occupancy detection circuit 35 CRC check circuit 36 LOW level packet detection circuit 37 ... address register 38 ... CRC check circuit 39 ... packet length detection circuit 91 ... pulse transmission circuit 92 ... transmission time detection circuit 93 ... voltage level detection circuit 94 ... packet transmission circuit 100 ... correspondence information

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Eiji Watanabe, Inventor 1-15-3, Chuocho, Meguro-ku, Tokyo Towa Computer System Co., Ltd. (72) Inventor Motosumi Kumazaki 504 Shinanomachi, Totsuka-ku, Yokohama-shi, Kanagawa 504-2 F-term (reference) in Hitachi Electronics Service Co., Ltd.

Claims (3)

[Claims] 1. A network monitoring support device for detecting a transmission failure in a network, comprising: for each transmission failure in the network, transmission failure information indicating the transmission failure; and a waveform of a transmission signal when the transmission failure occurs. And storage means for storing correspondence information in which a combination of diagnostic parameter values representing abnormal patterns included in the network is stored; and pulse detection for sequentially capturing a transmission signal transmitted through the signal transmission medium from the signal transmission medium of the network. Means, from the waveform of the transmission signal captured by the pulse detection means,
A combination of diagnostic parameter detecting means for sequentially detecting a plurality of diagnostic parameter values representing an abnormal pattern included in the waveform; and a combination of diagnostic parameter values detected by the diagnostic parameter detecting means from correspondence information stored in the storage means. A failure diagnosing means for extracting transmission failure information associated with the transmission failure information and diagnosing a transmission failure of the network based on the transmission failure information; and an output means for outputting a diagnosis result by the failure diagnosis means. Network monitoring support device. 2. The network monitoring support device according to claim 1, wherein the diagnostic parameter value includes a network usage rate lower than a predetermined network usage rate lower limit value and a predetermined network usage rate upper limit value. Test execution means for executing a test of the signal transmission medium of the network at a timing at which the end of the transmission signal captured by the pulse detection means is captured when at least one of the large number of gap violations is detected. A network monitoring support device, wherein the failure diagnosis unit further diagnoses a state of a signal transmission medium of the network based on a test result by the test execution unit. 3. The network monitoring support device according to claim 1, further comprising: history storage means for storing history information of diagnostic parameter values sequentially detected by said diagnostic parameter detecting means, wherein said output means comprises: A network monitoring support device for outputting history information held by a history holding unit in a time-series manner.