JPH09186780A - Network fault detecting device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect the fault of a line network by collecting communication end state information from a previously started point in time until this time. SOLUTION: In the case of delivering information from a FAX delivery system of a computer 11 to a FAX terminal equipment via line network 15, 16, the communication quality of a line to be selected is checked. A FAX output control program 25 informs the transmission result of a facsimile equipment to a FAX transmission result reception program 24. The transmission result is stored to a transmission result recording section 23 for each line. A monitor program 26 started by a timer 27 for each prescribed time collects a FAX transmission result recorded in the transmission result recording section 23 and compares a failure rate and the number of continuous failure as to the entire network for each line with each reference value. As a result of comparison, when a fault is detected, an alarm program 28 is started and displayed on an alarm device 17 to raise attention of the operator.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a network fault detecting apparatus for detecting a fault between a line and a network based on a statistical processing result of communication completion state information in a system for delivering data to a plurality of communication destinations through a line network.

[0002]

2. Description of the Related Art Conventionally, a FAX distribution system, EDI (E
electronic Data Interchang
e) In a system that distributes a lot of electronic data such as electronic data exchange), if a failure occurs on the computer side that sends the data, the occurrence of the failure can be detected by the function of the computer side. When a failure occurs, the computer side cannot distinguish it from the failure of the network or the terminal device and the busy (in use) state of the terminal device, and it may take many hours, for example, several hours until the failure is found.

In the case of a normal manual transmission, the FAX also requires human intervention on both the transmitting side and the receiving side. Therefore, the above failure can be detected relatively quickly, and recovery by an alternative means can be carried out immediately, but it has been automated. The FAX delivery system takes time as described above, and a large amount of forms are transmitted to an unspecified number of users, so that the influence of a failure is very large.

FIG. 6 shows an explanatory view of an embodiment of the prior art. The occurrence and detection of a failure in the conventional FAX distribution system will be described with reference to the drawings. 6 in FIG.
Reference numeral 2 denotes a FAX distribution system, and 61 is a computer system for operating the FAX distribution system 62. 6
F is for converting the data sent by the FAX distribution system into a format that can be handled by the FAX terminal device through the line F
It is an AX modem.

Reference numeral 64 denotes a communication line for transmitting FAX data, which is a private branch exchange PBX or public exchange network 6 shown at 65.
6 to connect to the FAX terminal device of the terminal. When the abnormality detected by the FAX distribution system is classified according to the occurrence location, four types shown in to of FIG. 6 can be considered.

However, the FAX delivery system can distinguish only two types of abnormalities (-). The reason is that, like a general FAX, the FAX distribution system side can only recognize the line to be transmitted and the FAX terminal device of the other party, and as is generally the case, the inside of the network including the PBX is a black box, Since the internal structure of the network cannot be recognized, the location of failure cannot be specified.

For example, a delivery destination FAX (A in FIG. 6)
When the delivery is attempted from a certain line (shown by a in FIG. 6) but the delivery fails due to a failure, the information that can be recognized by the FAX delivery system is the delivery destination FAX ( A), the used line (a), and the failure event (line disconnection, frequent occurrence of bit errors, etc.) detected on the FAX distribution system side. Therefore, it is impossible to distinguish where the failure occurred in.

In practice, there are many temporary failures, and in most cases transmission is possible without resending the cause of the failure, by resending several times. Therefore, in the case of a FAX distribution system having a retry function, it is often necessary to collectively confirm the distribution result once a day.

However, in rare cases, it may not be possible to deliver even if it is tried many times, and in some cases, it will be known when confirming the delivery result once a day in the conventional system. In such a case, if the network abnormality shown by occurs,
The failure detection will be delayed and the fax distribution operation will be stopped for a long time. The reliability of the network is relatively high even when considering the PBX, and it is rare for an abnormal network to occur. It is economically feasible to always install a supervisor in order to detect an abnormal network early. Not done.

[0010]

As described in the above-mentioned conventional technique, it is rare when data is distributed to a large number of unspecified terminals through a line network by automation such as a FAX distribution system. When communication is not normally performed due to the failure of the network that causes the problem, there is a problem that the detection of the failure is delayed and the operation of the distribution system is seriously affected.

In view of the above points, the present invention utilizes a public line, a switched line network such as ISDN, or the like, and a FAX distribution system or an ED for transmitting data to a large number of different destinations.
It is an object of the present invention to provide a means for improving serviceability by early detection of a network failure in a data exchange system such as I.

[0012]

The above problems can be solved by a network failure detecting apparatus and method configured as follows.

FIG. 1 is a block diagram of the present invention. Reference numeral 1 is a collecting means for collecting success or failure of the distribution result as communication completion status information, 2 is a recording means for accumulating and storing the collected communication completion status information, and 3 is predetermined communication completion status information stored therein. Monitoring means 4 for counting the time intervals and monitoring the occurrence of a failure in the network based on the result of the counting is an alarm means for notifying the operator when a failure in the network is detected as a result of the monitoring.

When transmitting a block of data, the collecting means 1 tries to repair a failure such as a specified number of retries at each stage of communication control, and then determines whether the communication processing is completed successfully or fails. The communication result of the data shown is collected as communication completion state information together with the condition of communication. The communication completion status information obtained here is accumulated as a log of normal communication processing, and the information necessary for facilitating the aggregation of the communication completion status information is recorded by the recording means 2 each time the communication completion status information is obtained. Totaled and stored.

The monitoring means 3 is activated at a predetermined time interval, and the failure of the line network is detected on the basis of the total result of the communication completion state information from the time of the last activation to the current time. The failure is notified to the alarm means 4 and an alarm is issued to the operator. (Claim 1) The failure of each line is detected by totalizing the communication completion status information for each line, and the failure rate of data distribution or the number of continuous failures exceeds each preset standard value. There is. (Claim 2) The failure of the circuit network is detected by totaling the communication completion status information of all the circuits used, and the failure rate of data distribution or the number of consecutive failures exceeds each preset standard value. I have to. (Claim 3) In a distribution system using a FAX terminal on the data receiving side, there is a failure in distribution due to an operation by a terminal side operator, and it is difficult to separate from a network failure, and the present invention is particularly effective. The data to be delivered is facsimile data, and FA
The X distribution system is targeted. (Claim 4) In this way, it is possible to detect a failure in the network from the communication completion status information of a large number of data.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A FAX distribution system will be taken as an example of a concrete system of an embodiment of the invention.

The FAX delivery system is a system for delivering form data created on a computer to a designated facsimile machine via a telephone line or an exchange line such as ISDN, and uses a plurality of lines to simultaneously and in parallel. The form can be sent to a plurality of destinations.

The FAX distribution function is a function of automatically calling the G3 / G4 FAX terminal device to perform FAX transmission and notifying the user of the FAX distribution system of the transmission result. , Paper, etc.)
If it cannot be sent normally, the form retransmitting function for retransmitting the set number of times is also provided.

In the present invention, the "collection and delivery result", that is, the "delivery result" here is for each delivery of one form (one transmission may be to the same destination and may be a plurality of sheets). The result is successful or unsuccessful. If the result of several times of transmission processing by the form retransmitting function is successful, success or failure is recorded as a delivery result at each retransmission.
That is, the number of transmission opportunities is treated as including the number of times of retransmission transmission.

Further, in the communication processing operation for distribution, the communication control function confirms the operation in units of one block on the line, and the block is the other party due to a temporary failure of the communication line or each communication equipment of transmission and reception. Even if it is not received by the terminal, the block is retransmitted and the failure is recovered in block units.However, as a result of this, the failure is temporary and an error log is taken for maintenance separately when repaired. In the present invention, this level of recovered failure is not visible as a delivery result and the result is considered successful. That is, it is recognized as a communication completion state whether the purpose of collectively transmitting a set of forms to a certain destination has succeeded or failed.

FIG. 2 is a block diagram of the embodiment of the present invention, and FIG. 3 is an explanatory diagram of network failure detection based on the transmission failure rate and the number of consecutive failures. FIG. 2 has the same network configuration as the FAX distribution system of FIG. 6 described in the conventional example. That is,
From the FAX distribution system of the computer 11 to the network (P
The image of the form is transmitted to the FAX terminal via the BX15, public network 16 or the like), and the line selected and connected for each distribution is connected to the FAX distribution system.
It is used by the communication function of the AX terminal.

By selecting a part of the line route of the line network for each delivery, for example, even if the communication quality of part of the line route is poor, the above-mentioned block control of communication control and resending of the form delivery unit are performed. It is often the case that another line route is selected and restored due to the function.

In addition, when the FAX terminal is not ready to receive data, it may fail even if a certain line is selected and an attempt is made to make a connection, but if the cause is busy busy or out of paper, for example, repeat the transmission. It will succeed while returning.

Therefore, in the FAX distribution system, the line is selected again by the retransmitting function, so that even if a part of the line is defective, the processing capability is reduced but the system operation seems to be normal and continues.

Although the conventional logging of line errors in block units of communication control is also utilized for system inspections such as regular maintenance, the above-mentioned state of the line network is not recognized as a fault in daily operation, and a fault may occur. Detection will be slow.

As described above, in the switched network, means for detecting the failure of the network early on the side of the user of the network by monitoring the status of the failure on a line-by-line basis or the operating status of the entire network. Will be required.

Before explaining the structure, the principle of fault detection in the present invention will be described with reference to FIG. The figure shows the types of faults that can be detected for each line or for the network (whole) using two indicators, the failure rate and the number of consecutive failures.

First, the failure rate and the number of consecutive failures will be described. As described above, the failure rate considers the process of delivering the form to one destination as one “delivery opportunity”, and if the result is failure, considers it as one “failure” and “the number of delivery opportunities”,
The "failure count" is counted, and the ratio of failure to the number of distribution opportunities is expressed as a percentage. The failure rate is calculated for each line or for all lines, and the original data uses the above-mentioned numerical value of the delivery result recorded at the time (for example, 60 minutes) divided by a preset timer.

The "number of consecutive failures" is obtained by counting 1 when the status of the result of the delivery is failure. Counting is performed for each line and all lines, and if the delivery result recorded for each line in each count unit or all lines is recorded in chronological order and there is "success", the counter for each corresponding line or all lines is cleared. By doing so, it records how many "failures" are consecutive for each unit of count. The original data is the distribution result recorded at the time divided by the timer, but the maximum value of the count during that time is recorded for each count unit.

By the way, as shown in FIG. 3, the intermittent failure occurring in a specific line can be detected relatively frequently in the failure rate of each line. This is because the failure rate increases when the number of failures recorded for each line increases for a particular line, and it relates to the corresponding physical line from the computer to the PBX or the local exchange of the public network. Failures such as line deterioration and poor contact at the connection such as modular jacks are possible.

Similarly, with the number of consecutive failures on a line-by-line basis, a permanent failure occurring on a specific line can be detected early. This is a case in which when a delivery process is attempted, a phenomenon occurs in which delivery fails continuously for a specific line.
X If there is some trouble on the terminal side, such as running out of paper, or if the terminal is busy sending or receiving other data, it will be recognized as a delivery failure, but redelivery will be the same by selecting the line again. There is a possibility that there will be continuous re-delivery to the same terminal on a fax machine of a certain scale.
It is unlikely to occur in the distribution system, and it is considered that there is a problem in the line where failures are recorded continuously. That is,
It will be possible to detect that the line cable from the computer to any switch is physically faulty.

Next, from the failure rate of the network, the intermittent failure occurring in the entire network can be detected relatively frequently. As mentioned above, if the ratio of the number of delivery failures to the number of delivery opportunities is higher than usual for all the lines used by the delivery system of the lines connected to the computer, it does not mean that the particular line is a failure. Means that a specific line is to be detected separately as the above-mentioned line-by-line fault detection, indicating that an abnormality has occurred intermittently in the entire network regardless of the line. For example,
There may be an intermittent failure in the common part of the network, but there may be an error in setting some telephone number transfer destinations in the PBX instead of the failure in the network itself.

When the distribution fails for the entire network, that is, for all the lines used by the FAX distribution system, in a chronological order, the distribution process that attempts to sequentially send to various lines fails one after another. Common equipment, cables, etc. may be impaired, that is, paralysis of the network may occur. At this time, of course, the failure rate is also detected for each line, but the number of continuous failures for a large number of distributions in the entire system enables the detection of a serious failure even earlier. A complete network down can be immediately detected even at the stage of counting the number of consecutive failures in the recording unit.

As described above, the failure of the line network is not affected by the situation of the partner FAX terminal due to the failure rate of the delivery to the entire line (network) and the number of continuous failures for each line for which the status information of the delivery result is recorded and summarized. The mechanism for early detection is a system configuration diagram as shown in FIG.

In FIG. 2, reference numeral 11 denotes a computer on which the FAX distribution system operates. Reference numeral 12 denotes an operation unit of a computer, which performs input and output according to an operator's instruction. Reference numeral 13 indicates the form data on the computer, which is the result of the processing by the distribution system, according to the CCITT regulations.
It is an image expansion device that expands to a format that can be handled by X. The image information developed here is the FAX output device 1
A call is made by means of 4 and is sent to the FAX terminal through the PBX 15 and the exchange of the public network 16 through the line.

Reference numeral 17 is an alarm device for notifying an operator when a network fault related to the present invention of the distribution system is detected. 18 is a file device for holding various data to be delivered, and 19 is a FAX for extracting various data to be delivered.
It is a program that formats and edits it so that it is sent to a terminal, and outputs it to the spool file 21. The FAX transmission program 22 is a program for taking out form data from the spool file 21 and giving a FAX transmission instruction. The FAX output control program 25 is a program for executing FAX transmission according to the CCITT protocol. Is to control.

The FAX output control program 25 also sends the FAX transmission result to the FAX transmission result receiving program 2
Notify 4. The present invention detects a network fault based on the delivered transmission result.

The configuration and operation will be described with reference to the flow chart of the network failure detection operation shown in FIG.
The format of the received FAX transmission result data is shown in FIG. That is, the distribution end time, distribution result, line number, and terminal number for each mail. For example, in the first line of the figure, the result of the distribution completed at 15:34:26 is OK, the line number used is 07, and the FAX terminal number of the destination is 8
It was shown that it was 9. In the next line, the fact that the result was unsuccessful is indicated by NG in the delivery result column. As shown in FIG. 4A, the FAX transmission result reception program 24 receives the distribution results of all the lines of the FAX distribution system in the order of time (step S51).

The distribution result is accumulated and recorded as a log in the transmission result recording section 23 as it is (step S51), and at the same time, it is sorted according to the line number as shown in FIG. Are counted and counted in the column of the number of delivery opportunities (step S52). If the result is OK, the number of consecutive failures in the corresponding line number field is stored in the maximum number of failures field and then cleared (step S54). If the result is NG, the number of failures and the number of consecutive failures are calculated. The number is counted by adding 1 to the column (step S55). At this time, it is checked whether the number of consecutive failures is within the standard value, and if it is found that the standard is exceeded, the failure detection is immediately performed by notifying the alarm program of the network network failure detection (step S50, Step S59).

Simultaneously with the processing for each line, it is recorded in each column of the number of delivery opportunities, the number of failures, the number of consecutive failures, and the maximum number of consecutive failures of the entire network (step S52, step S54, step S5).
5). Explaining with the example entered in FIG. 4 (b), regarding the line 2, the count record value from the time when the totalization for the previous monitoring is made is the FAX distribution 1 at the present time.
You can see in the column of the number of delivery opportunities that you tried 2 copies, 5 of which
The number of failures is recorded in the number of failures column. Furthermore, the number of consecutive failures is recorded as 4 that the last four failures have failed consecutively. Also, the maximum failure count column is 4 times. The failure rate is to be processed collectively at the next aggregation, but it is entered for explanation, but the failure rate (5) is divided by the number of delivery opportunities (12) 4
It is 1.7%.

Next, the monitoring program is started by the timer 27 at regular intervals, and the FAX transmission result recording table recorded in the transmission result recording unit 23 is totaled (step S56), and the failure rate and continuous failure for each line and the entire network are collected. The number is compared with each of the monitoring reference values (steps S57 and S58) to detect a network abnormality. The standard value table is shown in FIG. In the example shown in the figure, the failure rate is 50 for each line.
%, The number of consecutive failures is 10 and the failure rate is 5 for the entire network.
0% means that the number of consecutive failures is 100.

This reference value is set as follows. The failure rate of normal FAX distribution is about several percent, but 50% is adopted as a value that is sufficiently larger than the above value, assuming the detection of a line or network hang-up state. In addition, the number of consecutive failures is within the maximum number of transmittable documents (not counting the number of retransmissions) that can be processed by a certain line or the entire network within a certain period of monitoring, and more than several times the number of automatic retransmissions per message. Is set to 10 for each line and 100 for the entire network. Of course, these can be modified according to the actual results.

When the alarm program 26 detects an abnormality in the network by comparing it with the standard value, the alarm program 28 is activated (step S59) and displayed on the alarm device 17 to alert the operator and affect the entire system. Enables early response to obstacles.

[0044]

As is apparent from the above description, according to the present invention, a public line, a switched line network such as ISDN is used,
In a data exchange system such as a FAX delivery system or EDI that sends data to a large number of different destinations, it is possible to detect a network failure early and improve the service to the users of the delivery system and the reliability of system operation. It has a remarkable industrial effect.

[Brief description of the drawings]

FIG. 1 is a block diagram of the present invention.

FIG. 2 is a configuration diagram of an embodiment of the present invention.

FIG. 3 is an explanatory diagram of network failure detection based on a transmission failure rate and the number of consecutive failures.

[Figure 4] Communication completion status information and standard value table

FIG. 5 is a flowchart of the operation of network failure detection.

FIG. 6 is an explanatory diagram of an embodiment of a conventional technique.

[Explanation of symbols]

 1 collecting means 2 recording means 3 monitoring means 4 alarm means

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code Agency reference number FI Technical display location H04N 1/32 H04N 1/32 J

Claims (5)

[Claims] 1. In a system for distributing data to a plurality of communication destinations through a network, a collecting means for collecting success or failure of a distribution result as communication completion status information, and a record for accumulating and storing the collected communication completion status information. Means and a monitoring means for aggregating the stored communication completion status information at a predetermined time interval and monitoring the occurrence of a failure in the network based on the result of the aggregation, and as a result of the monitoring, the occurrence of a failure in the network is detected. A network failure detection device, comprising: an alarm means for notifying an operator when a failure occurs. 2. The communication completion status information is aggregated for each line, and the line failure is detected when the failure rate of data distribution or the number of consecutive failures exceeds a preset reference value. The network fault detection device according to claim 1. 3. Completion of communication completion status information of all lines used, and detection of a line network failure by detecting the failure rate of data distribution or the number of consecutive failures exceeds a preset standard value. The network fault detection device according to claim 1 or 2. 4. The network failure detection device according to claim 1, wherein the data to be distributed is facsimile data. 5. In a system for delivering data to a plurality of communication destinations through a network, the success or failure of the delivery result is collected as communication completion status information, and the collected communication completion status information is stored and stored, and then stored. The stored communication completion status information is aggregated at a predetermined time interval, and the occurrence of a line network failure is monitored based on the aggregated result, and the operator is notified when a failure is detected in the line network as a result of the monitoring. A method for detecting a network fault, comprising: