JP3523979B2 - Fault tracing method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION This invention performs cause analysis of failures in the system, which relates to the fault trace how tracing the fault information, in particular, AT
Multiple computers with M interface, router, CL
ATM terminals such as AD (cell assembling and disassembling equipment) are UNI
ATM via (user network interface)
Is connected to the ATM switch device constituting the (Asynchronous Transfer Mode) network, but about the fault trace how applicable to ATM communication system to communicate with each other using ATM.

[0002]

2. Description of the Related Art FIG. 18 shows, for example, Japanese Patent Laid-Open No. 63-83843.
FIG. 3 is a conceptual diagram showing a conventional fault tracing method disclosed in Japanese Patent Publication No. JP-A-2003-264, in which 31 is a trace buffer writing unit for writing trace information to a trace buffer 32, and 33 is a trace buffer writing unit for each trace buffer 32 when the trace buffer 32 is full. A trace file writing unit that writes the contents of the trace buffer 32 to the trace file 34, and a trace control unit 35 that instructs only the trace file writing unit 33 to start and stop the operation. Reference numeral 36 denotes a failure information buffer writing unit that notifies the failure type determination unit 38 of the completion of the writing after writing the failure information in the failure information buffer 37, and 38 denotes the latest failure information written in the failure information buffer 37. Of these, the near-side trace unit 39 determines whether or not it is a necessary fault, and is a fault type determination unit that activates the near-trace unit 39 if necessary. 39
Reads out the near trace information at the present time from the trace buffer 32 and writes it in the fault information buffer 37, and then writes this write end to the fault information file writing unit 40.
This is a trace part in the vicinity to notify the. Reference numeral 40 is a failure information file writing unit that writes the latest failure information written in the failure information buffer 37 and related trace information to the failure information file 41. The failure information buffer writing unit 36, the failure information buffer 37, the failure type determination unit 38, the neighborhood trace unit 39, the failure information file writing unit 40, and the failure information file 41 constitute a failure information collection unit.

Next, the operation will be described. First, the trace buffer writing unit 31 writes the corresponding trace information into the trace buffer 32 each time an event requiring collection of the trace information occurs in the system to which this method is applied. Then, each time the trace buffer 32 becomes full, the trace file writing unit 33
The contents are written in the trace file 34. When the trace control unit 35 receives a trace stop command from outside the system, the trace control unit 35 operates only in the trace file write unit 33 (write operation in the trace file).
Command to start and stop. Therefore, even if the trace file writing unit 33 stops its operation, the operation of the trace buffer writing unit 31 is continued.
As a result, even when the trace buffer 32 is full, the trace file 34 is not saved, the latest trace information is sequentially overwritten with the oldest trace information, and only the latest trace information of the capacity of the trace buffer 32 is stored. The trace buffer 32 remains.

In parallel with this, the failure information collecting units 36-4
1, the fault information with the near trace information is collected with respect to the predetermined information generated in the system. That is, the failure information buffer writing unit 36 writes failure information including a failure code and time of occurrence to the failure information buffer 37 every time a failure occurs in the system, and then terminates the writing. Type determination unit 3
Notify 8. Upon receiving the notification of the completion of writing, the failure type determination unit 38 reads the latest failure information written in the failure information buffer 37 and compares the failure code included therein with a predetermined failure code, It is determined whether or not the near trace is a necessary fault, and if necessary, the near trace unit 39 is activated. The activated neighborhood trace unit 39 reads out a predetermined number of neighborhood trace information from the trace buffer 32 for writing in the fault information buffer 37, and then notifies the fault information file writing unit 40 of the completion of the writing. The failure information file writing unit 40 that has received the write end notification writes the latest failure information written in the failure information buffer 37 and the near trace information related thereto in the failure information file 41.

Incidentally, as another prior art document related to the present application, there is JP-A-5-298144.

[0006]

Since the conventional fault tracing method is configured as described above, when applying such a trace information collecting method to an ATM communication network,
TM cell discard, cell switch buffer overflow, or UPC (Usage Parameter)
A network failure such as a user cell being discarded due to the control function of the Control) is likely to be repeated frequently because the cause of the failure does not necessarily exist in the device that has detected the failure. When the error occurs, most of the trace files are filled with these failure information, and the duration of the failure and the information when it first occurred are lost, which hinders the analysis of the cause of the network failure. was there.

The present invention has been made to solve the above problems, and when tracing frequently occurring faults, while tracing the fault information is suppressed, the number of fault occurrences and the latest or oldest fault are traced. important failure information including failure time is traced to prevent dropped and the and splicing
Keep trace area down for long-lived failures
It is an object of the present invention to provide fault trace how the capable.

[0008]

According to a first aspect of the present invention, there is provided a fault tracing method including a detecting step of detecting a fault occurring in a system, and a tracing step of tracing the fault content detected by the detecting step. when said detecting step occurs continuously repeated failure detected, the fault trace a trace suppression step of suppressing the fault contents above trace step traces
In the method, the trace suppression step is a trace step.
First, the failure that occurred first is the content of the failure and when the failure occurred.
Trace the time and monitor after the first failure
The failures that occurred during the period are as follows:
And the latest failure occurrence time within the specified monitoring period
If a race is continued and an obstacle that occurs continuously continues,
Failure of failures that occurred during the monitoring period, which is obtained by multiplying the monitoring period by n times.
Traces the contents of harm, the number of occurrences, and the latest failure occurrence time.
It was made to be able to .

In the fault tracing method according to the second aspect of the present invention, the trace inhibiting step includes the tracing step,
For the first failure, trace the failure description and failure occurrence time.
Within the specified monitoring period after the first failure.
The failure that occurred is the content of the failure, the number of times the failure occurred, and
The earliest failure occurrence time that occurred within the fixed monitoring period is traced.
In addition, the faults that have occurred continuously are for the specified monitoring period.
The failure content of the failure that occurred within the monitoring period that is sequentially multiplied by n,
The number of occurrences, the oldest failure occurrence time, and the latest failure occurrence time
It is designed to be traced.

In the fault tracing method according to the third aspect of the present invention , an upper limit value is set for the monitoring period, and the monitoring period is above the upper limit.
When the limit value is reached, the upper limit value is fixed.
is there.

In the fault tracing method according to the present invention, the fault detected by the detecting step does not occur for a predetermined period.
Trace suppression that cancels the trace suppression
A release step is provided.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION The following is not an embodiment of the present invention.
However, a reference example related to this will be described. Reference example 1. FIG. 1 is a block diagram showing an ATM communication system. In the figure, 11 is an ATM switch device of the ATM communication system, 11-1 is a microprocessor of the ATM switch device 11, 11-2 is a memory of the ATM switch device 1,
Reference numeral 1-3 is an ATM switch card of the ATM switch device 11, 11-4 is a user interface card of the ATM switch device 11, and this user interface card 11-4 is an ATM interface (UNI (user network interface) or NNI (network network)). Interface)) providing port 1
A plurality of 1-4-1 are mounted. In this way, ATM
The switch device 11 includes one microprocessor 11-
1, memory 11-2, ATM switch card 11-3,
And an ATM interface card 11-4.

The microprocessor 11-1 has a memory 1
1-2, the ATM switch card 11-3, and each port 11-4-1 of the user interface card 11-4 are accessible, and the protocol processing required for ATM communication is performed by the microprocessor 11-1 and the ATM. The switch card 11-3 and the user interface card 11-4 are shared. Also, the microprocessor 11-
1 detects a failure of the memory 11-2, the ATM switch card 11-3, and the user interface card, which are the peripheral circuits, and the ATM switch card 11-3.
Also, the failure information detected by interruption or polling from the user interface card 11-4 is traced.

Reference numeral 12 denotes a terminal device of an ATM communication system, 1
2-1 is the CPU of the terminal device 12, 12-2 is the terminal device 1
2 is a main memory, and 12-3 is an ATM interface of the terminal device 12 for interfacing with the ATM switch device 11. Reference numeral 3 is a data transmission line formed by an optical fiber or a twist bear cable, which connects the ATM switch device 11 and the terminal device 12 and the ATM switch device 11 together. Thus, the terminal device 12 has the CPU 12-
1, main memory 12-2, and ATM interface 12
-3, the terminal device 12 is UNI-connected to the ATM switch device 11 via the data transmission line 3.

The CPU 12-1 can access the main memory 12-2 and the ATM interface 12-3, and the protocol processing required for ATM communication is this CP.
U12-1 and ATM interface 12-3 share. Further, the CPU 12-1 detects a failure of the main memory 12-2 and the ATM interface 12-3 which are its peripheral circuits, and traces failure information detected by interrupt or polling from the ATM interface 12-3.

FIG . 2 is a conceptual diagram showing a fault tracing method according to the first reference example . In FIG . 2, 1 writes fault trace information in a fault trace buffer 2 and outputs fault information to a fault trace identifying section 5. It is a writing unit to the failure trace buffer. 3 is a write-in portion of the fault trace file for writing the contents of the fault trace buffer 2 to the fault trace file 4.
Reference numeral 5 identifies the failure information output from the failure trace buffer writing unit 1 (determines whether or not the failure is to be traced), and if the failure is to be traced, the failure information is sent to the trace suppression control unit 6. This is a fault trace identification section for output. Reference numeral 6 is a monitoring timer 7 (occurrence monitoring timer 7a, based on the failure information output from the failure trace identifying unit 5).
This is a trace suppression control unit that executes the trace suppression process using the non-occurrence monitoring timer 7b). Reference numeral 8 is a near trace unit that, when the trace suppression control unit 6 determines that it is a fault to be traced, writes additional information such as the fault occurrence time and the number of occurrences to the contents of the fault trace buffer 2.

The fault trace buffer writing unit 1, fault trace file writing unit 3, fault trace identifying unit 5, trace suppression control unit 6, and monitoring timer 7 are used by the microprocessor 1 of the ATM switch device 11.
1-1 and the CPU 12-1 of the terminal device 12, the fault trace buffer 2 and the fault trace file 4 are realized by the memory 12-2 of the ATM switch device 11 and the main memory 12-2 of the terminal device 12. It is what is done.

[0018] Next, the operation will be described. Figure 3 is this reference
FIG. 4 is a flowchart for explaining the processing operation of the fault tracing method according to Consideration 1 , FIG. 4 is a flowchart for explaining the occurrence monitoring timer processing operation, and FIG. 5 is a flowchart for explaining the non-occurrence monitoring timer processing operation. The process flow will be described below with reference to FIGS. 3 to 5.

When a failure occurs in the ATM communication system (step ST1, detection step), the microprocessor 11-1 or the CPU 12-1 collects the failure information that has occurred. The fault information collected by the fault trace buffer writing unit 1 is written in the fault trace buffer 2 (step ST2, detection step), and at the same time, the fault information is output to the fault trace identifying unit 5.

The fault trace identification unit 5, it is determined whether the failure to be traced from the type of disorder when receiving the fault information (in which those variously changed according to the type system) ( Step ST3, trace inhibiting step). (A) When it is not a failure to be traced When the failure trace identifying unit 5 determines that it is not a failure to be traced, the processing is terminated without performing any processing.

[0021] When the (B) If a fault should be traced fault trace identification unit 5 is determined to be failure to be traced, the fault trace identifying unit 5 outputs the failure information to the trace inhibit control unit 6. Trace suppression control unit 6
Upon receiving the fault information output from the fault trace identifying section 5, the device determines whether or not the fault is under trace inhibition (step ST4, trace inhibition step).

[0022] If the case is not in (a) trace deterrence is not in trace suppression (that is, if the failure of the first generation), the vicinity of the trace unit 8, the failure count 1
(Step ST5, trace suppression step), and the failure occurrence time is set to the current time (step ST6,
Trace suppression step), the occurrence frequency and the occurrence time of the failure are written in the contents of the error trace buffer 2, and the error trace file writing unit 3 writes the failure occurrence number and the occurrence time together with the failure information. Is written in the failure trace file 4 (step ST7, trace step). Then, the process of the occurrence monitoring timer 7a is activated (started) (step S
T8, trace suppression step).

The process of generating the monitoring timer 7a is performed as shown in FIG. First, when the process of the occurrence monitoring timer 7a is started, it is determined whether or not the monitoring timer 7 is activated (step ST101, trace inhibiting step). If it is not activated by the monitoring timer 7 (that is, if the occurrence monitoring timer 7a processing is directly activated when the trace suppression is not being performed), the monitoring timer 7 is set to the trace suppression (trace suppression state) (step ST102, trace suppression). Step), the failure occurrence count is reset to 0 (step ST106, trace inhibiting step), and the initial value of the occurrence monitoring timer 7a is set to T1 (step ST1).
07, trace inhibiting step) The occurrence monitoring timer 7a is started (step ST108, trace inhibiting step).

[0024] Thereafter, since a failure to release the trace inhibited state of the trace inhibit control unit 6 when a predetermined time (T2) does not occur, set T2 to the initial value of the non-occurrence monitoring timer 7b (step ST9, trace inhibit release Step),
The non-occurrence monitoring timer 7b is activated (step ST10,
Trace suppression release step).

The processing of the non-occurrence monitoring timer 7b is performed as shown in FIG. When the non-occurrence monitoring period of the non-occurrence monitoring timer 7b expires, the corresponding fault occurrence monitoring timer 7a
Stop (step ST201, trace suppression release step), and the non-occurrence monitoring timer 7 for the corresponding fault.
b is stopped (step ST202, trace suppression cancellation step), and the trace suppression in progress of the corresponding fault is reset (step ST203, trace suppression cancellation step).

[0026] (b) in the case when it is being traced deterrence is in trace deterrence, already occurrence monitoring timer 7
Since the process of a is activated, the failure occurrence count is updated without directly performing the trace process. That is, the failure occurrence count is incremented by the new failure occurrence count (step ST11, trace suppression step), and the current time is set as the failure occurrence time (step ST12,
Trace suppression step).

The process of generating the monitoring timer 7a in the case is in the trace suppression is performed as shown in FIG. First, it is determined whether or not the monitoring timer 7 is activated (step 101). If it is determined that the monitoring timer 7 is activated (in this case, the monitoring timer 7 is activated), then the failure occurrence count is determined. It is determined whether or not it is 0 (step ST103, trace suppression step). If the failure occurrence count is 0, trace processing is not performed and
Occurrence monitoring timer 7 to start a new occurrence monitoring period
The initial value of a is set to T1 (step ST107), and the generation monitoring timer 7a is started (step ST108). If the failure occurrence count is 0, the trace contents (contents of the failure trace buffer) are assembled from the failure code, the failure occurrence count, and the failure occurrence time (step ST104,
(Trace suppressing step), and writing to the failure trace file 4 is executed by the failure trace file writing unit 3 (step ST105, trace step).

In the reference example 1 , since the failure occurrence time is updated with the current time every time a failure occurs, the traced failure occurrence time is the latest failure occurrence time. FIG. 6 is a diagram for explaining a case where the number of times of failure occurrence and the time of failure occurrence are traced during trace suppression. In this case, the number of times of failure occurrence (the number of failures that occurred during the occurrence monitoring period T1 of the occurrence monitoring timer 7a) is 5, and the latest failure occurrence time t5 is recorded (traced) as the trace occurrence time.

[0029] Then, reset the fault count to zero (step ST 106, trace suppression step), the initial value of the occurrence monitoring timer 7a as T1 (step ST
107) Starts the occurrence monitoring timer 7a (step ST
108).

[0030] Thereafter, similarly to the case not being traced deterrence, set T2 to the initial value of the non-occurrence monitoring timer 7b (step ST9), activates the non-occurrence monitoring timer 7b (step ST10).

FIG . 7 is a diagram showing an example of tracing by the fault tracing method according to the first reference example . In FIG. 7, the occurrence monitoring period of the occurrence monitoring timer 7a is T1, and the non-occurrence monitoring period of the non-occurrence monitoring timer 7b is T2. The first failure is unconditionally traced, but the two failures that occurred in the next occurrence monitoring period T1 are collectively traced, and similar trace suppression continues three times.
After that, although no failure has occurred for a while, the failure has occurred before the non-occurrence monitoring period T2 of the non-occurrence monitoring timer elapses, so the trace suppression continues. However, since no fault has occurred in the second occurrence monitoring period T2, the trace inhibition state is released, and the trace is unconditionally performed at the first fault thereafter.

As described above, according to the first reference example , the trace suppression control unit 6 controls the monitoring timer 7 to trace all the fault information of the fault that has occurred, and the fault code of the fault that has occurred. Since the number of occurrences and the time of occurrence are traced, even for frequently occurring faults, the trace area is not filled with this, the tendency of occurrence can be grasped, and the cause of fault occurrence (or failure analysis) can be investigated. It can be done easily.

Reference Example 2. In Reference Example 1 above, the latest failure occurrence time was traced among the failures that occurred during the trace suppression period, but in Reference Example 2 , among the failures that occurred during the trace suppression period, the oldest failure occurred. The time of occurrence is traced. FIG. 8 is a flow chart for explaining the processing operation of the fault tracing method according to the second reference example .
Note that the configuration for realizing the fault trace method, the ginseng
Since it is similar to FIG. 2 described in Consideration 1 , the description thereof is omitted.

[0034] Reference Example fault information generated in 1 (step ST1), writing to the fault trace buffer (step ST2), the determination whether or not a failure to be traced (step ST3) (A) is not a fault to be traced when the processing End (B) If it is a fault to be traced, it is judged whether or not the fault is being trace-suppressed (step ST4). (A) If trace is not being suppressed, the fault occurrence count is 1 (step ST5), and the fault occurrence time is set to the current time. Setting (step ST6), writing to the fault trace file 4 (step ST7), starting processing of the occurrence monitoring timer 7a (step ST8), setting T2 to the initial value of the non-occurrence monitoring timer 7b (step ST9), non-occurrence monitoring timer 7b startup processing (step ST10) (process steps), of the reference example 2 shown in FIG. 8 The same applies to the physical behavior.

[0035] (b) in the case when it is being traced deterrence is being traced suppression, without direct tracing, updates the failure count. That is, the failure occurrence count is added by the new failure occurrence count (step ST11), and it is determined whether or not the failure occurrence count is 1 (step ST12A, trace inhibiting step).
The current time is set to the failure occurrence time only when the failure occurrence count is 1 (step ST13A, trace inhibiting step).

[0036] In this manner, in this Reference Example 2, failure time to be traced is a failure time of the oldest. Figure 9
FIG. 9 is a diagram for explaining a case where the number of times of failure occurrence and the time of failure occurrence are traced during trace suppression. in this case,
Failure occurrence frequency (occurrence monitoring period T1 of the occurrence monitoring timer 7a
The number of failures that occurred during the recording is 5 times, and the oldest failure occurrence time t1 is recorded (traced) as the trace occurrence time.

The subsequent processing sets the T2 to the initial value of the non-occurrence monitoring timer 7b (step ST9), start the non-occurrence monitoring timer 7b (step ST10) is the same as in Reference Example 1.

As described above, according to the second reference example , since the first fault occurring can be unconditionally traced, the time interval from the time when the first fault occurs to the time when the next fault occurs is important. It is possible to exhibit high failure analysis performance for various failures.

Embodiment 1. Next, an embodiment of the present invention will be described. Above reference
In the example 1 , in the process of the occurrence monitoring timer 7a, the occurrence monitoring timer 7a is unconditionally started with the initial value of the occurrence monitoring timer 7a set to T1, but in the first embodiment of the present invention , the trace suppression continues. As long as the initial value of the occurrence monitoring timer is the same as the initial value of the previous occurrence monitoring timer, n (real number)
It doubles.

[0040] Figure 10 is an embodiment of the such the present invention
6 is a flowchart for explaining the processing of the occurrence monitoring timer according to No. 1 ; Monitoring timer 7 in this processing step
(Step ST101), if the monitoring timer 7 is not activated, the monitoring timer 7 is set during trace suppression (step ST102), and it is determined whether the failure occurrence count is 0 (step ST103). Assembling the trace contents when the occurrence count is 0 (step ST104), writing to the failure trace file (step ST105), and setting the failure occurrence count to 0
The resetting (step ST106) is similar to the process of the occurrence monitoring timer 7a according to the first reference example shown in FIG.

In the first embodiment , after resetting the failure occurrence count to 0 (step ST106), the initial value of the occurrence monitoring timer 7a is set to n times the initial value of the previous occurrence monitoring timer (step ST107B). , Trace inhibiting step) and the generation monitoring timer 7a is started (step ST108, trace inhibiting step).

[0042] Figure 11 is a diagram showing an example of tracing the fault trace method according to the first embodiment of the present invention. As shown in FIG. 11, as long as the failure occurs during the trace suppression period, the occurrence monitoring period of the occurrence monitoring timer 7a is multiplied by n. If no failure occurs for a certain period of time, the trace suppression status will be released.

As described above, according to the first embodiment, since the trace inhibition period can be exponentially increased, the trace area is further suppressed for a fault having a long continuous generation period once it occurs. be able to.

Embodiment 2. Second embodiment of the invention, the above-described embodiment in writing newest failure time the oldest fault occurrence time of the failure during the trace prohibition period (trace) point 1
Is different from. FIG. 12 shows such an invention.
9 is a flowchart for explaining the processing operation of the fault tracing method according to the second embodiment . The description of the processing operations that overlap with the processing operations described in the first reference example and the first embodiment will be omitted.

[0045] (B) when the case if a fault should be traced (a) not being traced deterrence not being traced suppression, Near trace section 8,
The failure occurrence count is set to 1 (step ST5), the current time is set to the oldest failure occurrence time and the latest failure occurrence time (step ST6C, trace suppression step), and the failure occurrence count and occurrence time are set to the failure trace buffer 2 Writing to the contents of and writing to fault trace file 3
However, the contents of the fault trace buffer 2 in which the number of times the fault has occurred and the time when the fault occurred are written to the fault trace file 4 together with the fault information (step ST7). Next, the process of the occurrence monitoring timer 7a is activated (started) (step ST8). After that, in order to cancel the trace suppression state of the trace suppression control unit 6 when the failure does not occur for a predetermined time (T2), T2 is set to the initial value of the non-occurrence monitoring timer 7b (step ST9), and the non-occurrence monitoring timer 7b is set. Is activated (step ST10).

[0046] (b) in the case when it is being traced deterrence is being traced suppression, after updating the failure count, i.e., after only a new failure number of times adding the failure count (step ST11), The current time is set to the latest failure occurrence time (step ST12C, trace suppression step), and the current time is set only to the oldest failure occurrence time when the failure occurrence count is 1 (step ST13C, trace suppression step) (step ST14).
C, trace suppression step). Thereafter, as in the case where the trace suppression is not being performed, T2 is set to the initial value of the non-occurrence monitoring timer 7b (step ST9), and the non-occurrence monitoring timer 7 is set.
b is started (step ST10).

As described above, according to the second embodiment , the oldest failure occurrence time and the latest failure occurrence time among the failures that occurred during the trace inhibition period in the first embodiment are traced. Therefore, during the long trace suppression period in the above-described first embodiment, it is possible to obtain information as to whether the faults are occurring on average or in bursts, and for a fault with a long continuous occurrence period. The trace area can be suppressed and high failure analysis performance can be exhibited.

Embodiment 3. Third embodiment of the invention, an upper limit is provided to the initial value of the occurrence monitoring timer 7a, the occurrence monitoring timer 7a initial value at the point of fixing to the upper limit value of the above embodiment reaches the upper limit value
The second embodiment is different from the first embodiment and the second embodiment . FIG.
Is a flow chart for explaining the process of the occurrence monitoring timer according to the third embodiment of the present invention.
In addition, the reference example 1 , the first embodiment , and the above embodiment
Regarding the processing operation that overlaps with the processing operation described in 2 ,
The description is omitted.

In the third embodiment, after resetting the failure occurrence count to 0 (step ST106), it is judged whether the occurrence monitoring period of the occurrence monitoring timer 7a has reached the upper limit value (step ST107D, trace inhibition). If the upper limit has not been reached, the initial value of the occurrence monitoring timer 7a is set to n times the initial value of the previous occurrence monitoring timer (step ST108D), and the occurrence monitoring timer 7a is activated ( On the other hand, if the upper limit value is reached, the upper limit value is reset to the initial value of the occurrence monitoring timer 7a (step ST109D, trace suppression step), and the occurrence monitoring timer 7a is activated (step ST110D). ST110D).

As described above, according to the third embodiment, in the first embodiment and the second embodiment ,
An upper limit value is set for the initial value of the occurrence monitoring timer 7a, and when the upper limit value is reached, the initial value of the occurrence monitoring timer 7a is fixed to the upper limit value, so that the trace suppression period becomes extremely large and is not useful for failure analysis. This can be prevented in advance, and an increase in H / W cost due to an increase in the count value of the timer H / W can be suppressed.

Reference Example 3. Reference examples related to the present invention will be described again.
Reference example 3 of the above is during the trace suppression period (occurrence monitoring timer 7
the ginseng disorders that occurred occurred during the monitoring period) of a, instead of tracing every factor, each bit together multiple factors in that traces to correspond to a factor
This is different from Case 1 . Figure 14 is such a reference
11 is a flowchart for explaining processing of an occurrence monitoring timer according to example 3 . The description of the processing operation that is the same as the processing operation described in Reference Example 1 will be omitted.

[0052] (b) in the case when it is being traced deterrence is being traced suppression, without direct tracing, updates the failure count. That is, the failure occurrence count is added by the new failure occurrence count (step ST11). Then, the failure occurrence time is only set to the current time (step ST12). Thereafter, it is judged whether or not the generated faults should be traced together (step ST13E, trace step), and if they are traced together, the contents of the fault factor buffer are changed to the accumulated contents of the fault factor buffer. And the contents of the failure factor register are updated by a bitwise logical sum (step ST14E, trace step). After that, as in the case where the trace is not being suppressed, the non-occurrence monitoring timer 7b
T2 is set to the initial value of (step ST9), and the non-occurrence monitoring timer 7b is started (step ST10).

As described above, according to the third reference example, in the first reference example , the failure occurring during the monitoring period of the occurrence monitoring timer 7a is not traced for each factor but a plurality of factors are detected. Since each bit is traced in association with one factor, a plurality of factors may be set when an error occurs (eg SDH).
Failures defined in (Synchronous Data Hierarchy) (B
(IP8 error, LOS, LOF, etc.) will be traced together, so it is possible to prevent the trace size from becoming enormous if traces are taken for each factor although they are related to each other. it can.

Reference Example 4. A reference example 4 related to the present invention is a microprocessor 1
The timing at which the 1-1 or the CPU 12-1 detects a fault is not detected every time a fault occurs, but the fault detection timing is only performed every T3 hours to suppress the detection frequency, and the trace timing is also generated every occurrence monitoring period. This is different from Reference Example 1 above.

FIG . 15 is a flow chart for explaining the fault detection processing and detection interrupt mask timer processing according to the reference example 4 . The fault detection process shown in FIG. 15A corresponds to the fault information generation process (step ST1) shown in FIG. In this process,
When a fault detection interrupt occurs (step ST301,
Detection step), it is determined whether or not the detection interrupt is being suppressed (step ST302, detection step). If the detection interrupt is not suppressed, nothing is done, but if the detection interrupt is being suppressed, the detection interrupt of the corresponding fault is masked (step ST303, detection step), and T3 is set to the initial value of the detection interrupt mask timer. While setting (step ST304, detection step), the detection interrupt mask timer is activated (step ST305, detection step). The detection interrupt mask timer process shown in FIG. 15B is started when the detection interrupt mask timer expires (step ST401, trace step), and the detection interrupt mask of the corresponding fault is released (step S).
T402, trace step).

As described above, according to the fourth reference example , as the timing of detecting a fault, the fault detection timing is only performed every T3 hours, and the detection frequency is suppressed.
Since the trace timing is also set for each occurrence monitoring period, it is possible to reduce the load of the detection process for a failure that is more frequently occurring and the occurrence tendency is more important for failure analysis than the detailed occurrence time.

Reference Example 5. Different from Reference Example 1 described above, Reference Example 5 related to the present invention includes a failure tracing method setting table 9. In this failure tracing method setting table 9, whether failure tracing is performed for each monitoring cause for each failure factor, It is possible to selectively set whether to trace every occurrence detection or not to repeat the failure detection which is repeated until the device is reset once the failure detection method setting table 9 indicates that the maintenance tool 10 is set. It is possible to change it at any timing from the user.

FIG . 16 is a conceptual diagram showing the structure of the fault tracing method according to the reference example 5 . 16 that are the same as or equivalent to those in FIG. 2 are assigned the same reference numerals and overlapping explanations are omitted. In the figure, 9 indicates setting items for various fault tracing methods (whether the fault that has occurred is a trace target, a trace every time, a trace execution only once until reset, or a trace suppression target, as shown in FIG. 16). Is a maintenance tool that can update the failure trace method setting table 9 at any timing.

[0059] Next, the operation will be described. Figure 17 is for reference
11 is a flowchart for explaining the processing operation of the fault tracing method according to example 5 .

When a failure occurs in the ATM communication system (step ST1), the microprocessor 11-1 or the CPU 12-1 collects the failure information that has occurred. Writing to the fault information buffer The fault information collected by the writing unit 1 is written to the fault trace buffer 2 (step S
At the same time as T2), the failure information is output to the failure trace identifying section 5.

[0061] fault trace identification unit 5, it is determined whether the failure to be traced from the type of disorder when receiving the fault information (in which those variously changed according to the type system) ( Step ST3). (A) When it is not a failure to be traced When the failure trace identifying unit 5 determines that it is not a failure to be traced, the processing is terminated without performing any processing.

[0062] When the (B) If a fault should be traced fault trace identification unit 5 is determined to be failure to be traced, the fault trace identifying unit 5 outputs the failure information to the trace inhibit control unit 6. Trace suppression control unit 6
When receiving the fault information output from the fault trace identifying unit 5, the device refers to the fault tracing method setting table 9 and determines whether or not this fault is not detected until device reset (whether it is a trace target or not) ( Step ST4
F).

[0063] (I) is the trace inhibit control unit 6 if applicable to not detected until reset, if the fault output from fault trace identifying unit 5 determines that corresponds to no detected until system reset, the corresponding fault detection mask After applying (step ST5F), the neighboring trace unit 8 sets the failure occurrence count to 1 (step ST6).
F), the current time is set as the failure occurrence time (step S
T7F), the occurrence frequency and the occurrence time of the fault are written in the contents of the fault trace buffer 2, and the error trace file writing unit 3 writes the fault occurrence count and the fault occurrence time together with the fault information in the fault trace buffer 2. Is written to the fault trace file 4 (step ST8F). Then, the process of the occurrence monitoring timer 7a is activated (started) (step ST9F).

(II) In the case where it does not correspond to not detecting until reset : When the trace suppression control unit 6 determines that the failure output from the failure trace identifying unit 5 does not detect until device reset and does not correspond, the trace The suppression control unit 6
By referring to the fault tracing method setting table 9, it is determined whether or not this fault is traced every time (step ST
10F).

[0065] (1) the trace inhibit control unit 6 if applicable to trace each time, when it is determined that corresponds to trace each, near the trace portion 8 is thereby set to 1 failure count (step ST6F), disorders The present time is set as the occurrence time (step ST7F), the error trace file writing unit 3 writes the error trace file (step ST8F), and starts (starts) the process of the occurrence monitoring timer 7a (step S).
T9F).

[0066] (2) if not applicable trace inhibit control unit 6 to trace each time, if it is determined not to correspond to trace each, then the trace inhibit control section 6 refers to the fault trace method setting table 9 Then, it is determined whether or not this failure is in the trace inhibition (step ST11).
F).

[0067] If the case is not in (a) trace deterrence is not in trace suppression (that is, if the failure of the first generation), the vicinity of the trace unit 8, 1 to failure count
Is set (step ST12F), the failure occurrence time is set to the current time (step ST13F),
The number of times the error occurred and the time at which it occurred
And the failure trace file writing unit 3 writes the content of the failure trace buffer 2 in which the failure occurrence count and the occurrence time are written in the failure trace file 4 together with the failure information (step ST14F).
Then, the process of the occurrence monitoring timer 7a is activated (started) (step ST15F). The process of the occurrence monitoring timer 7a is performed as shown in FIG. 4, as in the first reference example .

[0068] Thereafter, since a failure to release the trace inhibited state of the trace inhibit control unit 6 when a predetermined time (T2) does not occur, set T2 to the initial value of the non-occurrence monitoring timer 7b (step ST16F), non-occurrence The monitoring timer 7b is activated (step ST17F). The process of the non-occurrence monitoring timer 7b is performed as shown in FIG. 5, as in the first reference example .

[0069] (b) in the case when it is being traced deterrence is being traced suppression, without direct tracing, updates the failure count. That is, the failure occurrence count is added by the new failure occurrence count (step ST18F). Then, only the current time is set as the failure occurrence time (step ST19F). The processing of generating the monitoring timer 7a when it is being traced suppression, similar to Reference Example 1 is performed as shown in FIG.
After that, as in the case where the trace suppression is not being performed, T2 is set to the initial value of the non-occurrence monitoring timer 7b (step ST1
6F), the non-occurrence monitoring timer 7b is started (step S
T17F).

As described above, according to this reference example 5 , by providing the fault tracing method setting table 9 storing the setting items of various fault tracing methods, the selective setting is possible, and this fault tracing method is also possible. Setting table 9
Can be changed at any timing from the user via the maintenance tool 10, so that a trace method that considers the occurrence of a failure can be flexibly selected, and the failure analysis capability is not reduced.
The overall fault trace processing load can be reduced and the trace capacity can be reduced.

In each of the above reference examples and each of the above embodiments, the case where the fault tracing method is applied to the ATM communication system has been described, but the present invention is not limited to this, and various systems can be used. It is also possible to apply to.

[0072]

As described above, according to the first aspect of the present invention, a detecting step for detecting a fault occurring in the system, a tracing step for tracing the content of the fault detected by the detecting step, and the above detecting If the fault detected by the step repeatedly occurs continuously, the trace inhibiting step includes a trace inhibiting step for inhibiting the content of the error traced by the trace step.
In the first step, the failure that occurred first is
Trace the time of the harm, and after the first failure
The failure that occurred within the specified monitoring period is the failure content and failure occurrence.
Number of times and latest failure occurred within the above specified monitoring period
The time is traced, and the fault that occurred continuously is
Occurred within the monitoring period, which is obtained by multiplying the above predetermined monitoring period by n times.
The failure details, the number of occurrences, and the latest failure occurrence time of the failure are displayed.
It is configured to race. Thus, it filled with information of the failure that the trace region occurs frequently, to prevent the information when generated in the duration and the first fault disappears, prevent hinders cause analysis of network failure it not only can, the monitoring period exponential function
Since it can be increased numerically, once it occurs
The trace area is further suppressed for failures with a long continuous period.
There is an effect that can e Rukoto.

[0073] According to the second aspect of the invention, it traces depression
The stop step causes the trace step to
Harm causes the content of the failure and the time when the failure occurred to be traced.
The failure that occurred within the specified monitoring period after the failure occurred,
The nature of the failure, the number of times the failure occurred, and the
Trace the oldest failure time that occurred and continue
The failure that occurred was monitored by multiplying the above specified monitoring period by n times.
Fault details, number of occurrences, and oldest faults that occurred during the viewing period
Trace the time when the damage occurred and the time when the latest failure occurred.
It is configured as Therefore, the trace area frequently
When the failure continues, it is filled with information on the failure that occurs.
To prevent the loss of information during
To prevent it from interfering with the cause analysis of
Not only can it fail during long trace suppression periods
Is occurring on average or in a burst
Information that can be obtained for a long time
The trace area can be suppressed even for harm.
In addition, there is an effect that high failure analysis performance can be exhibited.
It

[0074] According to the third aspect of the present invention, further, superintendent
The upper limit is set for the viewing period, and the monitoring period reaches the upper limit.
When it reaches the upper limit, it is fixed to the upper limit.
The control period becomes extremely large and is not useful for failure analysis.
Can be prevented in advance, and the timer H
Increase in H / W cost by increasing the count value of / W
This has the effect of suppressing the increase.

[0075] According to the fourth aspect of the present invention, further, search
The fault detected by the outgoing step no longer occurs for the specified period.
If you want to cancel the trace suppression
Since it also has a backup, if the failure does not occur continuously
By removing the trace suppression state,
Detailed obstacle details if the race area is not full
(Fault information) can be traced and the cause of fault occurrence
The effect is that the investigation of can be performed as usual.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an ATM communication system.

FIG. 2 is a conceptual diagram showing a configuration of a fault tracing method of reference example 1 related to the present invention.

FIG. 3 is a flowchart for explaining the processing operation of the fault tracing method according to the first reference example .

FIG. 4 is a flowchart for explaining an occurrence monitoring timer processing operation according to the first reference example .

FIG. 5 is a flowchart for explaining a non-occurrence monitoring timer processing operation according to the first reference example .

FIG. 6 is a diagram for explaining a case in which the number of times of failure occurrence and the time of failure occurrence are traced during trace suppression.

FIG. 7 is a diagram showing an example of tracing by the fault tracing method according to the first reference example .

FIG. 8 is a flowchart for explaining the processing operation of the fault tracing method of the second reference example related to the present invention.

FIG. 9 is a diagram for explaining a case where the number of times of failure occurrence and the time of failure occurrence are traced during trace suppression.

FIG. 10 is a flowchart for explaining the process of the occurrence monitoring timer according to the first embodiment of the present invention.

11 is a diagram showing an example of tracing the fault trace method according to the first embodiment of the present invention.

FIG. 12 is a flowchart for explaining the processing operation of the fault tracing method according to the second embodiment of the present invention.

FIG. 13 is a flowchart for explaining processing of an occurrence monitoring timer according to the third embodiment of the present invention.

FIG. 14 is a flowchart for explaining a process of an occurrence monitoring timer according to a reference example 3 related to the present invention.

FIG. 15 is a flowchart for explaining a fault detection process and a detection interrupt mask timer process according to a reference example 4 related to the present invention.

FIG. 16 is a conceptual diagram showing the configuration of a fault tracing method according to Reference Example 5 related to the present invention.

FIG. 17 is a flowchart for explaining the processing operation of the fault tracing method according to the fifth reference example .

FIG. 18 is a conceptual diagram showing a configuration of a conventional fault tracing method.

[Explanation of symbols]

1 failure trace buffer write section, 2 failure trace buffer, 3 failure trace file write section, 4 failure trace file, 6 trace suppression control section, 7 monitoring timer, 7a occurrence monitoring timer (monitoring timer), 7b no occurrence Monitoring timer (monitoring timer), 8 near trace part, 9 fault tracing method setting table, S
T1, ST2, ST301 to ST305 Detection step, ST7, ST105, ST13E, ST14E, S
T401, ST402 Trace step, ST3-S
T6, ST8, ST11, ST12, ST101-ST
104, ST106 to ST108, ST12A, ST1
3A, ST107B, ST108B, ST6C, ST1
2C to ST14C, ST107D, ST109D Trace suppression step, ST9, ST10, ST201 to S
T203 Trace suppression cancellation step.

Continuation of front page (56) Reference JP-A-8-69395 (JP, A) JP-A-5-95357 (JP, A) JP-A-9-65458 (JP, A) JP-A-7-321782 (JP , A) (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04L 12/26 G06F 11/34 H04L 29/14

Claims (4)

(57) [Claims] 1. A detection step for detecting a fault occurring in the system, a trace step for tracing the content of the fault detected by the detection step, and a case where the fault detected by the detection step repeatedly occurs continuously. , fault trace method odor and a trace suppression step of suppressing the fault contents above trace step traces
The trace inhibiting step includes the trace step.
First, the failure that occurred first is the content of the failure and the time when the failure occurred.
Traced to the specified monitoring period after the first failure.
The failures that occurred in the interval are the details of the failure, the number of failures, and
Trace the latest failure occurrence time that occurred within the specified monitoring period
The above-mentioned specified monitoring
Failure of a failure that occurred within the monitoring period that is n times as long as the visual period
Trace contents, number of occurrences, and latest failure occurrence time
A fault tracing method characterized by the above. 2. A detection step of detecting a fault occurring in the system, a trace step of tracing the content of the fault detected by the detection step, and a fault detected by the detection step in a case where the fault repeatedly occurs continuously. , fault trace method odor and a trace suppression step of suppressing the fault contents above trace step traces
The trace suppression step is
First, the failure that occurred first is the content of the failure and the time when the failure occurred.
Traced to the specified monitoring period after the first failure.
The failures that occurred in the interval are the details of the failure, the number of failures, and
Trace the oldest failure occurrence time that occurred within the above specified monitoring period
The above-mentioned specified monitoring
Failure of a failure that occurred within the monitoring period that is n times as long as the visual period
Content, number of occurrences, oldest failure occurrence time, and latest failure occurrence
Failure characterized by tracing the time
How to trace. 3. An upper limit value is set for the monitoring period, and the monitoring period is set.
Is fixed to the upper limit when
The fault trace according to claim 1 or 2, characterized in that
Method. 4. The fault detected by the detecting step is for a predetermined period.
A trace that cancels the trace suppression when it no longer occurs.
And a step of releasing the source deterrence.
The obstacle tray according to any one of claims 1 to 3.
Way.