JPH02236738A - Fault diagnosis processing system - Google Patents

Abstract

PURPOSE:To execute diagnosis without lowering the efficiency of the data processing of a system in current use by developing data for diagnosis consisting of a pair of execution processor name and the data for diagnosis on two processors, and executing the data for diagnosis according to the same load by taking synchronism mutually on both processors. CONSTITUTION:A main processor 1 and an opposite processor 2 are provided with diagnostic data storage means 20, diagnosis execution requesting means 40, diagnosis execution transmission means 50, and fault analyzing means 60, respectively according to the same constitution. The data 30 for diagnosis for inter-processor communication control equipment consisting of the pair of execution processor name and data for diagnosis is developed on both processors commonly. And the analysis of the fault of the inter-processor communication control equipment is executed by executing developed data 30 for diagnosis according to the same load as taking the synchronism mutually by both processors. Thereby, the diagnosis of the inter-processor communication control equipment is executed without lowering the efficiency of the data processing in current use being operated.

Description

[Detailed Description of the Invention] [Summary] Regarding a fault diagnosis processing method for an interprocessor communication control device used in a multiprocessor system adopting a duplex configuration, it is possible to perform diagnosis without reducing the data processing efficiency of the active system. With the aim of but,
When the processor is not displayed by the specified flag,
Diagnostic execution requesting means for requesting the executing processor to execute the diagnostic data via the active communication line;
When this execution request is made, the diagnostic data is executed to its own interprocessor communication control device, and, on the condition that the requesting processor is in reception request mode, the diagnostic data is sent via the active communication line. Diagnosis execution transmitting means transmits data related to diagnosis to the request source, and from this received data and status data of its own interprocessor communication control device,
and a failure analysis means for performing failure analysis of the inter-processor communication control device. [Industrial Application Field] The present invention relates to a fault diagnosis processing method for an inter-processor communication control device used in a multiprocessor system having a duplex configuration, and in particular, to reduce the load on the currently operating system as much as possible. The present invention relates to a failure diagnosis processing method that enables failure analysis of a standby interprocessor communication control device that is in a failure state.

For example, in a telephone exchange system, a configuration is adopted in which network control is executed according to a multiprocessor system having a duplex configuration. In a multiprocessor system that adopts such a duplex configuration, when a failure occurs in the interprocessor communication control device that controls communication between processors, the active system is immediately switched and predetermined data processing continues according to the switched active system. As we continue to
Communication system between standby processors in failure state? ! ! l Equipment failure analysis will be performed. In such failure diagnosis processing, it is necessary to take measures to reduce the load placed on the currently operating system as much as possible. [Prior Art] FIG. 8 shows the system configuration of a multiprocessor system that adopts a duplex configuration. In the figure, l is the main processor,
2 is the opposing processor. Each of the main processor 1 and the opposite processor 2 includes a central processing unit 3, a main storage device 4, a channel control device 5, and an interprocessor communication control device 6 according to a duplex configuration of an active system and a standby system.

In the figure, the symbols "#O" and "#1" indicate which system each of these devices belongs to. Next, assuming that a failure occurs in the "l-system" inter-processor communication control device 6 (either one or both may occur), a failure diagnosis for the conventional inter-processor communication control device 6 will be performed. We will explain the processing method. In FIG. 9, in order to execute the failure diagnosis process of the conventional technology,
The unit configuration of the diagnostic program that will be deployed on main processor 1 and opposing processor 2 is shown below. In the figure, 7 is an execution control unit deployed in the active system of the main processor 1, which executes control of the entire diagnostic process;
is a data execution unit deployed in the standby system of the main processor l, and according to diagnostic data from the execution control unit 7,
A device for diagnosing the inter-processor communication control device 6 on the main processor side which is in a failure state, 9 is a data transfer unit deployed in the active system of the opposing processor 2,
A device that executes data transfer with the execution control unit 7, lO
is a transfer data execution unit deployed in the standby system of the opposite processor 2, which diagnoses the inter-processor communication control device 6 on the opposite processor side in a failure state according to the transfer data sent from the execution control unit 7. This is what executes the following. In the conventional technology in which a program is developed in this way, the execution control unit 7 first passes data necessary for diagnosis to the data execution unit 8, and the data execution unit 8 executes the following data according to the received diagnostic data. A diagnostic process is executed for the interprocessor communication control device 6 to be diagnosed. Subsequently, the data execution section 8 receives the response result from the interprocessor communication control device 6 and instructs the execution control section 7 to transfer the necessary data. Upon receiving this instruction, the execution control unit 7 transfers the requested data to the data transfer unit 9 via the active communication line, and the transfer data execution unit 10 performs the following according to the received transfer data: A diagnostic process is executed for the interprocessor communication control device 6 to be diagnosed. Subsequently, upon receiving the response result of the inter-processor communication control device 6 from the transfer data execution unit 10, the data transfer unit 9 transfers the data to the execution control unit 7 via the active communication line, and the execution control unit 7 passes this transferred data to the data execution unit 8. Finally, the data execution unit 8 diagnoses the failure by checking the data from both interprocessor communication control devices 6, and if necessary, the execution control unit 7 processes to display the diagnosis results. ．． As described above, in the conventional technology, the main processor l takes the initiative in the diagnostic processing and executes everything such as setting data necessary for diagnosis and checking responses from the interprocessor communication control device 6. was taken.

[Problem to be solved by the invention]

However, with such conventional technology, there is a problem that the program size for fault diagnosis developed in the main processor 1 increases, which increases the burden on the main processor 1. There was a problem in that the amount of data transferred between the two devices was large, resulting in a long diagnosis time.

This caused the problem that the load on the currently running system would increase and the efficiency of data processing would drop significantly. The present invention has been made in view of the above circumstances, and is designed to reduce the load on the currently operating system in the failure diagnosis process of the interprocessor communication control device of a multiprocessor system.
The purpose is to provide a fault diagnosis processing method that makes it possible to perform the following tasks. [Means for Solving the Problems] FIG. 1 is a diagram showing the basic configuration of the present invention.

In the figure, the same components as those explained in FIG. 8 are indicated by the same symbols, and 6a is the inter-processor communication control device on the active side, and 6b is the inter-processor communication control device on the standby side which is in a failure state. represents. 20 is a diagnostic data storage area in which diagnostic data 30 is stored. The stored diagnostic data 30 has a designation flag 31 that displays the name of the processor on the execution side and diagnostic data 32 to be executed on the interprocessor communication control device 6b as basic units, and these basic units are arranged in the order of execution. It consists of Reference numeral 40 denotes a diagnosis execution requesting means, which itself requests the designation flag 3.
5, which requests the execution side processor to execute the diagnostic data 32 forming the basic unit via the communication line of the active system when the processor is not displayed in 1;
0 is a diagnostic execution transmitting means, which is a diagnostic execution requesting means 40.
When there is an execution request from the processor, the diagnostic data 32 forming the basic unit is executed to its own inter-processor communication control device 6b, and on the condition that the requesting processor is in the reception request mode, the current system via the communication line,
60 is a failure analysis means that transmits the data related to the diagnosis to the requesting processor, and when it receives the data from the diagnosis execution transmission means 50 via the active communication line, it A failure analysis of the inter-processor communication control device 6b is executed based on the status data of the inter-processor communication control device 6b.

In the present invention, a main processor 1 and a counter processor 2
Each of them has the same configuration as the diagnostic data storage area 20, the diagnostic execution request stage 40, and the diagnostic execution transmitting means 50.
and failure analysis means 60, and the same diagnostic data 30 is expanded to the diagnostic data storage area 20 of both the main processor 1 and the opposite processor 2.

[Effect]

In the present invention, when reading out the basic units of the diagnostic data 30 one after another, if it is determined that the designation flag 3l does not display its own processor name, the diagnostic execution requesting means 40 requests a Communication control device 6a
After making a request to the diagnostic execution transmitting means 50 of the processor on the execution side to execute the diagnostic data 32 following the designation flag 31, the process is performed to enter the reception request mode. Upon receiving this execution request, the diagnostic execution transmitting means 50 of the processor on the execution side, that is, the processor indicated by the designation flag 3l, sends the diagnostic data 32 requested for execution to its own inter-processor communication in the failure state. rli! to the control device 6b, and subsequently indicates that the requesting processor is in receive request mode. After the diagnosis is confirmed, data related to the diagnosis is transmitted to the requesting processor via the active inter-processor communication system 1B device 6a. Upon receiving this data, the failure analysis means 60 of the requesting processor determines the cause of the failure of the inter-processor communication control device 6b based on the received data and the status data of the inter-processor communication control device 6b which is in its own failure state. Run the analysis. As described above, in the present invention, the diagnostic data 30 for the inter-processor communication control device, which is composed of a pair of execution processor name and diagnostic data, is commonly expanded to both processors, so that both processors can communicate with each other. By executing this expanded diagnostic data 30 according to equal loads while maintaining synchronization, a failure of the interprocessor communication control device is analyzed. From now on, by significantly reducing the amount of data transferred required for fault diagnosis, it will be possible to significantly shorten diagnosis time, and
This eliminates the possibility of the load being biased towards one processor or the other. Therefore, it is possible to diagnose the interprocessor communication control device without reducing the efficiency of data processing in the currently active system.

〔Example〕

The present invention will be explained in detail below according to an embodiment applied to a multiprocessor system of a telephone exchange system. Figure 2 shows the system configuration of a telephone exchange system to which the present invention is applied. As shown in this figure, a plurality of opposing processors 2 are configured to perform distributed control of the telephone exchange system by executing switching processing for the network 19 that each is responsible for under control from the main processor 1. It will be done. In the telephone exchange system configured as described above, in order to cope with a system down due to a failure of the inter-processor communication control device 6, the main processor 1 and the opposite processor 2 adopt a duplex configuration, and each opposite processor 2 normally performs the following functions. Control of each network 19 is executed according to the active system, and when a failure occurs in the interprocessor communication control device 6, the standby system is switched to the active system and control of the network 19 is continued. I'm going to go. and,
When a failure occurs, the fault diagnosis processing of the inter-processor communication control device 6 is executed using the switched active communication line. FIG. 3 shows the configuration of an embodiment of the fault diagnosis processing of the inter-processor communication control device 6 according to the present invention. This figure shows the unit configuration of a diagnostic program that will be deployed in the main processor 1 and opposing processor 2 in order to execute the failure diagnosis process of the present invention. In the figure, No. 9
Components that perform the same functions as those explained in the figures are indicated by the same symbols, and as explained in FIG.
Let b represent the interprocessor communication control device on the standby side that is in a failure state. Reference numeral 11 denotes a reception request flag developed in the memory of the active system of the main processor 1 and the opposite processor 2, and when the data related to failure diagnosis is in the reception request state, the flag is set to sesoto by the own execution control unit 7. 12 is a flag monitoring unit deployed in the active systems of the main processor 1 and the opposing processor 2, and is activated by the execution control unit 7 of the opposing processor. , and monitors the flag status of the reception request flag 1l. As shown in this figure, in the present invention, the execution control unit 7 that controls the diagnostic processing is configured to be deployed equally in the active systems of both the main processor 1 and the opposing processor 2, and According to the diagnostic data from the section 7, the inter-processor communication control device 6b is in a failure state.
The data execution unit 8 that executes diagnostic processing on the standby system of both the main processor 1 and the opposing processor 2
It is characterized by being configured so that it is expanded equally. Fourth
The figure shows the data structure of diagnostic data 30 included in the present invention to enable such a configuration. As shown in this figure, in the present invention, diagnostic data 30 required for executing a failure diagnosis of the interprocessor communication control device 6b in a failure state is stored in a designation flag 31 indicating the name of the processor on the execution side. and diagnostic data 32 specifying specific diagnosis contents as basic units, and these basic units are arranged in the order of execution. To explain specifically using the example of FIG. 4, the designation flag 31 of ■ is set to "1°," the diagnostic data 32 of ■ is set to "write command," and the designation flag 31 of ■ is set to "1°." Since “θ” is set and a “read command” is set in the diagnostic data 32 of
The data structure is such that a "read command" is executed to the inter-processor communication control device 6b on the opposing processor 2 side. And, in the present invention,
The diagnostic data 30 having such a structure is commonly expanded to both the main processor 1 and the opposing processor 2. In this way, in the present invention, the main processor 1 and the opposing processor 2 are configured to have exactly the same diagnostic data 30 for fault diagnosis. Next, the operational processing of the present invention will be explained in detail according to the flowcharts shown in FIGS. 5 to 7. Here, the flowchart in FIG. 5 is a flowchart executed by the data execution unit 8, the flowchart in FIG. 6 is a flowchart executed by the execution control unit 7, and the flowchart in FIG. 7 is executed by the flag monitoring unit 12. This is a flowchart.

As shown in the flowchart of FIG. 5, the data execution unit 8 that executes the diagnostic process first determines the processor type in the SteSob 1 and determines that its own processor is the main processor 1. In step 2, it assigns "1" as a processor variable, and when it determines that its own processor is the opposing processor 2, it executes a process in which it assigns "O" as a processor variable in step 3. Then proceed to step 4,
Diagnostic data 3 to be processed given from the execution control unit 7
It is determined whether the designation flag 31 of 0 indicates the processor variable of the processor itself or, conversely, the processor variable of the other party.

Based on the judgment in step 4, the specified flag 3 to be processed
1 is displaying the other party's processor variable, this means that the data execution unit 8 of the other party's processor executes the diagnostic data 32 following the specified flag 31. Therefore, the data execution unit 8 , the process proceeds to step 5 and sends a transmission request to its own execution control unit 7, and in the subsequent step 6, the diagnostic data 32 up to the next specified flag 31 is skipped without being executed. On the other hand, when it is determined that the specified flag 31 to be processed indicates its own processor variable, it means that it executes the diagnostic data 32 following the specified flag 31, so the data execution unit 8 Proceeds to step 7, sends a reception request to its own execution control unit 7, and after accepting the request from the other party, executes the diagnostic data 32 up to the next specified flag 31 in step 8. It turns out. Finally, in step 9, it is determined whether or not all of the diagnostic data 30 has been executed, and when it is determined that the execution has been completed, the process is terminated, and when it is determined that the execution has not been completed, the process returns to step 4. Process. That is, the data execution unit 8 of the processor that is not indicated by the designation flag 31 processes to send a startup request to the data execution unit 8 of the other processor, and
When the data execution unit 8 of the processor indicated by the designation flag 3l receives an activation request from the other processor, it processes the diagnostic data 32 following the designation flag 3l to execute it. Next, according to the flowchart in FIG.
The control processing executed by the data execution unit 8 will be explained. When the data execution unit control process is entered as shown in step 1l, it is determined in step 12 whether the request from the data execution unit 8 is a transmission request or a reception request.

When it is determined that this is a transmission request in step 12, it means that the data execution unit 8 of the other party's processor is requested to transmit the execution result of the diagnostic data 32, so the execution control Part 7 follows Step 1
3, the flag monitoring unit 12 of the other party's processor is activated via the active inter-processor communication control device 6a.
When it is confirmed that a reception request flag is sent to the reception request flag 1l of the other processor, the data requested to be transmitted is sent to the execution control unit of the other processor via the active inter-processor communication control device 6a. 7.

On the other hand, when it is determined in step 12 that it is a reception request, it means that the request is to receive data regarding the diagnostic data 32 of the data execution unit 8 of the other party's processor, so the execution control unit 7 continues. In step 14, a reception request flag is set in its own reception request flag 11.

Next, in step 15, a wait for data reception is executed, and in the next step 16, it is determined whether or not data has been received. When it is determined that data has been received, the process proceeds to step 17, and data execution is performed on the received data. Transfer to Section 8.
Finally, in step l8, it is determined whether or not the data execution unit control process has ended, and when it is determined that it has ended, the process is ended, and when it is determined that it has not ended, the process returns to step l2. That is, in accordance with a request from the data execution unit 8, the execution control unit 7 transmits the execution results of the diagnostic data 32, etc. to the data execution unit 8 of the other party's processor via the active communication line, or Processing is performed to receive the execution results of the diagnostic data 32 from the data execution unit 8 of the other party's processor. Next, the processing of the flag monitoring section 12 will be explained according to the flowchart shown in FIG. When activated by the execution control unit 7 of the other party's processor, the flag monitoring unit 12 determines in step 21 whether or not a reception request flag is set in the reception request flag 11. With this judgment,
When it is determined that the reception request flag is not set, it means that the data cannot be received, so the process proceeds to step 22, performs timing processing, and then returns to step 21. On the other hand, step 2
When it is determined that the reception request flag is set, it means that the data can be received, so the process proceeds to step 23, and the data is sent from the execution control unit 7 of the other processor. The received data is passed to the execution control unit 7 of its own processor, and in the following step 24, the reception request flag 11 is cleared and the process is terminated.

That is, when the flag monitoring unit 12 detects that the receiving state is entered, the flag monitoring unit 12 transfers the sent data to the execution control unit 7.

Next, using the diagnostic data 30 shown in FIG. 4 as an example, a diagnostic process for the inter-processor communication control device 6b in a failure state will be specifically described.

If the designation flag 31 of ■ is the designation flag 31 of the processing target, the designation flag of The data execution unit 8 of the opposing processor 2 determines that it is the request receiving side, and determines that it is the non-executable activation request sending side. From now on, the data execution unit 8 of the opposite processor 2 will be connected to the inter-processor communication control device 6 of the active system.
a, executes a startup request to the data execution unit 8 of the main processor l, and also executes the “
Executes the skip process for "write command" and enters the reception wait state according to the specified flag in the next ■.Meanwhile,
Upon receiving this activation request, the data execution unit 8 of the main processor 1, which had entered the reception waiting state due to the designation flag 3l of ■, sends the diagnostic data 32 of ■ Executes a “write command”. That is, the data execution unit 8 of the main processor 1 is in a failed state.
Data is transferred to the data execution unit 8 of the opposing processor 2 via the processor 2.

Subsequently, since the specification flag 31 of ■ is displaying "O", the data execution unit 8 of the main processor 1 determines that it is the sender of the non-execution activation request, and executes the data of the opposite processor 2. The unit 8 determines that it is the receiving side of the execution start request. From now on, the data execution unit 8 of the main processor 1 will be connected to the active inter-processor communication control device 6a.
, executes a transmission activation request to the data execution unit 8 of the opposing processor 2, and also sends the diagnostic data “
Executes the skip process for the read command and enters the reception wait state according to the following specified flag. On the other hand,
Upon receiving this transmission activation request, the flag monitoring unit 12 of the opposite processor 2 monitors the reception request flag 11 and confirms that the data execution unit 8 of the opposite processor 2 is in a reception waiting state, and then the active processor A startup request sent from the main processor 1 via the intercommunication control device 6a is passed to the data execution section 8. Then, in accordance with the activation command, the data execution unit 8 of the opposite processor 2 performs the following for the inter-processor communication control device 6b in the failed state:
The "read command" of the diagnostic data 32 in (2) is executed. In other words, the data execution unit 8 of the opposing processor 2 performs the reading process of the data sent by the 1 write command (2). The data execution unit 8 of the processor 2 compares this reading result with the sent correct reading value to analyze whether or not the inter-processor communication control device 6b in the failure state is malfunctioning. It will be processed as follows.

In this way, in the present invention, the main processor 1 and the opposite processor 2 can use the commonly developed diagnostic data 3.
0 and follow the instructions displayed with the specified flag 31.
By sequentially executing this diagnostic data 30 in synchronization with each other, a failure analysis of the inter-processor communication control device 6b in a failure state is executed. Although the illustrated embodiments have been described above, the present invention is not limited thereto. For example, the present invention is not limited in its application to multiprocessor systems in telephone exchange systems. Further, for convenience of explanation, a specific example has been described using diagnostic data of a write/read command, but the present invention is not limited to this.

〔Effect of the invention〕

As described above, according to the present invention, diagnostic data for an inter-processor communication control device consisting of a pair of an execution processor name and diagnostic data is expanded to both processors,
Both processors synchronize with each other and execute the developed diagnostic data according to the same load, thereby analyzing the failure of the interprocessor communication control device. From now on, by significantly reducing the amount of data transferred required for fault diagnosis, diagnosis time can be greatly shortened, and the load will no longer be biased towards one processor or the other. Therefore, it is possible to diagnose the interprocessor communication control device without reducing the efficiency of data processing in the currently active system.

[Brief explanation of drawings]

Fig. 1 is a diagram of the principle configuration of the present invention, Fig. 2 is a system configuration diagram of a telephone exchange system to which the present invention is applied, Fig. 3 is a diagram of the configuration of an embodiment of the present invention, and Fig. 4 is a diagnostic town of the present invention. Zi-Yu structural diagram, Figures 5 and 6
7 and 7 are flowcharts executed by the present invention, FIG. 8 is a system configuration diagram of a multi-processor system adopting a duplex configuration, and FIG. 9 is an explanatory diagram of the prior art. 1 is a main processor, 2 is an opposing processor, 6 is an inter-processor communication control device, 7 is an execution control section, 8 is a data execution section, 11 is a reception request flag, 12 is a flag monitoring section, 2
0 is a diagnostic data storage area, 30 is diagnostic data, 3l is a designation flag, 32 is diagnostic data, 40 is a diagnostic execution requesting means, 50 is a diagnostic execution transmitting means, and 60 is a failure analysis means. Patent applicant Fujitsu Limited (one other person) representative
Person Patent attorney Hiroshi Morita (2 others) The present invention is applicable to the telephone communication remote system system, Kazuo Ina (Fig. 2) Sae, 46th day of the moon's first look now wr month of the month of the year of the year of the year of the year Flow ÷ Yard (I System 4 Mangaru Figure 8

Claims (1)

[Claims] In a fault diagnosis processing method for diagnosing a fault occurring in an inter-processor communication control device (6) that controls communication between a main processor (1) and an opposite processor (2), Diagnostic data (30) is prepared by using a specified flag that displays the processor name and diagnostic data to be executed for the interprocessor communication control device (6) as basic units, and arranging these basic units in the order of execution. and is configured to be deployed to both the main processor (1) and the opposite processor (2), and the main processor (1) and the opposite processor (
When each of 2) is a processor that is not indicated by the above specified flag, it requests the executing processor to execute the diagnostic data forming the above basic unit via the active system communication line. Diagnosis execution request means (40
), when this execution request is made, the diagnostic data forming the basic unit is executed to its own inter-processor communication control device (6) which is in a failure state, and the requesting processor is in the reception request mode. A diagnostic execution transmitting means (50) that transmits data related to the diagnosis to the requesting processor via the active communication line, and a diagnostic execution transmitting means (50) that receives data from the diagnostic execution transmitting means (50). , its own inter-processor communication control device (6
), the inter-processor communication control device (6
) a failure analysis means (60) for performing a failure analysis of the failure diagnosis processing method.