JPH0352048A - Automatic resending processing system for message - Google Patents

Abstract

PURPOSE:To reduce the load for realizing a fault tolerant function upon a person who makes a message communication by providing a kernel with an automatic resending processing part, a return message monitoring processing part, and a loop detection processing part. CONSTITUTION:The automatic resending processing part 21 recognizes the occurrence of abnormality in a reception space 3 and resends a request message, sent previously from a transmission space 1, to the reception space 3. The return message monitoring processing part 22 selects only correct return message sent from the reception means 3 and sends them to the transmission space 1 when resending is performed. The loop detection processing part 23 detects a loop at the time of the resending operation. Consequently, the load for realizing the fault tolerant function upon the person who makes a message communication is reduced.

Description

[Detailed Description of the Invention] [Summary] Regarding the automatic message retransmission processing method by the kernel of the operating system, the purpose of this invention is to reduce the burden on those who communicate messages by realizing a fault-tolerant function. Sending a request message to the receiving space and sending a reply message from the receiving space to the sending space via a kernel, and retransmitting the request message to the receiving space when an abnormality occurs in the receiving space. In the data processing device that is used, an automatic retransmission processing unit and. Reply message monitoring processing unit. A loop detection processing unit is provided in the kernel. The automatic retransmission processing unit verifies the occurrence of the abnormality in the reception space and retransmits the request message that was previously transmitted from the transmission space to the reception space, and the reply message monitoring processing unit When the retransmission is performed, a correct one is selected from among the reply messages transmitted from the reception space and transmitted to the transmission space. The loop detection processing unit is configured to detect the loop during the retransmission. [Industrial Application Field] The present invention relates to an automatic message retransmission processing method. This paper concerns the automatic message retransmission processing method by the operating system kernel. Data is sent and received in the data processing system in the form of five messages. Therefore, a fault tolerant (FT) function has been implemented for sending and receiving messages, that is, a function that continues to provide service to the network without stopping even if an abnormality occurs in the system. [Prior Art] Figure 4 is an explanatory diagram of the prior art and shows the conventional message retransmission process. In Fig. 4, 1 is the transmission space. 2 is the kernel and 3 is the reception space. 31 is a working space of the reception space 3. 32 is a standby space in the reception space 3. A sending side, such as a processing program or process, ie, a sending space 1, sends a message requesting processing (request message) to a receiving side, such as another processing program or process, via the kernel 2 of the operating system. j In particular, it is transmitted to the currently active space (working space 31) among the receiving spaces 3 (
(■ in the figure). After this, if an abnormality such as a crash occurs in the current space 3l during message processing (■ in the figure), the current space 31
Kernel 2, which is monitoring
．． Furthermore, due to the occurrence of a crash, the Ijl space 31 switches to the standby space 32, which is the receiving space 3 paired with it (duplexed), and becomes active (■ in the figure). Sending space 1, which learned of the occurrence of the crash, sends a request message via kernel 2. It is retransmitted to the standby space 32, which is the currently active receiving space (■ in the figure). The standby space 32 receives the request message. Upon completion of the process, a reply message is sent to sending space l via kernel 2 (■ in the figure), and sending space 1 receives this (■ in the figure). As described above, the fault-tolerant function against crashes in the working space 31 is mainly realized by the transmitting space 1. Also. If the cause of a crash in the active space 31 is a bug in the processing program on the receiving side, such as an input-dependent bug, the same crash will occur in the standby space 32,
It cannot be processed normally. As a result, an (infinite) loop may occur where the request message is repeatedly retransmitted or clanked. Therefore, loop detection processing is performed before retransmission (O in the figure). When the number of loops exceeds a predetermined value, appropriate processing is performed to eliminate the loop. Therefore, a fault-tolerant function for the loop is realized by the transmission space l. (Object to be Solved by the Invention [111]) According to the prior art described above, the process (■ in the figure) of retransmitting a request message by recognizing reception space switching based on a crash and crash notification is as follows. Transmission space 1 must do this. on the other hand. The content of the retransmitted request message must be managed by the side that uses it, that is, the receiving side (receiving space 3). Therefore, the receiving side must have a function to check whether the contents of the request message received from the second time onwards are the same as the first one. Also. As mentioned above, there is a possibility that a loop may occur, so it is necessary for the transmission space 1 to detect this loop. Conventionally, like this. Fault-tolerant functions for sending and receiving messages are provided by those who communicate messages. That is, it was realized by a transmission space 1 and a reception space 3 (processing program or process). That is. From the perspective of those who communicate messages, especially from the transmission space l, there is no structure in which request messages are automatically retransmitted for fault tolerance. Therefore, for those who communicate messages. There was a problem in that the burden of implementing fault-tolerant functionality was large. The purpose of the present invention is to provide an automatic message retransmission processing method that reduces the burden on those who communicate messages to realize a fault-tolerant function. [Means for solving the problem] Figure 1 is a diagram showing the principle structure of the present invention. 1 shows a data processing device according to the present invention. In FIG. 1, 1 is a transmission space, 2 is a kernel, 3 is a reception space, and 21 is an automatic retransmission processing unit. 22 is a reply message monitoring processing section, 23 is a loop detection processing section, 31 is a working space in the receiving space 3, and 32 is a standby space in the receiving space 3. The sending space l sends a request message (a message requesting processing) to the receiving space 3 via the kernel 2. The reception space 3 is. When a request message is received, perform the processing requested by the score request message. And once the process is complete. The receiving space 3 transmits a reply message to the request message to the transmitting space 1 via the power channel 2. This will notify you that the process has completed. The receiving space 3 is duplicated in case an abnormality such as a crash occurs. It consists of a working space 3l, which is currently active, and a standby space 32, which is waiting. If an abnormality occurs in the working space 3l of the receiving space 3. After the reception space 3 is switched from the working space 31 to the standby space 32. The request message is retransmitted to the waiting space 32, which is the receiving space 3. The requested process can be completed. Kernel 2 performs message communication services such as sending the above-mentioned request messages and reply messages, and also performs automatic retransmission processing of request messages. For this. Automatic retransmission processing unit 21. A reply message monitoring processing section 22 and a loop detection processing section 23 are provided in the kernel 2. The automatic retransmission processing unit 2l is a working space 3 of the receiving space 3.
Recognize that an abnormality such as a crash has occurred in l, and that this has caused the space to switch as described above.
Then, the request message sent from the sending space 1 to the receiving space 3 before the occurrence of the abnormality is retransmitted to the standby space 32 of the receiving space 3 (which is considered to be the working space). The reply message monitoring processing unit 22. Monitor reply messages. When the request message is retransmitted, the correct reply message sent from receiving space 3 (working space 3l and/or standby space 32), Wi, and retransmitted at the closest time. Select only the reply message to the request message and send it to transmission space 1. The loop detection processing section 23. When retransmitting a request message. Detects the occurrence of an abnormality such as a crash in the reception space 3, or a loop in which request messages are repeatedly retransmitted. [Operation] Kernel 2 is configured such that the transmission space l is the working space 31 of the reception space 3.
It receives the request message sent with the destination as the destination and sends it to the working space 31. At this time. The automatic retransmission processing unit 2l is
Copy and save the request message. If no abnormality such as a crash occurs in the working space 31, the working space 31 completes the requested processing and then sends a reply message to the request message with the sending space 1 as the destination. The reply message monitoring processing unit 22 of the kernel 2 that receives this recognizes that the cough reply message is correct and sends it to the transmission space 1.
When an abnormality such as a crash occurs in the active space 31, the active space 3l is switched to the standby space 32. The automatic retransmission processing unit 21. When it recognizes that an abnormality has occurred and the space has been switched, it sends the stored request message to the standby space 32. Therefore, the identity of the request message is maintained, and no burden is placed on the transmission space. You can resend the request message. During this retransmission. The loop detection processing unit 23 detects loops. Perform necessary processing such as eliminating loops.
In this way, as the automatic retransmission processing section 21 of the kernel 2 performs retransmission, the rube detection processing section 23 of the kernel 2 performs loop detection, so that nine loops are detected reliably, and
The burden on the transmission space 1 can be eliminated. After retransmission, if no abnormalities such as crashes occur in the receiving space 3, the receiving space 3 (working space 31 and/or standby space 32)
A reply message to the retransmitted request message is sent from On the other hand, depending on the timing of an abnormality such as a crash, a reply message may be sent to the request message before the crash. In such cases. The reply message monitoring processing unit 22 selects a reply message to the request message retransmitted at the closest time, and transmits it to the transmission space 1. Therefore, since the sending space 1 receives only one reply message for one request message, there is no need to be aware of any abnormalities in the receiving space 3. in this way. Kernel 2 retransmits the request message for fault tolerance and guarantees the identity of the request message at that time. Transmission space 1
And the burden on the receiving space 3 can be reduced. [Embodiment] Figure 2 is an explanatory diagram of an embodiment, and shows data processing device I. In FIG. 2, 11 is a request message transmission processing section with automatic retransmission (request message transmission processing section), 12 is a reply message reception processing section, 24 is a copy area, and 33 is a reply message transmission processing section. Transmission space 1 and reception space 3 are . In the illustrated data processing device which is a message-based system. A person who communicates messages, such as a processing program or process. Central processing units, terminals, etc. The transmission space 1, which is the transmission side of message communication, includes a 1-request message transmission processing section 11 and a reply message reception processing section 12. The request message transmission processing unit l1 sends the request message with or without automatic retransmission (request) according to the user's specifications, and then performs transmission processing. The former is. For kernel 2. This is used to request automatic retransmission processing for request messages. Therefore, even if Kraftshe occurs in the working space 31 of the receiving space 3 (if the number of loops is small), a reply message can be received. On the other hand, the latter sends only request messages. Therefore, in the case of the above crash. The processing request for reception space 3 is canceled. The reply message reception processing section 12. Even if automatic retransmission processing of the request message is performed, only one correct reply message selected by the reply message monitoring processing section 22 is received. In addition, if a request message without automatic resending is sent, the reply message reception processing unit 12. Receive crash notifications when a crash occurs. This releases the state of waiting for a communication message. The receiving space 3, which is the receiving side of message communication, is redundant so that even if a crash occurs in the working space 31, the standby space 32 can immediately continue its activities. Therefore, the standby space 32 has the same logical structure as the current space 3l, and the reply message transmission processing section 33 exists in the current space 31 as well as the standby space 32. Switching between the working space 3l and the standby space 32 is as follows. This is done by the kernel 2 or operating system. The reply message transmission processing section 33. According to the processing in the space. Send a reply message. Kernel 2 is. It is the core or supervisor of the operating system. The automatic retransmission processing unit 2l. Request message transmission processing unit 1l
Receives the request message sent from and sends it to reception space 3. At this time. It is determined whether this request message includes automatic retransmission or not, and if automatic retransmission is included, the request message is copied and stored in the copy area 24 in the kernel 2. and. After adding the number of retransmissions indicating the number of times the request message has been sent. Send this to reception space 3. For the first transmission, the number of retransmissions is 10. In addition, if the request message does not have automatic resending. The above-mentioned copying and retransmission counts are not added. Furthermore, when a clafsier occurs in the reception space 3 (active space 31), the automatic retransmission processing unit 21 extracts the request message from the copy area 24 and retransmits it to the reception space 3 (standby space 32). At this time. Count up the number of retransmissions by +1 and add this to the request message. During this retransmission. Prior to this, loop detection processing section 2
3 performs loop detection. That is. If the number of retransmissions is 2 or more, a loop may be occurring. Therefore. The loop detection processing unit 23 regards the number of retransmissions, that is, the number of loops, and checks whether this exceeds a predetermined value, and if it does, performs processing to eliminate the loop. The reply message monitoring processing unit 22. It receives the reply message sent from the reply message transmission processing section 33 and sends it to the reply message reception processing section 12. At this time. It is checked whether this reply message is correct as described above. For this purpose, the number of retransmissions in the request message is compared with the number of retransmissions in the reply message. The reply message transmission processing unit 33 adds the number of retransmissions (the number of retransmissions added to the request message to which the reply message corresponds) to the reply message. Send this. The reply message monitoring processing unit 22. 2 obtained in this way
If the two retransmission counts match, the reply message is judged to be correct (for the request message retransmitted at the closest time) and sent to transmission space 1. If the number of retransmissions does not match, the reply message is not sent to transmission space 1. Figure 3 is. This is the message automatic retransmission processing flow. ■ The automatic retransmission processing unit 21. Receive the request message from sending space 1 and check whether automatic retransmission is enabled. ■ The automatic retransmission processing unit 21 copies the request message to the copy area 24 when automatic retransmission is included. Copying is not performed unless automatic retransmission is enabled. ■ The automatic retransmission processing unit 21
When making a request with retransmission, a request message with the number of retransmissions added is sent to the working space 31. ■ The reply message monitoring processing section 22 receives the Crassier notification from the reception space 3 (standby space 32), and notifies the automatic retransmission processing section 21 and the loop detection processing section 23. That is, in the case with automatic retransmission, no crash notification is sent to transmission space 1. If the above-mentioned automatic retransmission is not included, the processing [phase] described below is performed. ■ The loop detection processing section 23. Perform loop detection processing.
Check whether the number of retransmissions exceeds a specified value. ■ If the predetermined value is not exceeded, the automatic retransmission processing unit 21 receives this notification. Count up the number of retransmissions by +1. And after this. Process ■, that is, the request message is retransmitted. If it exceeds the specified value. The process ■ described below is performed. Also. Transmission space 1 receives an error message indicating that the loop has failed. ■ The reply message monitoring processing unit 22. Receive the reply message from reception space 3. ■ The reply message monitoring processing unit 22 checks whether the number of retransmissions of the reply messages matches. If they do not match, terminate the process without performing the following process ■ and [Stock]. Wait for the reply message to be received. ■ If they match. There is no longer a need for retransmission. The copy area 24 is released by the reply message monitoring processing section 22. In addition. Even if the number of retransmissions exceeds a predetermined value in process (2), no further retransmissions are performed and the copy area 24 is released. ■ The reply message monitoring processing unit 22 receives the reply message and transmits it to the destination transmission space 1. Furthermore, if the request message does not have automatic resending, when a crash notification is received, the crash notification message reception process is performed. When a reply message is received, the reception process for the reply message is performed. [Effects of the Invention] As explained above, according to the present invention. When an abnormality occurs in the 5 receiving space during automatic message retransmission processing, the request message is automatically retransmitted by the kernel without making the sending space aware of this. Fault-tolerant functions for message communication can be easily realized, and
This burden can be avoided on the sending space and receiving space for message communication.

[Brief explanation of drawings]

Figure 1 shows the principle structure of the present invention. Fig. 2 is an explanatory diagram of an embodiment, and Fig. 3 is a message automatic retransmission processing flow. Fig. 4 is an explanatory diagram of a conventional technique. 1 is the transmission space, 2 is the kernel. 3 is the receiving space. 21 is an automatic retransmission processing section, 22 is a reply message monitoring processing section, 23
3 is a loop detection processing unit, and 31 is a working space in the receiving space 3. 32 is a standby space in the reception space 3.

Claims (1)

[Claims] Transmission of a request message from the transmission space (1) to the reception space (3) and from the reception space (3) to the transmission space (1)
A reply message to the request message is sent to the receiving space (2) via the kernel (2).
3) in a data processing device that retransmits the request message to the reception space (3) when an abnormality occurs, comprising: an automatic retransmission processing unit (21) that retransmits the request message; and an automatic retransmission processing unit (21) that retransmits the request message; The kernel (2) is provided with a reply message monitoring processing unit (22) that monitors the automatic retransmission, and a loop detection processing unit (23) that detects the occurrence of the abnormality or a loop in which the retransmission is repeated. A processing unit (21) recognizes the occurrence of the abnormality in the reception space (3) and retransmits the request message that was previously transmitted from the transmission space (1) to the reception space (3). , the reply message monitoring processing unit (22) selects a correct reply message from among the reply messages sent from the reception space (3) when the retransmission is performed, and sends it to the transmission space (1). An automatic message retransmission processing method, characterized in that the loop detection processing unit (23) detects the loop at the time of the retransmission.