JPH02135831A - Redundancy system for communication protocol - Google Patents

Abstract

PURPOSE:To keep the consistency of a transmission sequence even if an error occurs in an opposite direction from the result of an event by checking the series of a data sequence number and aborting or processing the data based on the result. CONSTITUTION:A sender station adds a data sequence number to data 1, 2...n in addition to header information having been added to them and sends the result and a receiver station returns ACK/NACK data added with the data sequence number is addition to the header information. When the data sequence number added to the ACK/NACK data is dissident with the data sequence number of the sent data, the sender station aborts the ACK/NACK data. When the data sequence number is equal to or smaller than the data sequence number added to the preceding data, the receiver station aborts the data received at present. Thus, even if the occurrence of error takes place in an opposite direction to the result of event, the consistency of the transmission sequence is kept.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to a distributed processing system in which a plurality of CPUs mutually exchange information through a transmission path, and particularly relates to a distributed processing system in which a plurality of CPUs exchange information with each other through a transmission path. The present invention relates to a communication protocol redundancy method that maintains normal processing functions and the consistency of a database distributed among the CPUs against failures caused by noise or the like.

(Prior Art) FIG. 8 is a functional block diagram of one station configuring this type of distributed processing system. This station mainly consists of a part executed by the main CPU 7n and a part executed by the transmission control CPU 7m, and the general data processing and control are performed as follows.

First, when the user task 7a-1 of the main CPU 7n makes a transmission request, the transmission data and control signal are sent to the transmission task 7.
sent to b. After adding appropriate header information to the transmission data, the transmission task 7b writes it into the transmission buffer 7h, and further activates the transmission driver 7e. The transmission driver 7e transfers the data in the transmission buffer 7h to the local buffer 7j in the transmission control CPU 7m, and further activates the transmission control section 7g in the transmission control CPU 7m. The transmission control unit 7g adds text length control information, CRC redundancy information, etc. in the local buffer 7j to the transferred transmission data, and then sends the D/A converted data to the transmission path 71.

Next, the transmission control unit 7g notifies the transmission driver 7e of a transmission completion interrupt signal when the local buffer 7j becomes empty, and at that timing the transmission driver 7e reads the status information 7k and sends the result via the transmission task 7b. and notifies the user task 7a-1.

On the other hand, in the case of reception, the transmission control unit 7g in the transmission control CPU 7m checks the CRC redundancy information on the A/D converted data every time the received data arrives, and stores it in the local buffer 7j. , if a text end code is detected in the received data, the main CPU 7
Interrupts the reception driver 7f in n.

Therefore, if the status information 7k is normal, the reception driver 7f selects the reception buffer 7 from the local buffer 7j.
1, and then starts the reception task 7C. After defining the data chain structure, the receiving task 7C activates any of the user tasks 7a-2, . . . , 7a-n to transfer the data.

Here, if the status information 7k is abnormal, processing is generally performed according to the following procedure.

If a status information abnormality is detected during the transmission process,
The transmission initialization task 7d is activated to initialize the transmission control CPU 7m, and the transmission driver 7e, reception driver 7f,
Also, sending task 7b.

Initialize the reception task 7C. When detected during the reception process, if the status abnormality is fatal, it is initialized in the same way as the transmission status abnormality, and if not, the received data is discarded. However, user task 7a-2,
..., 7a-n are not notified of the reception status.

In order to continue the transmission procedure in response to errors occurring in the transmission and reception procedures as described above, various redundancy processes are performed at the user task level in the upper layer of the transmission procedure. This redundancy processing will be explained using FIG. 9.

After transmitting the transmission data 1d, the transmission source station 1a (St) checks the transmission completion status 1j, and if this status is abnormal, executes a retransmission procedure. If the transmission completion status is normal, the diagnostic timer 1c (T
) by giving a set signal 1g to set this diagnostic timer IC, the ACK from the destination station is
The time until data 1e or NACK data 1f is received is monitored. In monitoring with the diagnostic timer IC, A
Diagnostic timer IC before receiving CK/NACK data
If the timeout occurs, the retransmission procedure is executed by retry 11 after the timeout occurs.

On the other hand, at the destination station 1b (Sj),
When transmission data 1d is received, the validity of this data and data loss is checked, and if it is normal, ACK is sent.
The data 1e and, in the case of an abnormality, the NACK data 1f are returned to the transmission source station 1a.

FIG. 10 is a time chart showing this.

That is, when the transmitting station 6a in the idle state receives data 6d from the user task, the receiving station 6b in the idle state
At the same time, a set signal 6m is given to the monitoring timer 61 to set it, and the transmission status 6e is also received.

If a transmission status error 6j occurs, the data is retransmitted.

On the other hand, when the receiving station 6b in the idle state receives or accepts data, it generates a data receiving/accepting signal 6g, and transmits either data to the transmitting station 6C waiting for ACK data or NACK data.

Here, if the transmitting station 6c receives the ACK data before the time-up of the monitoring timer 61, it will give the reset signal 6n to the monitoring timer 61 at this point and also send the received signal 6f.
occurs and returns to the idle state. When NACK data is received, a data retransmission signal 6k is generated. Also,
If the monitoring timer 61 times up before receiving or receiving ACK data or NACK data, the monitoring timer 61 generates a data retransmission signal 6o.

On the other hand, when the receiving station that returned the ACK data receives the transmission status 6h, the user program writes the received data 61
and return to idle state.

FIG. 11 is a flowchart showing the processing procedure when transmitting data at the transmitting station. That is, when data is transmitted in step 101, it is determined in step 102 whether the transmission status is normal.

If the sending status is normal, step 1
At step 03, a time monitoring timer is set. On the other hand, if the transmission status is not normal, the number of data transmission attempts is determined in step 104, and the process continues in step 1 until a predetermined value is reached.
The data transmission at step 01 is repeated, and when the data exceeds a predetermined value, transmission error processing is performed at step 105.

FIG. 12 is a flowchart showing the corresponding processing procedure when receiving data at the receiving station. in this case,
When data is received in step 201, step 202
It is determined whether the received data is normal or not. If it is determined to be normal here, the ACK data is sent in step 203.
If not normal, NACK data is set in step 204, and these data are transmitted in step 205. Then, in step 205, it is determined whether the transmission status is normal or not. If it is normal, the received data is passed to the user program in step 207, and the state returns to the idle state in step 208. Conversely, if the transmission status is not normal, step 209 The number of data transmission attempts is determined in step 205, and the data transmission in step 205 is repeated until a predetermined value is reached, and when the predetermined value is exceeded, transmission error processing is performed in step 210.

Also, FIG. 13 shows the ACK data or N at the transmitting station.
This shows the processing procedure when receiving ACK data. If these data are received in step 111, it is determined in step 112 whether or not ACK data has been received, and the ACK data is
If it is K data, it is returned to the initial idle state in step 113, and if it is not ACK data, the number of data transmission attempts is determined in step 114, and the 11th
The data transmission process shown in the figure is repeated, and when a predetermined value is exceeded, transmission error processing is performed in step 115.

Next, FIG. 14 is a flowchart showing the processing procedure after the diagnostic timer times out. In the figure, if a timeout of the diagnostic timer is detected in step 121, the number of data transmission attempts is determined in step 122, and the data transmission process shown in FIG. If so, transmission error processing is performed in step 123.

(Problems to be Solved by the Invention) FIG. 15 is a diagram showing the relationship between the transmission statuses of the above-mentioned transmitting station. Here, the main CPUBa and the transmission control CP08C are connected by the data/address bus 8b, and the transmission control Assuming that the transmission line 8d is connected to the CPU 8C and a data transmission request is made from a user task, the following four patterns exist as transmission status transitions at the transmitting station.

■8e-P8h-”8f ■ 8e → 81 → 8g *■　8e groan 8h → 8g ■8 e-+8 i-8f however, 8e: Data transmission request route 8f: Normal transmission completion status route 8g = Abnormal transmission completion status route 8h: Normal data transmission route 81: Data abnormality sending/sending failure route.

FIG. 16 is a diagram showing the relationship between the transmission statuses of the above-mentioned receiving stations. When data is received from the transmission line 8d, the following four patterns exist as transmission status transitions at the receiving station.

*■　9a→9C ■9b→9d ■ 9a → 9d ■9b→9C however 9a: Normal data input route 9b=Data abnormal input route 9c: Normal data reception route 9d: Data abnormal reception/no reception route It is.

Here, pattern ■ at the transmitting station is a case where the data transmission is judged to have failed even though the data is being sent normally, and pattern ■ at the receiving station is a case where the data is normally input using the normal data input route. This is the case when the message is received.

Assuming that actual transmission and reception is performed using these patterns ■ and ■, the transmitting station will retransmit the data based on the determination that the transmission has failed, and the receiving station will have already received normal data and finished its processing. However, it also receives normal data and processes it. Furthermore, on the receiving station side, ACK/NACK data is returned to the source station for the first day's data reception, and ACK/NACK data is also returned to the source station for the second day's data reception.

On the other hand, on the transmitting station side, the reception of ACK/NACK data is 1
Even though I expected only one, I ended up receiving this twice, and I was unable to continue the sequence after that.

However, when the transmitting station transmits in pattern 2 and the receiving station receives in pattern 2, there is a problem in that the transmission sequence hangs up after the above process.

This invention was made to solve the above problems, and is a communication protocol redundancy method that can maintain the consistency of the transmission sequence even if an error occurs in the opposite direction to the result of the event. The purpose is to provide

[Structure of the invention]

(Means for Solving the Problems) In the present invention, the transmitting station side adds a data sequence number unique to each station to transmission data as redundancy information in the application layer and transmits the data, and the receiving station side adds the data sequence number to transmission data as redundancy information in the application layer. At the same time, the validity of the transmitted data is determined by checking the series of data sequence numbers, and the transmission data that lacks validity is rejected, and the sending side returns the data sequence number of the response data and the data at the time of transmission. It is characterized in that it is checked whether or not the sequence numbers match, and when the two do not match, the response data is rejected and a response wait state is created.

(Function) In the distributed processing system described as the prior art, as shown in FIG. 2, the transmitting station side generates data 10b...n in which header information is added to each of the data 1, 2...n of the transmission text 10a. IOC is sent sequentially, and the receiving station adds header information each time it receives the data.
It is designed to return CK/NACK data. In contrast, the present invention, as shown in FIG.
, etc.), and the receiving station adds the data sequence number in addition to the header information.
Send back the K data. For such a data format, the transmitting station side and the receiving station side respectively execute data validity judgment and subsequent pig processing using the following logic.

A. When receiving ACK/NACK data from the transmitting station, if the data sequence number added to it matches the data sequence number at the time of data transmission, this ACK/NACK data is
Processing is performed based on the NACK data, and if they do not match, this ACK/NACK data is rejected.

B. When receiving data from the receiving station side is received, if the data sequence NCL added to that data is increased by "1" compared to the previous data, the received data is considered valid and subsequent processing is performed, and the data is If the sequence number is the same as or smaller than the data sequence number added to the previous data, the currently received data is discarded.

This makes it possible to maintain the consistency of the transmission sequence even if an error occurs in the opposite direction to the outcome of the event.

(Embodiment) FIG. 3 is a time chart for explaining the outline of an embodiment of the present invention. When the transmitting station side adds a data sequence mark and sends data, the receiving station side Data rejection process 6 for determining the validity of transmitted data and remedial processing by examining the sequence of
p, and when the transmitting station side receives ACK/NACK data, the data sequence number added to it,
This method differs from the conventional method shown in FIG. 10 in that a process 6q is performed to discard the ACK/NACK data if it does not match the data sequence Nα at the time of data transmission.

FIG. 4 shows the processing at the time of data transmission at the transmitting station. In step 106 before executing data transmission, a data sequence Nα unique to the transmission data at that time is extracted, and in step 107, this data sequence No. is added. There is.

FIG. 5 is a flowchart showing the data reception process in the receiving station including the data rejection process 6p, and in contrast to FIG.
214 processing is added. In this case, step 2
When the data is received in step 211, the data sequence number is determined.

Then, in step 212, it is verified whether this data sequence number matches the previous value, and if it is "+1" with respect to the data sequence number received last time, that is, , if updated, step 2
02 and subsequent processes are executed, but in any other case, the received data is rejected in step 213, and the initial idle state is returned to in step 214.

Figure 6 shows ACK/NA including the above data rejection process 6q.
3 is a flowchart showing processing when receiving CK data.

This process includes steps 116 to 120 added to FIG. 13, which shows the conventional process procedure. Among these, in step 116, the data sequence Nα is extracted, and in step 117, it is checked whether this data sequence Nα matches the data sequence N at the time of transmission, and if it does not match, the received ACK/NACK data is is rejected in step 118, and ACK/
Return to idle state waiting for NACK data. On the other hand, if it is determined in step 117 that the data sequence Nα matches the data sequence No. at the time of transmission, ACK is sent.
At step 120 after determining that data has been received, the data sequence Nα is set to "+1" in preparation for the next data transmission.

If such a processing operation is expressed in correspondence with the processing operation of FIG. 9, it will be as shown in FIG. 7, and the transmission source station 1a
(St) side, AC due to data sequence Nα mismatch
A process 121 for rejecting K/NACK data is added,
On the destination station 1b (Sj) side, processing 12k is added for discarding A CK/N ACK data due to data sequence No. and mismatch.

In this way, even if the transmitting station processes the above pattern (i), i.e., determines that the data transmission has failed even though the data is being sent normally, and retransmits the data, it will not be possible to receive normal data. In the receiving station that is currently receiving the data, the data sequence No. is the same as the one received last time, so it is rejected. Therefore, ACK/NACK
The response is also executed only once, and the hang-up of the transmission sequence based on the combination of pattern (2) and pattern (2), which was a problem in the conventional method, is resolved.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, a data sequence number is added to each data, and the validity of the data is checked by checking the series of data sequence numbers at the transmitting station and the receiving station. Since the data is checked and data is discarded or processed based on the result, the consistency of the transmission sequence can be maintained even if an error occurs in the opposite direction to the result of the event.

[Brief explanation of drawings]

Figure 1 is a data format for explaining the present invention in detail, Figure 2 is a data format for explaining a conventional communication protocol redundancy system, and Figure 3 is for explaining a transmission procedure in an embodiment of the present invention. 4 to 6 are flowcharts showing the processing procedure of the CPU according to the same embodiment, FIG. 7 is an explanatory diagram for explaining the transmission procedure of the same embodiment, and FIG. 8 is a conventional Fig. 9 is an explanatory diagram for explaining the transmission procedure of the system, and Fig. 10 is an explanatory diagram for explaining the transmission procedure of the system. FIGS. 11 to 14 are flowcharts showing the processing procedure of the CPU constituting the system, and FIGS. 15 and 16 are transition diagrams for explaining the occurrence of a transmission abnormality. 7a-1 to 7a-n...user task, 7b...transmission task, 7c...reception task, 7d...transmission initialization task, 7e...transmission driver, 7f...reception driver, 7g...Transmission control unit, 7h...Transmission buffer, 71...Reception buffer, 7j...Local buffer, 7k...Status information, 71, 8d...
・Transmission line, 7m18c...Transmission control CPU, 7n
, 8a...Main CPU0 Applicant's agent Mr. Sato Figure 1 Figure 2 Figure 6P Figure 3 Figure 4 Figure 6 Figure 12

Claims (1)

[Scope of Claims] A distributed processing system in which a plurality of CPUs exchange information with each other through a transmission path, and initializes a transmission sequence within an interface device in response to an abnormality in the transmission sequence due to noise intrusion, etc. The transmitting station side adds a data sequence number unique to each station to transmission data as redundancy information in the application layer and transmits the data, and the receiving station side returns response data with the data sequence number added, and also sends the data sequence number. determine the validity of the transmitted data by examining the sequence of
rejecting the transmission data that lacks validity; the transmission side checks whether the data sequence number of the response data matches the data sequence number at the time of transmission;
A communication protocol redundancy method characterized in that when the two do not match, the response data is discarded and a response wait state is created.