KR910003941B1 - Calling and call-restoring method by use of communications between processors - Google Patents

Abstract

The stand by processor receives the data base of the current acting microprocessor to maintaining the loop when a fatal abnormality occurs. The method includes steps: (A) initializing the system by the stand-by microprocessor; (B) maintaining the data base of special service function and the current control loop and initializing other data; (C) reconstructing the loop according to the special service function and loop data stored in a memory; (D) checking another call; and (E) processing the call, storing the call data on the memory and transmitting the data to the stand- by microprocessor.

Description

Call recovery method of private exchange system

1 is a conventional system flow diagram.

2 is a system block diagram of a private exchange for carrying out the invention.

3 is a flowchart of a system operation in accordance with the present invention.

4 is a call recovery flow diagram in accordance with the present invention.

5 is a flowchart illustrating another example of a wake-up function registration according to the present invention.

* Explanation of symbols for main parts of the drawings

10: first control unit 20: second control unit

30: bus monitoring and selection circuit 40: trunk line subscriber circuit

50: currency exchange circuit 60: extension subscriber circuit

The present invention relates to a method for recovering a call of a private exchange, and more particularly, to a method of maintaining a call state and a special function even when a system error occurs during operation by a redundant processor of a private exchange.

1 is a flowchart of a call and function service in a conventional private exchange, which proceeds as follows.

In the step (A1), the system initializes all the data in the hardware and memory of the system at the moment of starting. When a call is generated, a call path is formed in the system in step (A2) to perform a call service of the corresponding subscriber. When a special function is generated, each function is serviced in step (A3).

At this time, if an unexpected serious error occurs while the system is running, the system re-performs the system initialization process again.In this case, all the lines in the service state are disconnected by initializing all databases that were in hardware and service. All special functions registered by users are also deleted. In addition, when a fatal abnormality occurred in the processor, there was a problem that the whole system is stopped.

Accordingly, an object of the present invention is to provide a message to the standby processor, the processor operating in the private exchange of the redundant processor, the call recovery can maintain the call state by taking over control of the waiting processor even if an error occurs in the running processor In providing a method.

Another object of the present invention is to provide a method for recovering and servicing a special function by operating a standby processor when an abnormal phenomenon occurs in a running processor.

Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 2 is a system block diagram of a private exchange. The first and second controllers 10 and 20 control the system by operating as an active processor and a standby processor, and the first and second controllers ( A bus monitoring and selection circuit 30 for monitoring the buses of 10 and 20, switching control rights when an error occurs, and displaying an error state, and a trunk line subscription circuit for interfacing the trunk line with the system under the control of the controllers 10 and 20. 40, a call exchange circuit 50 for switching a communication path under the control of the controllers 10 and 20, and an extension subscriber circuit 60 for interfacing the subscriber and the system under the control of the controllers 10 and 20. It consists of.

3 is a flowchart illustrating an operation of a system according to the present invention, the process of initializing the hardware of the system at system startup, saving the call and function related data in the storage device after initializing the hardware, and initializing the remaining data; After performing data initialization, a third process of transmitting a call and a function message generated in a normal state to a waiting processor is performed.

4 is a flowchart of a call recovery method according to the present invention, and FIG. 4 (a) is a call flow diagram of an operating processor. 4b is a call flow diagram of a standby processor.

Here, the operation processor initializes the hardware configuration of the system at system startup. After that, the data stored in the storage device is initialized except for call and function related data, and the call state is restored by the call and function related data. If a call state occurs after the initialization process, the service related to the call is performed, and the service result is transmitted to the standby processor. At this time, if a fatal error occurs during the call service process, the control transfers data to the standby processor and the operating processor is switched to the standby state.

In addition, the standby processor initializes the hardware configuration of the system at system startup, and initializes the remaining data except for call and function related data from data also stored in the storage device. Thereafter, when the service processing result message is received from the operating processor, the received message is stored in the storage device. If the control switching data is received from the operating processor while the above process is repeated, an acknowledgment is sent and the control is switched to the operating processor. Argument, and then initializes the operating processor.

5 is a flowchart of a wake up function according to an embodiment of the present invention. FIG. 5A is a flowchart of registering a wake up function of a running processor, and FIG. 5B is a flowchart of a waiting processor.

Based on the above-described configuration, the present invention will be described in detail with reference to FIGS. 2, 3, 4, and 5.

2 is a block diagram of a private exchange with a redundant processor.

In the above configuration, the first and second controllers 10 and 20 are redundant processor units for controlling the private exchanger. Here, the first controller 50 assumes the operation controller and the second controller 20 as the standby controller. Accordingly, the first controller 10, which controls the overall operation of the system by incorporating the first processor 11, controls the trunk line subscriber circuit 40, the call exchange circuit 50, and the internal subscriber circuit 6. When a call occurs, a call path is formed and the call and function data related thereto are stored in the RAM 13 as a storage device. In addition, the bus monitoring and selection circuit 30 selects the first processor 11 operating among the two processors 11 and 21 to control the switching unit, the service unit, the termination unit, and the like. 11) In case of a failure, a relay switching circuit, an auto / manual switching circuit, a CPU switching circuit, an error display circuit, etc., which automatically switch to the waiting second processor 21 are built in.

According to the present invention, when an error occurs in the operating first processor 11 by performing interprocessor communication (IPC) between the operation and the standby processors, the waiting second processor 21 may perform a subsequent job. It takes over and implements the functions of the system without interrupting service. Therefore, even in the event of a fault in the system, a system that provides a service without interruption to a user may be implemented. The operation proceeds as follows.

First, a process of preserving data required for system initialization and transmitting and receiving a message by the redundant processor will be described with reference to FIG.

First, in the step (B1), the FDD (FLOPPY DISK DRIVER) 17, the DMA circuit 15, the trunk line subscriber circuit 40, and the call exchange circuit 50 in the first control unit 10 at the system start. The first processor 11 initializes relevant hardware parts, such as the extension subscriber circuit 60. In addition, the second processor 21 in the standby state initializes the hardware components in the second control unit 20. After that, the processor 11 or 21 proceeds to step B2, and among the data stored in the RAMs 13 and 23, which are memory devices, data related to a call and data related to a function are considered in consideration of self restart. The current data value is kept as it is without initialization at startup. Each of the first processor 11 and the second processor 21 initializes all data except for data related to a call and data related to a function among data in the RAM 13 and 23 in step (B3). After the normal state (B4) through the operation state of the first processor 11, the call state message, the function state message, the function state message, etc. generated by the information exchange bus 18 21, and if necessary, the second processor 21 in the standby state also transmits a message to the first processor 11 in the operating state through the information exchange bus 18. In addition (B5) provides a call service through the call exchange circuit 50, the bus drive circuit 52 and the extension subscriber circuit 60, the trunk line subscriber circuit 40 and the like.

A call recovery process by the operation processor will be described with reference to FIG. 4A.

First, in step C1, the first processor 11 initializes related hardware such as the call exchange circuit 50, the FDD 17, the DMA circuit 15, and the like. In step (C2), all data except for data related to a call and a function are initialized from the data stored in the RAM 13. Thereafter, in step (C3), the first processor 11 recovers the call state by referring to the data stored in the RAM 13. That is, when the system self-restarts or the waiting processor takes over and starts the operating processor, the operating processor refers to the data stored in the RAM and makes a call to the subscriber in the active state. The exchange circuit 50 is controlled to form a communication path for the CO line subscriber circuit 40 and the internal subscriber circuit 60, thereby saving the call state.

Then, when a call state occurs, the first processor 11 recognizes this in step (C4), provides a call service function for the call in step (C5), and stores information related to the call state in the RAM 13. Store in In step (C6), the result of the call service is transmitted to the second processor 21 in the standby state through the information exchange bus 18. This is for the first control unit 20 to take over the control right and to control the system in its original state.

However, if an unexpected error condition occurs during the call service, the first processor 11 switches to the standby processor.

Secondly, the operation of the second processor 21 on the standby side will be described with reference to FIG. 4B.

First, in step D1, the second processor 21 initializes the hardware configurations related to the FDD 27 and the DMA circuit 25. In step (D2), all of the RAM 23 and the data currency and functions being stored are initialized except for the related data.

Thereafter, the second processor 21 checks whether a message is received through the information exchange bus 18. If a call state message generated from the first processor 11 is received through the information exchange bus 18, (D3). In step) it recognizes and receives the message received in step (D4) and stores in the RAM (23). Since the message processed by the operating processor through the above process is stored in the same form in the standby processor, even after the control right is switched, the standby processor is switched to the operating processor, and by the data stored in the RAM 123 Through this, the state can be plural.

Here, the information exchange bus 18 is connected to the fatigue (FIFO) so that the first and second control units (10, 20) can communicate with each other.

If the control switching data is received through the bus monitoring and selection circuit 30 while receiving the home message and storing the message in the RAM 23 as described above, the second processor 21 recognizes this in step D5 and recognizes it (D6). In step), the verification data is sent to the first processor 11 and the process proceeds to the operation processor. Then, the second processor 21 switched to the operation processor performs the flow as shown in FIG. 4A to restore the call state by the call data stored in the RAM 23.

A process of transitioning the processing of the operating processor and the standby processor from the operating state to the standby state (standby state to the operating state), and a series of operations performed by the operating processor and the standby processor, respectively, between call generation and termination. In this section, we will look at the sequence of processes that are performed in the operation processor and the standby processor at the end of the call.

First, the operation processor transfers control to the standby processor. If an unexpected error occurs while the first processor 11 in the operating state is in service, the first processor 11 recognizes this at step (C7), and at step (C8) the bus monitoring and selection circuit ( The control switching data is output to the second processor 21 through 30). In this case, first, the first processor 11 writes a specific data (1100 0000: C0 _H ) to a specific address (0000 _H ) to enable a switchover state, and then again to the specific address (0030 _H ). 0000 0000: 00 _H ), and the first processor 11, which was an operating processor, transitions to the standby processor, and then transfers the control right to the second processor 21 in the standby state.

In addition, the second processor 21 in the standby state recognizes the data transmitted from the first processor 11 in step (D5) in step (C8), and specifies the specific address (0000 _H ) in step (D6). By writing the data (0000 0100: 05 _H ), the second processor 21 can be restarted by software, and by writing the specific data (0000 0001: 01 _H ) at the specific address (0020 _H ) again, the second processor Let 21 restart as the operating processor.

Therefore, the first processor 11, which is an operating processor, outputs data for switching control to the second processor 21, which is a standby processor, and transitions itself to a standby state, and the second processor 12, which is a standby processor, writes to the data. Is shifted to the operating state to restore the call state of the system.

Thereafter, the second processor 21 recovers the call state by performing the same process as in FIG. That is, the second processor 21 initializes the trunk line subscriber circuit 40, the call exchange circuit 50, the internal subscriber circuit 60, the second control unit 20, and the like in step (C1), and the step (C2). In the RAM 23, since call-related data are stored, the call state is restored with reference to the call-related data in step (C3). In addition, the first processor 11 switched to the standby state performs an initialization process through steps (D1) and (D2).

Second, we look at a series of processes in which a call is generated and then processed by the operating processor and the standby processor.

First, when a call is generated, the second processor 21 in the operating state recognizes it in step (C4), performs a process for the call in step (C5), and generates a message as follows in step (C6). The information exchange bus 18 transmits the data to the first processor 11.

MSG ID: Message Identification Number

MSG-counter: Total number of blocks transmitted

Serial no: Current block serial number

data 1

data 2

data 3 data

data 4

data 6

The message counter (MSG counter) of the data type refers to the total number of blocks transmitted in association with the call, and the serial number (serial NO) refers to the serial number of the currently transmitted block among the total number of blocks.

Then, the first processor 11 in the standby state recognizes the above message in step (D3), and stores the message received in step (D4) in the call-related buffer of the RAM 13. Therefore, the second processor 21 and the first processor 11 to maintain a database related to the call to each other.

Thirdly, we look at the sequence of processing to the working processor and the standby processor at the end of the call.

When the call ends first, the second processor 21 in the operating state recognizes this at step (C4), performs a series of services related to call termination at step (C5), and at step (C6), a message type as follows. Is transmitted to the first processor 11 in the standby state through the information exchange bus (18).

MSG-ID

Call-ID 8byte

dumny

data

.

.

.

Then, the first processor 11 in the standby state recognizes this in step (D3), and converts the call identification number (call-ID) in step (D4) to find the call-related buffer position of the RAM 13 to determine the corresponding buffer. Clear the contents. Therefore, it can be seen that the operating processor and the standby processor maintain the same database even for a closed call.

Private exchanges can also perform special functions. The special functions include a wake-up service function, call forward registration, call rejection registration, and message registration. 5 illustrates a registration process of a wake-up service function among the special functions.

First, FIG. 5A is a flowchart of an operation processor, in which step (E1) initializes related hardware, and in step (E2), a value in a database RAM related to a function is not initialized and remains as it is. After that, it checks whether a wake up service is registered in step (E3), and when a wake up service is registered in step (E3), time, minute, etc. for the service in step (E4). Data is stored in RAM. In addition, in order to maintain the already registered wake-up service even if the system is down, the time and minute required to perform the wake-up service to the waiting processor through the information exchange bus 18 in step (E5). data such as number) is sent to the standby processor.

5A is a flowchart of a waiting processor, wherein steps (F1) and (F2) are the same as steps (E1) and (E2). After performing the steps (F1) and (F2), it is checked in step (F3) whether a message related to wake-up registration is input through the information exchange bus 18. When the message related to the wake-up registration is transmitted in step (F3), it is received through the information exchange bus 18 and stored in the RAM. Through this process, both processors can maintain the same database of wake-up services, and based on this, the processors can be switched to maintain the service even when the system is down.

As described above, a fatal abnormality occurs in the system during operation, and even when the system is down, the waiting processor can maintain the same database and take control of the call without interruption, and wake-up registered by the user. Various special function services can also be performed continuously, which has the advantage of improving the reliability of the system.

Claims (2)

A method for recovering a call in a private switching system having a redundant processor of an operating processor and a standby processor, the method comprising: performing a first process (C1) of initializing hardware by an operating processor switched at system initialization and after performing the first process (C1) A second process (C2) for preserving a database of calls and special functions in the storage device and initializing other data; and data related to calls and special functions being stored in the storage device after performing the second process (C2). In the third process C3 of restoring the call state, the fourth process C4 of checking whether another call state occurs after the third process C3, and the fourth process C4 When the status occurs, the call-related data is stored in the storage device after service of the corresponding call, and the service result message is transferred to the fifth process (C5, C6) of transmitting and returning the same counter call-related data to the standby processor. A method of recovering a call using a private exchange system, characterized in that it is carried out. A method for recovering a call in a private switching system having a control unit of a redundant processor comprising an operating processor and a standby processor, wherein the operating processor generates control conversion data when an error occurs during system control, transmits to the standby processor, and switches to a standby state. After the first process (C7, C8, D5, D6) and the first process (C7, C8, D5, D6) that the standby processor receives the control switch data and transitions to the operating state, the switch is performed. The operating and standby processors, respectively, retained the call and function related data stored in the memory and initialized the rest of the data, so that the divert operation processor could resume the service prescribed by the call and function related data. After performing the second process (C1, C2, C3, D1, D2) to restore the state to the original state and the second process (C1, C2, C3, D1, D2), the operation The processor controls the operation of the system, and when the call and function data is generated, the processor processes the data and stores the data in a storage device and transmits the data to a standby processor. The standby processor receives the data and stores the data in a storage device. A method for recovering a currency of a private exchange system, comprising a third process (C4, C5, C6, D3, D4) for maintaining the same data.

Images