KR930007239B1 - Dual trunk control method - Google Patents

Abstract

The method is for smoothly performing the initialization of processor and improving the facility of achierement, the accuracy of control, efficiency of performance and the reliability of hardware unit operation by checking the status of state flags setin several steps needed to activate a relay line control processor. It is composed of the stages of: re-starting self processor until receiving all data correctly; operating as exclusive responsibility mode of load; and operating as shared responsibility mode of load when self-processor is in response stand-by mode.

Description

How to check status during start-up for redundant redundant control

1 is a hardware configuration to which the present invention is applied.

2 is a flowchart illustrating a state transition in a process of transmitting and receiving a signal between upper processors according to the present invention.

3 is a flowchart illustrating a state transition in a process of transmitting and receiving signals between counterpart processors of the present invention.

* Explanation of symbols for main parts of the drawings

100: subscriber switching processor (ASP) 101, 102: relay line control processor

103 to 106: trunk line hardware unit

The present invention relates to a method for checking processor status during start-up for relay line control redundancy of an electronic switchboard.

In large-capacity electronic switchboards, real-time processing and minimal call processing loss and quick reconfiguration are required in case of failure. To this end, the trunk line control processor, which handles relay call processing in an electronic switch, adopts a distributed structure redundancy method in which two processors control four hardware units separated from each other.

Accordingly, an object of the present invention is to initialize the operation of the processor smoothly by checking the state by placing a status flag in each step until the relay line control processor is activated, as well as the ease of implementation, accuracy of control, and performance in the relay line control redundancy. It is possible to realize the redundancy of the load distribution structure of the trunk line control processor more effectively by using the state check method during the start-up process for the trunk line control redundancy, which improves the efficiency and reliability of hardware unit operation.

In order to achieve the above object, the present invention provides a trunk line control processor having a load-balancing redundancy structure, which communicates with an upper processor ASP and a relative redundancy control processor through IPC, and controls the relay line uni- plex of the relay switch. A method for checking a state of a computer, the method comprising: checking whether a state of a processor is waiting to receive data from a higher level so that the processor can transition to a next state only when data necessary for operating a relay line hardware unit is correctly received from a higher level processor. In this case, the first step is to restart the own processor until all data is received correctly. Step 2, when receiving a response to a request for a partner processor state, checks whether the processor is in a wait state for the other processor state and if true, loads the processor in dedicated load mode if the state of the other processor is less than the state of the processor. In the third step of operating the processor, if the state of the counterpart processor is the same as the state of the own processor, the counterpart reports the completion of the relay line hardware unit initialization of the share, makes the state of the own processor in the load sharing check state, and then distributes it from the counterpart processor. The fourth step of receiving a report of the completion of the hardware unit initialization completion and operating in the load sharing mode. If the counterpart processor is operating in the load-sharing mode, the request is made to the counterpart processor for load sharing, and then the support processor status is waited for the load sharing response. In a state And a fifth step of restarting the processor when the response wait time for receiving the load-sharing request is timed out and checking whether the processor is in the load-sharing response waiting state when the response is received. It is done.

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

In the hardware configuration to which the present invention is applied, as shown in FIG. 1, a relay control processor has a load-balancing redundancy structure, and an AP (Access Switching Processor) 100 and a relative redundancy control, which are higher processors, through IPC (Inter Processor Communication). It communicates with the processor 102 and controls the corresponding trunk line hardware units 103, 104, 105, 106. If both trunk line control processors 101 and 102 operate in load sharing mode and one of the trunk line control processors fails and cannot operate, the other trunk line control processor controls the hardware unit that was in charge of the failed trunk line control processor. It takes over and operates in dedicated load mode.

In the present invention, a description will be given of the state check method based on the A side trunk line control processor based on FIGS. 2 and 3.

First, in FIG. 2, when the data necessary for operating the relay line hardware unit (the hardware unit mounting information to be controlled, the time data necessary for the various signals, and the information on each type of the relay line) are correctly received from the upper processor, the state transitions to the next state. If not, restart the magnetic processor until all data is received correctly.

Referring to FIG. 2 step by step, the first processor requests the data required by the upper processor (201), and then sets the self processor state to the appropriate data wait state from the different processor (202). The processor is restarted (204), and when data is received (203), it is checked whether its processor is in a proper state of waiting for data reception (205), and if it is true, it is checked whether it has received all necessary data from the higher processor (206). . If the result of the data reception check step 206 is true, the partner processor requests the partner processor state (207), and then places the own processor state in the wait state for the partner processor status response (208).

In FIG. 3, the state of the counterpart processor is checked to operate in a load-only mode when the counterpart is in an idle state and less than the state of the own processor. In the case of operating in the load sharing mode and the opponent is operating in the dedicated load mode, the load sharing mode is requested by the counterpart and the load sharing mode is operated when the response is received.

Referring to FIG. 3 step by step, first, when the response reception wait timeout for the relative processor state request requested in the relative processor state request step 207 of FIG. 2 is reached (301), the opponent is determined to be in an operation stop state. It operates in the dedicated load mode (302) and, if a response is received (303), checks whether the own processor is in a wait state for the relative processor state response (304) and if true, checks the relative processor state (305).

If the state of the counterpart processor is less than the counterpart of the own processor in the step of checking the counterpart processor state 305, the selfprocessor is operated in a dedicated load mode (306), and the state of the counterpart processor is the same as that of the own processor. In case the other party reports the completion of the shared trunk line hardware unit initialization (307), puts the state of its own processor in the load sharing check state (308) and receives the report of the completion of the shared hardware unit initialization from the other processor. In operation 309, the load sharing mode is operated. Here, the reason for setting the load sharing check state is to avoid duplication of hardware unit initialization in case the counterpart and the own processor start at the same time to initialize the shared hardware unit and one side immediately stops working due to a failure. will be.

When the counterpart processor is operating in the dedicated load mode in the counterpart processor state checking step 305, a request is made for the load sharing request to the counterpart processor (311), and then the self processor state is placed in a load sharing response wait state (312). When the response wait timeout for a sharing request is timed out, the self processor is restarted (313), and when a response is received (314), it is checked whether the self processor is in a load sharing response waiting state (315) and if it is true, it operates in the load sharing mode. 316. A status plug is provided at each stage until the relay line control processor is activated to wait for data reception from an upper flag, the processor status response wait state, load sharing increment state, load sharing response wait state, operation state in the load sharing mode, And check the dedicated load mode operation state to perform effective load balancing redundancy control.

As described above, the present invention firstly accurately receives data required for operation of the relay line hardware unit from one block of the upper processor ASP, and then performs a load or load-only operation state according to the state of the counterpart processor to duplicate the load balancing structure. To realize.

The effect by the present invention is as follows.

By using the state check method in the start-up process for redundancy control, the redundancy of the load distribution structure of the repeater control processor can be more effectively initialized, and the ease of implementation and control in relay control redundancy is achieved. It improves the accuracy, performance efficiency, and reliability of the operation of the hardware unit to ensure the reliability of service with minimal call processing loss and quick reconfiguration in case of failure.

Claims (5)

The trunk line control processor (101, 102) has a load balancing redundancy structure, and communicates with the upper processor ASP (100) and the partner redundancy control processor (101 or 102) through the IPC, and the corresponding trunk line unit (103, 104, 105, 106). A state checking method for relay line control redundancy of an all-electronic exchanger configured to control; It checks whether the state of its own processor is waiting to receive data from the upper level so that it can transition to the next state only if it receives the data necessary for operating the trunk line hardware unit from the upper level processor correctly. The first step of restarting the own processor until the wait timeout for response to the request of the counterpart processor state is determined, the second step of determining that the counterpart is in the stopped state and operating in a load-only mode; In response to receiving the response, the processor checks whether the own processor is in a waiting state of the relative processor state. If true, if the state of the other processor is less than the state of the seat processor, the third step of operating the own processor in the dedicated load mode is performed. three If is equal to the state of his processor, report to the other party the completion of the shared-line hardware unit initialization, bring the state of its processor into the load-sharing check state, and then report the completion of the shared hardware unit initialization from the other processor. Step 4 in which the counterpart processor is operating in the load-sharing mode, if the counterpart processor is operating in the load-sharing mode, the request is made to the counterpart processor, and then the state of the processor is placed in the load-sharing response standby state and the response to the load sharing request is received. When the wait time is received, the self processor is restarted, and when a response is received, the self processor checks whether the self processor is in a load sharing response wait state, and if it is true, the fifth step of operating in the load sharing mode is performed.How to check the status of the same process. The start-up of the relay line control redundancy according to claim 1, wherein the received data of the first step includes hardware mounting information to be controlled, time data for various signals, and information on each type of relay line. How to check the status in the process. The method of claim 1, wherein an effective state of load balancing redundancy control of the relay line hardware unit is realized by placing a state flag in each step. 4. The status flag according to claim 3, wherein the status flag is a data reception wait status flag from a higher processor, a counterpart processor status response wait status flag, a load sharing check status flag, a load sharing response wait status flag, a load sharing mode operating status flag, and a load. A state checking method during start-up for relay line control redundancy, comprising a dedicated mode operating state flag. The hardware unit initialization of claim 1, wherein the load sharing check state of step 4 is set so that the counterpart and the self-processor start at the same time to initialize the shared hardware unit, and one side immediately stops operating due to a failure. Method of checking the state during start-up for relay control redundancy, characterized by avoiding the duplication of.