JP2000298599A - Automatic testing system for exchange software - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To realize an automatic test system of exchange software capable of efficiently and surely detecting a failure of an exchange application program and performing an automatic test even when a failure occurs. SOLUTION: Since a command of a debugger is periodically executed by a periodic command execution means 108, and a failure detection means 110 detects occurrence of a failure by a state transition of a real-time OS received from a debug monitor 124. The occurrence of a failure in the exchange application program 121 can be detected efficiently and reliably. In addition, when the failure detection unit 110 detects the occurrence of a failure, the command acquisition unit 101 reads the restart procedure file and performs the restart processing, so that the automatic test can be continuously performed even when the failure occurs.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic switching software test system for automatically testing switching software used in an electronic exchange.

[0002]

2. Description of the Related Art As shown in FIG. 6, a conventional automatic test system comprises a workstation 300 and an exchange 320, which are connected by a communication line 330. The workstation 300 includes a command acquisition unit 301, a command inspection unit 302, a command determination unit 303, a debugger command execution unit 304, and a communication path system simulation unit 30.
5. An online command execution unit 306 and an execution result storage unit 307 are provided. The exchange 320 includes exchange software 321, a bootstrap program 322, a real-time OS 323, and a debug monitor 324.

In the automatic test system having such a configuration, first, in the workstation 300, the command acquisition unit 301 acquires a command from a test procedure file in which a command for performing a test is described, and then the command inspection unit 302 acquires the command. Check the command you entered for errors. Then, the command discriminating means 303 discriminates the type of the command by an identifier such as "!" Or "<" added to the head of the command, and outputs the debugger command executing means 304, the communication path system pseudo means 305, The command is passed to the command execution means 306 and executed. The execution result storage unit 307 stores the execution results received from the debug monitor 324 and the exchange application program 321. The above-described processing is performed until a command cannot be obtained from the test procedure file.

[0004]

The above-described conventional automatic test system has the following problems (a) and (b). (A) The debug monitor 32 within a certain period of time after executing a command by the debugger command execution means 304, the communication path system simulation means 305, and the online command execution means 306.
4, and when the execution result cannot be obtained from the exchange application program 321, the failure is uniquely regarded as a failure. Therefore, if the load on the exchange 320 is high and it takes time to execute the command, the exchange application program 321 There is a problem that a fault is detected as having occurred even though it is operating normally.

(B) Online command execution means 306
Each time an online message is received from the exchange application program 321, it is determined whether or not the message is a failure message. If the message is a failure message, the failure is determined to be a failure. There is also a problem that the cost is low.

Accordingly, the present invention has been made in view of such circumstances, and it is possible to efficiently and surely detect the occurrence of a failure in an exchange application program, and to perform an automatic test continuously even when a failure occurs. It aims to provide an automatic test system for software.

[0007]

According to the first aspect of the present invention, there is provided an automatic test system for automatically testing exchange software executed on a real-time OS and embodying processing operations of an exchange. A failure detecting means for periodically executing a debugger command and detecting the presence or absence of a failure from the state of the real-time OS corresponding to the execution of the command; and a predetermined restart when the failure detection means detects the failure. Restart means for executing a restart command for restarting the exchange in a procedure.

According to the second aspect of the present invention, there is provided an automatic test system for automatically testing exchange software which is executed on a real-time OS and embodying a processing operation of an exchange. Means, state storage means for determining and storing the state of the real-time OS from the execution result of the periodic command execution means, and detecting presence / absence of a failure according to the state of the real-time OS stored in the state storage means And a restart procedure obtaining means for reading a command for restarting the exchange from the restart procedure file and executing the command when the fault detection means detects the occurrence of the fault.

According to a third aspect of the present invention, the debugger command outputs information indicating the cause of the inability of the real-time OS to operate.

According to the present invention, a debugger command is executed periodically, and a real-time debugger command corresponding to the command execution is executed.
The presence or absence of a failure is detected from the state of S, and when the failure is detected, the exchange is restarted by a predetermined restart procedure.
The occurrence of a failure in the exchange application program can be detected efficiently and reliably, and the automatic test can be continuously performed even when the failure occurs.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. (1) Summary of the Invention In the present invention, an automatic test system for exchange software detects a failure of an exchange application program by periodically executing a debugger command, and restarts the exchange when a failure is detected. It is intended to provide a configuration for performing an automatic test continuously. Specifically, the periodic command execution means 108 shown in FIG.
The state storage means 109 reads the real-time OS1 from the execution result.
The state of 23 (activated or not activated) is determined and stored. Then, the failure detection means 110 is used to
The occurrence of a failure is detected by determining the state transition at 23.

When the occurrence of a failure is detected, the command acquisition means 101 reads the command from the restart procedure file in which commands for restarting the exchange 120 are sequentially described and executes the command, thereby restarting the exchange 120 and continuing. Perform an automatic test. This makes it possible to detect the occurrence of a failure in the exchange application program and restart the exchange when the failure is detected, thereby continuously performing the automatic test.

(2) Configuration Next, the configuration of an automatic test system according to an embodiment will be described with reference to FIG. The automatic test system according to the present invention operates under program control,
It comprises a workstation 100, an exchange 120, and a communication line 130.

Configuration of the Workstation 100 The workstation 100 includes a command acquisition unit 10
1, command checking means 102, command discriminating means 10
3. Debugger command execution means 104, communication path system simulation means 105, online command execution means 106, execution result storage means 107, periodic command execution means 108, real-time OS state storage means 109, and failure detection means 1.
It consists of ten.

The command obtaining means 101 obtains commands from a test procedure file in which commands for executing a test are sequentially described and a restart procedure file in which commands for restarting when a failure has been detected are sequentially described.
The command checking unit 102 checks the acquired command for errors. The command determining unit 103 determines the type of the acquired command, and transfers the command to the debugger command executing unit 104, the communication channel system simulating unit 105, and the online command executing unit 106. The debugger command execution means 104 communicates with the debug monitor 124 via the communication line 130 to execute a debugger command.

The communication path system simulating means 105 is connected to the communication line 13.
Signals are transmitted to and received from the exchange application program 121 via the debug monitor 124 via the debug monitor 124 to simulate a communication path system device. Online command execution means 10
6 communicates with the exchange application program 121 via the communication line 130 and executes an online command. The execution result storage unit 107 stores the execution results of the commands executed by the debugger command execution unit 104, the communication channel system simulation unit 105, and the online command execution unit 106.

The periodic command execution means 108 periodically outputs a debugger command to the debugger command execution means 104.
Hand over to The real-time OS state storage means 109
The state of the real-time OS 123 is determined from the execution result of the command of the debugger executed periodically and stored. The fault detecting means 110 is provided in the real-time OS state storing means 10.
9, the transition between the state of the real-time OS 123 stored last time and the state of the real-time OS 123 stored this time
It is determined whether a failure has occurred in the exchange application program 121.

Configuration of Exchange 120 The exchange 120 is provided with the exchange application program 12
1, a bootstrap program 122, a real-time OS 123, and a debug monitor 124. The bootstrap program 122 calls an initialization process of the debug monitor 124 when the bootstrap program 122 itself is started. When the real-time OS 123 starts itself,
The initialization process of the debug monitor 124 is called. When a failure occurs in the exchange application program 121, an exception interrupt occurs in the real-time OS 123. The real-time OS 123 activates the bootstrap program 122 when a failure occurs in the exchange application program 121 and an exception interrupt occurs.

(3) Operation Next, the operation of the automatic test system having the above configuration will be described with reference to FIG. Overall Operation First, the command acquisition unit 101 acquires one command from a test procedure file in which commands for executing a test are sequentially described (step A1). Subsequent step A2
Then, the command acquisition procedure 101 determines whether or not the command has been acquired. If the command cannot be acquired, the automatic test ends. On the other hand, if the command can be obtained, the process proceeds to step A3,
02 checks whether the acquired command is correct.

The command discriminating means 103 discriminates the type of the acquired command based on an identifier such as "!" Or "<" added to the beginning of the command, and outputs the command to the debugger command executing means 104, The means 105 delivers the command to the online command execution means 106, executes the command, and acquires an execution result (step A4). Next, in step A5, the execution result storage unit 107 stores the execution result of the command.

Thereafter, the process proceeds to step A6, and the periodic command execution unit 108 requests the debugger command execution unit 104 to execute a panic command for referring to the setting information of the panic trap. The panic command is used to set a panic trap for outputting a message indicating the cause of the inoperability to the standard output when the real-time OS 123 becomes inoperable, cancel the panic trap, and refer to the panic trap setting information. This is the debugger command to be executed.

For reference of panic trap setting information, pa
When the nic command is executed, as the execution result of the panic command, if the real-time OS 123 is running, the panic trap setting information (panic trap is set or panic trap is not set) is transmitted to the real-time OS 123. Is not activated, the non-activated state can be received from the debug monitor 123 as the state of the real-time OS.

The debugger command execution means 104
Execute the ic command and display the execution result in the debug monitor 1
23 (step A6), the real-time O
The S state storage unit 109 stores the non-startup as the state of the real-time OS 123 when the execution result of the received panic command indicates that the real-time OS has not been started, and stores the real-time OS when the execution result is other than the real-time OS not started.
The activation is stored as the state of S123 (step A
7).

When the panic command is executed, the program of the running exchange 120 is stopped. Therefore, the periodic command execution means 108 executes the g command for running the exchange program by the debugger command execution means 1.
Request to 04. Debugger command execution means 105
The command is executed to run the program of the exchange 120 (step A8). Then, the fault detecting means 110
Determines whether a failure has occurred based on the transition between the state of the real-time OS 123 previously stored in the real-time OS state storage means 109 and the state of the real-time OS 123 stored this time (step A9).

Here, it is determined that a failure has occurred if the state transition is from activation to non-activation, and that no other failure has occurred in the other state transitions. Of test execution. If a failure has occurred, the command acquisition unit 101 proceeds with the process to step A10, acquires one command from the restart procedure file in which commands for resuming are sequentially described, and determines in step A11 whether a command has been acquired. Is determined. If the command cannot be obtained, the restart processing has been completed, and the test execution processing from step A1 is continuously performed. If the command can be obtained, the restart processing is continuously performed.

Next, the processing in steps A12 to A17 is as follows:
This is the same process as steps A3 to A8 described above. The failure detection means 110 determines whether a failure has occurred, as in step A9. If a failure has occurred during the restart processing, the automatic test ends. If no failures have occurred,
Continue the processing from step A10 (step A
18). After the restart, the test execution processing from step A1 is continuously performed. Steps A1 to A9 or A18 are repeated while a command can be obtained from the test procedure file.

Next, in accordance with the failure detection processing sequence shown in FIG. 3, steps A6 to A9 and steps A15 to A15 in FIG.
For the failure detection processing corresponding to A18, the exchange 120
This will be described together with the operation of the program. Exchange application program 121, bootstrap program 122, real-time OS 123, debug monitor 12
4 operates similarly to a conventional automatic test system.

First, when the exchange 120 is started up, the bootstrap program 122 assigned to the fixed address of the exchange 120 is automatically started, and the bootstrap program 122 calls the initialization processing of the debug monitor 124 and Reference numeral 124 stores the state of the real-time OS 123 as not-started (sequence B1 in FIG. 3). The periodic command execution means 108 internally activated by the automatic test system at the start of the automatic test requests the debugger command execution means 104 to execute the panic command, and the debugger command execution means 104 executes the pani command.
Execute the c command (sequence B2).

The debug monitor 124 stores the stored real-time OS as the execution result of the panic command.
The inactive state 123 is transmitted to the debugger command execution unit 104. Real-time OS storage means 109
Is the real-time O received from the debug monitor 124.
The non-startup state of S123 is stored (sequence B
3). When the panic command is executed, the program of the exchange 120 being executed is stopped.
Command execution request, debugger command execution means 1
04 executes the g command to run the exchange program. The failure detection means 110 is a real-time OS 1
Since the state of 23 has not transitioned from active to inactive, it is determined that no failure has occurred (sequence B4).

Until the real-time OS 123 is started, the sequences B2 to B4 are repeatedly performed. The bootstrap program 122 is installed in the exchange 12
Real-time OS1 stored at a fixed address of 0
Read the execution start address of the real-time OS
When the 123 is started, the real-time OS 123 calls the initialization process of the debug monitor 124, and the debug monitor 124 stores the start as the state of the real-time OS 123 (sequence B5). Periodic command execution means 1
08 indicates panic to the debugger command execution means 104
Command execution request, debugger command execution means 1
04 executes the panic command (sequence B
6).

The debug monitor 124 is a real-time O
The setting information of the panic trap stored in S123 is read (sequences B7 and B8). Debug monitor 1
24, the panic trap setting information as the execution result of the panic command;
Send to 4. Real-time OS state storage means 109
Stores the activation as the state of the real-time OS 123 (sequence B9). Periodic command execution means 108
Requests the debugger command execution means 104 to execute the g command, and the debugger command execution means 104 executes the g command to run the program of the exchange 120. Since the state of the real-time OS 123 has not transitioned from activation to non-activation, the failure detection unit 110 determines that no failure has occurred (sequence B10).

As long as no failure occurs in the exchange application program 121, the sequences B6 to B10 are repeated. Exchange application program 12
When a failure such as an address error or a bus error occurs in the first device, an exception interrupt occurs, and control is transferred to an exception handler of the real time OS 123 (sequence B11). When the exception handler of the real-time OS 123 executes the instruction at the fixed address of the exchange 120, the control is transferred to the bootstrap program 122. The bootstrap program 122 calls the initialization processing of the debug monitor 124, and the debug monitor 124 O
The inactive state is stored as the state of S123 (sequence B
12).

The cyclic command execution means 108 requests the debugger command execution means 104 to execute a panic command, and the debugger command execution means 104
The command is executed (sequence B13). The debug monitor 124, as an execution result of the panic command,
The stored state of the real-time OS 123 that is not yet started is transmitted to the debugger command execution unit 104. The real-time OS state storage means 109 is provided in the debug monitor 1
The non-start state, which is the state of the real-time OS 123, received from the server 24 is stored (sequence B14). The periodic command execution unit 108 requests the debugger command execution unit 104 to execute the g command, and the debugger command execution unit 104 executes the g command to run the program of the exchange 120. The failure detection unit 110 determines that a failure has occurred because the state of the real-time OS 123 has transitioned from started to not started (sequence B1).
5).

As described above, according to the present embodiment, the cyclic command executing means 108 periodically executes the debugger command, and the state transition of the real-time OS received from the debug monitor 124 causes the failure detecting means 11 to execute.
Since 0 detects the occurrence of a failure, the occurrence of a failure in the exchange application program 121 can be efficiently and reliably detected. In addition, when the failure detection unit 110 detects the occurrence of a failure, the command acquisition unit 101 reads the restart procedure file and performs the restart processing, so that the automatic test can be continuously performed even when the failure occurs.

It should be noted that, instead of the panic command executed in steps A6 and A15 (see FIG. 2) and sequences B2, B6 and B13 (see FIG. 3), an em command for setting information reference (step C6 in FIG. 4) is used. , C
15 and the sequence D2, D6, and D13) of FIG.
The em command is used when an exception such as an address error or a bus error occurs in the exchange application program 121, to set an exception trap for outputting a number corresponding to the type of the occurred exception to a standard output, and to set an exception trap. This is a debugger command for canceling and referring to the setting information of the exception trap.

That is, as shown in FIGS. 4 and 5,
When the em command is executed for referring to the setting information of the exception trap, as an execution result, the setting information of the exception trap when the real-time OS 123 is activated, and the non-start as the state of the real-time OS when the real-time OS 123 is not activated, Received from the debug monitor 123. Panic periodically executed in the above-described embodiment.
The command and the em command periodically executed in this embodiment are debugger commands that can receive the state of the real-time OS 123 from the debug monitor 124 without changing the setting information stored in the real-time OS 123. . Similarly, the occurrence of a failure can be detected by periodically executing a debugger command capable of receiving the state of the real-time OS 123 from the debug monitor 124 from the periodic command execution unit 108.

[0037]

According to the present invention, a debugger command is periodically executed, the presence / absence of a failure is detected from the state of the real-time OS corresponding to the execution of the command, and when the failure is detected, a predetermined restart procedure is performed. Therefore, the occurrence of a failure in the exchange application program can be detected efficiently and reliably, and the automatic test can be continuously performed even when a failure occurs.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an automatic test system according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an overall operation in the automatic test system.

FIG. 3 is a sequence diagram showing a failure detection processing sequence.

FIG. 4 is a flowchart illustrating an overall operation when an em command is used.

FIG. 5 is a sequence diagram showing a failure detection processing sequence when an em command is used.

FIG. 6 is a block diagram for explaining a conventional example.

[Explanation of symbols]

Reference Signs List 100: workstation, 101: command acquisition unit, 102: command inspection unit, 103: command discrimination unit, 104: debugger command execution unit, 105: speech path system simulation unit, 106: online command execution unit, 107: execution result storage Means, 108: periodic command execution means, 109: real-time OS state storage means, 1
DESCRIPTION OF SYMBOLS 10 ... Fault detection means, 120 ... Exchange, 121 ... Exchange application program, 122 ... Bootstrap program, 123 ... Real-time OS, 124 ... Debug monitor

Claims (3)

[Claims] 1. An automatic test system, which is executed on a real-time OS and automatically tests exchange software embodying a processing operation of an exchange, wherein a debugger command is periodically executed, and the real-time OS corresponding to the command execution is executed. And a restart means for executing a restart command for restarting the exchange in a predetermined restart procedure when the fault detection means detects a fault occurrence. And an exchange software automatic test system. 2. An automatic test system, which is executed on a real-time OS and automatically tests exchange software embodying a processing operation of an exchange, comprising: a cyclic command execution means for periodically executing a debugger command; State storage means for determining and storing the state of the real-time OS from the execution result of the means; failure detection means for detecting the presence or absence of a failure according to the state of the real-time OS stored in the state storage means; An automatic testing system for exchange software, comprising: a restart procedure acquisition unit that reads a command for restarting the exchange from a restart procedure file and executes the command when the failure detection unit detects the occurrence of a failure. 3. The automatic test system for exchange software according to claim 1, wherein the debugger command outputs information indicating a cause of the inability of the real-time OS to operate.