JPS60223283A - Fault processing system - Google Patents

Abstract

PURPOSE:To know the state of a fault and to perform analysis of the fault by saving the contents of an own system or an remote-side system to a magnetic memory and printing out them before the replacement processing is restarted when the fault occurs. CONSTITUTION:A single emergent working circuit EMA100 is connected to both center controllers 10 and 11 having double structures. Then the EMA100 detects the faults of each device and selects either one of controllers 10 and 11, main memories 20 and 21 and magnetic disk devices 30 and 31 respectively. The contents of the address areas set within the main memories are transferred to devices 30 and 31. Then an indication is given to the controllers 10 and 11 to restart the replacement processing. The system consitution is selected together with the designation of contents of either one of both main memories that should be transferred to the disk devices according to the value of a counter set in the circuit EMA100.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to failure restart processing for electronic exchanges, and particularly to the collection of data necessary for failure search.

Description of the Prior Art Because failures in electronic exchanges have a large impact, a redundancy system is adopted to prevent a single failure from causing a system shutdown. When a failure occurs, the priority is to make effective use of redundancy, select the minimum viable system configuration, and start the replacement process, and collect data on the failure situation for repair. , after reopening. This has the disadvantage that the data that can be collected is limited, and that the software accesses the failed device to collect data, thereby affecting the replacement operation again. In particular, in the case of a software failure (bug), it is difficult to reproduce the failure, so data at the time of failure is extremely important.

Purpose of the Invention The present invention has been made in view of the above-mentioned conventional circumstances.
Therefore, an object of the present invention is to save the contents of the main storage device in a magnetic disk device after a failure occurs and before handing over control to the software to restart the exchange operation, which was previously done by software in the process of restarting the exchange operation. A novel method that eliminates the shortcomings of the morbidity data saving method uses hardware and a microprogram built into the central control unit to save the contents of the bidirectional storage device to a magnetic disk device before handing over control to the software. The objective is to provide a failure handling method.

Structure of the Invention In order to achieve the above object, the failure handling method according to the present invention consists of a central control unit (aC), a main storage unit (HLM), and a magnetic disk unit (DKU) which are duplicated, and which of the central control unit In an electronic exchange having an interconnection means for accessing the main memory or magnetic disk device fkK of both systems, a single emergency operation circuit (gMA) is connected to both central control units, and the emergency operation circuit is connected to each device. The central control unit which detects the failure of each of the above and which is duplicated,
The central controller selects one system each, either the main storage device or the magnetic disk device, transfers the contents of the specified address area in the main storage device to the magnetic disk device, and then restarts the exchange process. The present invention is characterized in that the configuration of the system is selected and the contents of the main storage device to be transferred to the magnetic disk device are specified based on the value of a counter in the emergency operation circuit.

3. Detailed Description of the Invention Next, a preferred embodiment of the present invention will be specifically described with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of the present invention. The emergency operation circuit (EMA circuit) 100 is connected to the central control unit (
OCo ) 10 and central control unit (cal) 11 (7
) with connections to both central control units. The necessary power for the gMA circuit 100 is supplied by the two power supplies 12 and 13 of the central control unit (POW units provided in 10 and 11), so that the circuit does not stop even if one power supply is cut off.

FIG. 2 shows a more detailed block configuration of the central control circuit and KMA circuit. If a failure occurs in any central control circuit 10゜ll, it will be sent to D through the failure detection report line 14.15.
It is detected by the MA circuit 100K. Failure detection circuit 1
001 sets the counter 1003, and the counter 10
05-based system configuration setting circuit 1002 is driven.

Next, the system configuration setting circuit 1002 performs initial settings necessary for the hardware to pass control to the software, that is, clock selection, master/slave settings for the central controller OC, and operation mode settings according to Table 1. -Do. Table 1 is a correspondence table between the counter 1003 and status settings. After that, pass the control lines 16 and 17 to the central control unit [0
0 microprogram is started. Central control unit & aC
The microprogram transfers the contents of the main storage device to the magnetic disk unit DKU and loads the program in the magnetic disk unit DKU to the main storage device 1: MM according to the contents of the counter 1003 in the DMA circuit 100 according to Table 1. and then passes control to the software.

If there is a failure in the device in the system configuration set above, the DMA circuit 100 detects this condition again and increases the counter 1003.
1 and perform the following state settings.

Table 1 After restarting the software, the operator must use the magnetic disk device t.
By printing out the contents saved in the DKU, it is possible to know the failure situation and perform failure analysis.

Effects of the Invention As explained above, the present invention saves the contents of the main storage device of the own system or the partner system in the magnetic storage device before restarting the replacement process when a failure occurs.

(1) Raw data can be collected when a failure occurs; (2)
), by executing all combinations, it is possible to collect data from the main storage devices of not only the own system but also other systems without reducing the possibility of restarting the system. (This is particularly important because it is highly likely that this occurred in the central control system on the opposite side.)

[Brief explanation of the drawing]

FIG. 1 is a system configuration diagram showing an embodiment of the present invention. 10... Central control unit (GCjo), u... Central control unit (G(31), 12... Power supply unit for CaO, 13... Power supply unit for OG2), 100...
Emergency operation circuit (FfMA), addition...main memory device i
lt (MMO), 21-=main memory (MMl
), Ka...Magnetic memory bag [(DKUO), 31.
...Magnetic storage bag k (DKUl) FIG. 2 is a block configuration diagram showing details of the emergency operation circuit. 14.15...Report line for failure detection, 16.17...
System configuration setting control line, 18.19... Counter reading line, 101, 111... Micro program ROM,
1001...Failure detection circuit, 1002...System configuration setting circuit, 1003...Counter Patent applicant NEC Corporation Representative Patent attorney Yutabe Kumagai Figure 2

Claims (1)

[Claims] Each has a redundant central control unit, a main storage unit,
The only failure detection system reconfiguration function in an electronic switching system that consists of magnetic disk devices and has central processing units interconnected so that each of the central control units can access both main storage devices or magnetic disk devices. An emergency operation circuit having a function is connected to both of the central control units, and the emergency operation circuit detects a failure in the central processing unit and activates one of the duplexed central control unit, main storage device, and magnetic disk device. select one system, transfer the contents of the main storage device at a specified address to the magnetic disk device, and then instruct the central controller to restart the exchange process, and select the system and which main storage device. A failure handling method characterized in that the contents of a counter provided in the emergency operation circuit are used to specify whether or not the contents of the file should be transferred to the magnetic disk device.