JPS59123946A - System control system - Google Patents

Abstract

PURPOSE:To switch automatically connections of an input/output device to an in-use system central controller by performing time supervising by a timer unless an interrupting signal reporting that a stand-by system central controller establishes its normal state arrives within a specific time. CONSTITUTION:A fault occurs to the stand-by system central controller CC1, which is under repairs; and an external storage device FPD and a typewriter device TYP are connected to the conroller CC1. In this case, the inputerrupt signal reporting the establishment of the normal state is not sent out within the specific time unless the repair of the controller CC1 is complete, so the in- use system central controller CCO performs the time monitoring by the timer TO. Consequently, a switching control part SO switches connections of the devices FPD and TYP to the device CCO when the device CCO recognizes that the repair of the device CC1 is not complete. Thus, even when the repair of the device CC1 is incomplete, the input/output device is switched automatically.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field of the Invention The present invention relates to a system control system, and particularly in an information processing system equipped with an input/output device shared by an active central controller and a backup central controller. The present invention relates to a system control method in which input/output devices connected to a central control unit are controlled to be automatically connected to an active central control unit when it is determined that a standby central control unit is in a failure state.

Technical background In information processing systems that require high reliability, such as electronic exchanges, the central control unit that is the core of exchange processing is duplicated, and one side is operated as the active system.
A so-called standby redundant configuration is adopted in which the other system is kept on standby as a standby system.

In this type of information processing system, if the active central controller fails, the standby central controller immediately switches over to the active system and starts operating, minimizing the interruption of operation of the information processing system. .

The affected central control unit was disconnected from the relevant information processing system and underwent fault diagnosis and repair.

After the normality is confirmed, it will be re-incorporated into the information processing system.

Prior Art and Problems FIG. 1 shows an example of a conventional system control method for this type of information processing system. FIG. 1 shows duplicated central control units Cco and ccl, an external storage device FPD such as a floppy disk, and a typewriter device TYP shared by the central control units CCO and ccl. External nuclear storage unit F'PDu central control "location c0"
The operating system required for the operation of the CCI, the central control unit CCO, and a repair confirmation test program to be used after the CCI has been damaged and a fault has been repaired are stored. And the typewriter device TYP is this I
It is used to start up during PL, start up a repair confirmation test program, and output the test results. And external storage device FPD and typewriter device T
YP connects these central controllers CC via external storage controller FDC and typewriter controller TPO, respectively.
It is connected to either O or CCI. In this case, from the point of view of system economy, the external storage device FPD and typewriter device TYP are not duplicated.

The central control unit CCO is currently operating as the active system.

It is assumed that the external storage device FPD and the typewriter device TYP are connected to the central control device CCO.

On the other hand, it is assumed that the failed central control unit CCI has been repaired as a standby system. Then, in order to confirm the normality after repair, the central control unit CCO uses the typewriter device T.
In response to a command input from YP, the external storage device FPD and typewriter rehearsal TYP are connected to the central control device CC1. Then, the central control unit CC1 performs IPL from the external storage device FPD, stores the operating system in the main storage device (not shown), and performs a short test operation using the repair confirmation test program stored in the external storage device FPD. After completing the normality checking process, the central controller CCI sends the system controller 5
Sends the offline completion to the active system central control unit CCO via the COD and transmits the interrupt signal,
Report repair completion. When receiving this interrupt signal, the central control unit CCO reconnects the external storage device FPD and the typewriter device TYP to connect them to the active system.

However, if the central controller CC1 has not yet been completely repaired, the central controller CCO
Since the interrupt signal is not transmitted from C1 for any length of time, the external storage device FPD and typewriter device TYP remain connected to the central control device CCI, and these external storage device 11'PD and typewriter device TYP are To connect again to the active central control unit CCO.

For example, maintenance personnel had to operate switches on the operation panel. By the way, there are many other important switches installed on this operation panel, and if they malfunction, it could lead to a serious accident, such as a system failure. It is not recommended to do so.

Therefore, in the conventional system as shown in Fig. 1,
The active system central control unit CCO - indicates the external storage device FPD and typewriter device TYP as the standby system central control unit CC.
1, the external storage device FPD or typewriter device'rYP must be connected to the central controller CCO again until the offline completion interrupt signal is received from the central controller CCI. I don't have the means to do it,
If the central control unit CC1 cannot send out this interrupt signal, it is necessary to change the connection through human intervention.

OBJECTS OF THE INVENTION It is an object of the present invention to solve the above-mentioned problems by providing an interrupt signal that reports the establishment of a normal state from a standby central control unit if it is interrupted for a certain period of time.

To provide a system control method in which an input/output device is automatically connected to an active central control device without human intervention.

Structure of the Invention To achieve this object, the system control method of the present invention includes an active central controller, a backup central controller, and an input/output device shared by the active central controller and the backup central controller. An information processing system comprising: an instruction information display means for displaying instruction information for transmitting status information of the standby central control unit to the active central control unit; When the display is not performed, it is determined that a fault exists and the active central controller connects the input/output devices connected to the standby central controller to the active central controller. It is characterized by being controlled as follows.

Embodiment of the Invention An embodiment of the present invention will be described with reference to FIGS. 2 and 3.

FIG. 2 is a configuration diagram of an embodiment of the present invention, and FIG. 3 is an explanatory diagram of its operation. The same reference numerals as in other figures indicate the same objects.

The central control unit CCO is provided with a timer To, a switching control unit SO, and a register RO, and the central control unit CC1 is also provided with a timer Tl, a switching control unit S1, and a register R1.
is provided.

The timer TO monitors the set time of the offline completion signal (hereinafter referred to as OFC) written in the register R1 of the central control unit CC1. OFC written in this register R1 is an interrupt signal, and 5CO
D to the central control unit CCO.

The switching control unit SO controls switching of the external storage device FPD and the typewriter device TYP to the central control unit CCO.

The OFC output from the central control unit CCO is written in the register RO, and will be used when the central control unit CCO becomes a standby system.

A timer Tl provided in the central controller CC1.

The switching control unit S1 and the register R1 are respectively the timer TO1 switching control unit S of the central control unit CCO.

and performs the same operation as register RO.

Here, as in FIG. 1, the central control unit CCo is currently in operation, and the external storage device FPD and typewriter device TYP are connected to the central control unit cco. It is assumed that the affected central control unit CCI has been repaired as a standby system.

Then, control as shown in the flow shown in FIG. 4 is performed. That is, in order to confirm the normality after this repair, the current central control unit CCO sequentially reconnects the external storage device FPD and the typewriter unit TYP to the central control unit CC1 in response to a command input from the typewriter unit TYP.

Then, the central controller CCO performs 5COD and sends an IPL activation signal to the central controller cc1. As a result, the central control unit CC1 performs IPL, and then the central control unit cC1 executes a repair test based on the repair confirmation test program stored in the external storage device f'PD. At this time, if the central control unit CC1 has been completely repaired, the state will be as shown in FIG.

After a certain period of time To after starting the IPL operation, a flag OFC indicating IPL success is set in register R1.

This becomes an interrupt signal and is confirmed by the central control unit CCO via 5COD. Then, when the test by the above test program is completed, the flag OFC is again stored in the register R1 after a certain period of time Tl, and an interrupt signal is sent to the central control unit CCO.
CO recognizes that CC1 has been completely repaired. As a result, the central control unit CCO switches the connection of the external storage device PPD and the typewriter device TYP to the current central control unit CCO by the switching control unit SO. in this case.

Since the above To and Ti are known, the central control unit CCO can monitor the time after sending the IPL start start signal and can monitor this using the timer TO.

If the repair of the central control unit CC1 is not complete, the IP
If OFC is not set in the register R1 even after To has elapsed after the start of L, the first interrupt signal indicating IPL success will not arrive. Also, even if IPL is successful, then T
If the OFC is not set again after l, the repair confirmation process by the test program will not end, so if these conditions are monitored by the timer To, the status of the central control unit CC1 can be grasped.

In this way, when the central controller CCO recognizes that the repair of the central controller CC1 is not complete, the switching controller SO switches the external storage device FDC and typewriter device TYP to the current central controller COO side. Change the connection. This makes it possible to automatically switch these input/output devices even when the central control unit CCI is incompletely repaired.

As another embodiment of the present invention, if OFC is set in the register R1 (when the central controller CC1 is in the standby system) at each step as shown in FIG. can be monitored.

That is, first, the current central control unit CCO outputs an IPL start signal to CC1 and then starts time monitoring. And time 1. Later, the backup system central controller CC1
The first OFC indicating acceptance of LROM activation is set in register R1 as shown in FIG. Then, after the known tl, CCI bootstraps the program storage area for IPL as a result of testing the program storage area - the second OF which indicates memory test OK.
Stand C as shown in ■. Then, after t2, the third OFC, which indicates that the loading of the bootstrap program, which is a program for accessing the external storage device FPD and performing IPL, was successful, is stored in register R1 as shown in ■. . In this way, for each step thereafter, for example, OFC■ for memory test O' which shows the result of checking the program loading area of the memory, and OFC■ for the memory test O' which shows the result of checking the program loading area of the memory, and the spare system which shows the success of loading the spare program from the external storage device FPD. program·
OFC■ for loading OK, OFC■ for outputting a system establishment message OK to indicate that the system establishment message has been outputted to the typewriter device TYP in order to output the system establishment message to maintenance personnel as a type, and the backup system according to the test program. OFC for backup system processing and termination indicating the completion of processing
+click. This OFC■ allows the central control unit CCO to
The input/output device will be reconnected to the active system.

In this way, it is the time from the sending of the IPL start signal to the setting of OFC■ indicating the end of standby system processing.

That is, by monitoring the setting state of the OFC during the maximum monitoring time, the state in the middle can be checked. In this way, the failure status can be determined by the number of times the OFC bit is turned on. Note that this maximum monitoring time is known and can be input by maintenance personnel when inputting the offline system establishment command.

Other embodiments of the invention are shown in FIGS. 6 and 7.

In FIG. 6, processors CPRO~15/5VPO~4, which operate as call processors (hereinafter referred to as CPR) or service processors (hereinafter referred to as SvP), are managed by a management Φ processor (hereinafter referred to as MPR). For example, in this example, a maximum of 16 units can be connected for CPR connection, and a maximum of 5 units can be connected for SVP connection.
When P is mixed, a maximum of 21 units can be connected. The MPR has central control units CCO and CC1, one of which operates as an active system and the other as a standby system, and a system control unit (hereinafter referred to as 5).
(referred to as COD). CPR is a communication processor, and SVP performs billing management and mailbox control. And CPRO~15/5vPO~4 is also a processor CCσ.

It has a redundant configuration with CC1'', and is connected to the O-system 2-in LO and the l-system 2-in LL by a system control unit (hereinafter referred to as 5coc) as shown in FIG.

For the central control units CCO and CC1 on the MPR side.

Status display register ESR-M for each category shown in Figure 0)
is provided, and a processor CCO on the CPR/SVP side is provided.
CCI° is provided with a status display register MSR-C for each section shown in FIG. 7(b).

The IPBL of the status display register ESR-M shown in FIG.
1 indicates that a communication bus disable signal indicating that the communication bus (not shown) has become incapable of communication is received, and the CPR/S
VP writes. This communication bus also has a dual structure,
This results in switching the communication bus.

The EST is a system switching instruction that instructs the CPR/SVP to switch between the active system and the backup system, and is entered by the MPR side.

15TI is for the MPR side to write an IPL start signal instructing the CPR/SvP standby processor to start IPL.

When the main controller of the active system on this MPR side enters the classification,
It becomes an IPL (start signal) that instructs PL start-up, and becomes an OFC when written by the standby central control unit.

Therefore, if the current central control unit reads this, it will be able to recognize the same information in each of the above embodiments.

PN3 to PNO indicate the CPR/5VPO machine number, and since up to 21 CPR/SVPs can be connected with this system, a specific one can be selected using these 4 bits, which must be entered on the MPR side. It is.

IPBL of the status display register MSR-C shown in FIG. 7(b) is a communication bus disabled display that notifies the MPR that the communication bus has become disabled.

MPR will read this.

EST becomes a system switching instruction reception indication indicating that a system switching instruction has been received from MPR. CPR/SVP reads this and switches the reading system.

The present invention can also be implemented in such a system control unit as shown in FIG.

Effects of the Invention According to the present invention, the status of the standby system can be confirmed by simply monitoring the status notification signal sent from the standby system in the active system after notification of the IPL start signal.

If a certain number of devices are not sent out within a predetermined time, the input/output devices connected to the backup system can be automatically switched to the active system, improving system safety. .

[Brief explanation of the drawing]

Fig. 1 is a conventional system control method, Fig. 2 is a configuration diagram of an embodiment of the present invention, Figs. 3 and 4 are illustrations of its operation, and Fig. 5
The figure is an explanatory diagram of the operation of another embodiment of the invention, and FIGS. 6 and 7 are diagrams illustrating the configuration of still another embodiment of the invention. In the figure, cco and cal are the central control device, 5COD is the system control device, and FPD is the external storage device. TYP is a typewriter device, To and Tl are timers. RO and R1 are registers. Patent Applicant: Fujitsu Ltd. Representative Patent Attorney Akira Yamatani f'I IPL U)l--11N
oF (--OH S flash Cn (mouth> tcpg/sw) ESft-C Ento Sei=〒yo=;=]

Claims (1)

[Claims] In an information processing system comprising an active central controller, a backup central controller, and an input/output device shared by the active central controller and the backup central controller,
An instruction information display means is provided for displaying instruction information for transmitting status information of the standby central controller to the active central controller, and when the instruction information display means does not display the status information within a certain period of time, It is characterized in that, upon determining that a failure exists, the active system central controller controls the input/output devices connected to the standby system central controller so that they are connected to the active system central controller. System control method.