JPH03266133A - Line system fault processing system in hot stand-by system - Google Patents

Abstract

PURPOSE:To carry on the normal operation of a system by performing system switching, the initialization of a communication controller, and a reconnection with an external terminal when a communication controller (FNP) fault which does not shut down the system directly occurs and the number of lines degenerates exceeding some extent. CONSTITUTION:If a fault occurs to the FNP 7, an operating system (OS) 14 detects the fault in the process of communication with the FNP 7, decides the fault influence range, and informs a monitor program 11, where it is decided whether or not lines degenerate more than predetermined. If the lines degenerate exceeding the constant number, the monitor program 11 instructs the OS 14 to switch the system and OS 34 stops the on-line transaction processing program 12 and informs a stand-by OS 34 of the system shutdown. Then a monitor program 31 take the resource which is used by the in-use system in its system, initialize the FNPs 6 and 7 temporarily, and connects the FMP 7 to an external terminal again. Consequently, the normal operation is carried on.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a hot standby system, and is particularly characterized by a system switching control method when a line degeneracy failure equivalent to a failure occurs in system operation. Regarding the hot standby system.

[Conventional technology]

In recent years, systems are increasingly required to have high reliability, and even if a system goes down, it has become essential that the system be restarted immediately.

Conventionally, this type of information processing system has a standby system that is always operational as a standby redundant system, and when a failure occurs in the active system, it automatically switches from the active system to the standby system at high speed, and the system is maintained. There is a hot standby system that continues operation with a short online down time.

[Problem to be solved by the invention]

In the conventional hot standby system mentioned above,
Even if a failure occurs, if the system does not go down, system switching will not be actively performed. However, since the system is operated online, if a failure occurs such as a failure in the interface between the host and the communication control device, even if it is not a failure of the communication control device itself, the line will be significantly degraded. Although operations continued on the host side, it appeared to many operators on the terminal side that the system was down, causing a serious inconvenience from the perspective of system operation.

[Means for Solving the Problems] The configuration of the line failure handling method in the hot standby system of the present invention is such that the active system central processing unit and the standby system central processing unit are connected by an intersystem communication device, A communication control device that is interconnected from the active and standby central processing units and functions as an active or standby system is connected to a line switching device, and the standby system is connected to the same online network as the active system. 2. The unauthorized attack processing program is placed in a standby state immediately before the start of execution, and the active central processing unit is able to communicate with the standby central processing unit via the intersystem communication device using a monitoring means. The monitoring means, after switching to the backup communication control device due to a failure in the current communication control device, monitors the degeneration status of the lines owned by the backup communication control device, and detects whether the number of lines has degenerated below a certain number. If so, the active system's online (2) processing program is stopped, and a system down notification is sent to the standby system central processing unit via the intersystem communication device, and the standby system central processing unit is placed on standby. After starting the two online transaction processing programs of the standby system and initializing the line status of the standby communication control unit that was in use with the active central processing unit, the system was switched. The feature is that system operation can be resumed in the current state.

〔Example〕

Next, one embodiment of the present invention will be explained with reference to the drawings.

FIG. 1 shows a configuration diagram of an embodiment of the present invention, and the current system is as follows:
Main storage device i1, central processing unit (hereinafter abbreviated as CPU)
2. The standby system includes a main storage device 3 and a central processing unit 4. The CPU 2 and CPU 4 are connected by an intersystem communication device 5, and a communication control device (hereinafter abbreviated as PNP) 6.7.
They are connected to CPU2 and CPU4, respectively, and connected to line switching device 8, and one of them (in this example, FNP6
) is connected to an external terminal via a line and operated online.

In addition, the main storage device IK is loaded with a monitoring program 11, an online transaction processing program (hereinafter abbreviated as 0LTP) 12, a communication control program 13, and an operating system (hereinafter abbreviated as O8) 14. , 0814.

0814 controls the basic operations of the computer system, and the communication control program 13 receives messages from external terminals connected through the line via the FNP 6 and processes them so that they can be used within the computer. and handed over to 0LTP12. The 0LTP 12 performs specific processing on the received data as determined by the system, stores the processing results in a database (not shown), and transmits the processing result responses to the external terminal via the FNP 6 using the communication control program 13. is being sent to.

The monitoring program 11 monitors the occurrence of a system down, issues a system down notification via the intersystem communication device 5 when the system goes down, and also monitors the degeneration status of the number of lines. A similar program is loaded on the standby main storage device 3, and all files to be used are available for use.
When a system failure occurs in the active system, the monitoring program 31 receives the system down notification sent from the active system under the control of the 08340, and furthermore, the monitoring program 31 receives the system down notification sent from the active system. (not shown) into its own system, starts execution of 0LTP32, completes system switching processing, and is in a state where it can immediately resume processing on the active system.

Now, when a failure occurs in the active system PNP 6, a switching instruction is issued to the line switching device 8 via the FNP 7 under the control of the OS 14, and when the PNP 7 is connected as the active system PNP, the KFNP 7 If a failure occurs, there will be no backup PNP, so depending on the scope of the failure, operation will continue with some lines degraded, or operation will stop due to online downtime. .

However, there are cases where a PNP failure is recognized as a PNP failure even if there is no abnormality in the PNP itself, such as an intermittent PNP common part failure or a 7-part failure of the CPU-PNP interface. Therefore, in the present invention, when a PNP7 failure occurs, the system is controlled by performing K as described below.

When a failure occurs in PNP7, 0814
A failure is detected in the process of communication to 7, the range of influence of the failure is determined, and the monitoring program 11 is notified. Upon receiving the notification of FNP7 failure from 0814, the monitoring program 11 determines whether the line has been degraded more than a predetermined amount. A failure in the interface between the CPU2 and PNP7 is recognized as a total line failure or a partial line failure.

If the number is less than a predetermined number, the operation will continue as is, but if the number of lines has degraded to more than a certain number, it will be determined that the system function cannot withstand operation, and the monitoring program 11 will call 0814. Instructs the system to switch. Upon receiving the system switching instruction from the monitoring program 11, 0814 stops 0LTP12 and switches to standby system 083.
4, a system down notification is sent. When the standby system 0834 receives the system down notification, it notifies the monitoring program 31, and the monitoring program 31 also imports the resources (not shown) used in the active system into its own system, and
After initializing P6.7, PNP7 is reconnected to the external terminal. Furthermore, execution of 0LTP32 is started, system switching processing is completed, processing of messages sent from external terminals is started again, and system operation is continued.

4L, if the lines are degraded due to a failure in the interface between CPU2 and PNP7, these lines will be
The operations available via U4-FNP7 will continue as they are.

〔Effect of the invention〕

As explained above, even when a PNP failure occurs that does not cause a direct system down, the present invention can perform system switching, initialization of communication control equipment, etc. if the number of lines on a fixed board is degraded.
(K) and reconnecting with external terminals) can help in cases where system operation stops even if there is no failure in the communication control device itself, such as a failure in the interface between the central processing unit and the communication control device. This has the effect of allowing the system to continue operating normally.

FIG. 1 is a system configuration diagram of an embodiment to which the present invention is applied.

°1, 3... Main memory, 2.4...
Central processing unit, 5... Intersystem communication device, 6, 7.
...Communication control device, 8...Line switching device, 11.31...Monitoring blogger, 12,32
...Online transaction processing program, 13.33...Communication control program, 14
．． 34...Operating system.

Claims (1)

[Claims] An inter-system communication device connects an active system central processing unit and a standby system central processing unit, and communication between the active system and standby system central processing unit is connected to each other and functions as an active or standby system. and a control device are respectively connected to a line switching device, the standby system puts the same online transaction processing program as the active system in a standby state immediately before starting execution, and the active central processing unit controls the monitoring means. The monitoring means can communicate with the backup central processing unit via the inter-system communication device, and after switching to the backup communication control device due to a failure in the current communication control device, the monitoring means The communication control device monitors the degeneracy status of lines possessed by the communication control device, and when the number of lines degenerates below a certain number, it stops the online transaction processing program in the active system, and executes the central processing program in the backup system via the intersystem communication device. A system down notification is sent to the device, and the standby central processing unit starts the standby online transaction processing program, and the standby communication control unit that was being used by the active central processing unit A line system failure handling method in a hot standby system characterized by restarting system operation with the system switched after initializing the line status.