JPS585858A - System restarting system - Google Patents

Abstract

PURPOSE:To accomplish the restarting of a system within the shortest period by checking the status of each doubling device at the switching of the system in case of a system trouble by a means for storing the status of each doubling device and also using a storage device of which contents can be insured. CONSTITUTION:A control unit CC#1 is started by switching operation and then a discrimination circuit ANN1 is also started. A comparator CMP checks whether a main storage device MM#1 is normal or not. If no trouble is detected, a signal N is outputted and a phase 1 is restarted from the MM#1. If the contents of a main storage device MM#0 is incorrect, an output Y starts a discrimination circuit ANN4, which checks a status signal FM0. When a FM#1 is normal, restarting (RSFM) is started. When the FM#1 is incorrect, a discrimination circuit ANN5 checks the status of a magnetic storage device CMT. When the CMT is normal, restarting RSCMT is started. When the CMT is defective, the MM#1 executes the restarting (RSMM). When the MM#1 is defective, the FM#1 is checked and, when the FM#1 is normal, it starts restarting. If FM#1 is defective, the CMT is checked; the restarting is started in case of the normal CMT and the system is downed by the defective CMT.

Description

DETAILED DESCRIPTION OF THE INVENTION This article relates to system reconfiguration in the event of a system failure that requires switching of system systems including control equipment pipes in a duplex system. tK system 1 reprocessing lIK
It is related to

In general, electronic switching equipment, etc., has a high degree of reliability. Required: In an online system, common devices such as control system devices and input/output devices are duplicated, and in the event of a failure in multiple control system devices, it is possible to identify and disconnect the faulty device programmatically. There is. In response to the failure of the control system equipment at t7, the hardware including the duplicated control equipment is switched, the failure is identified, and the system is reconfigured to return to normal #1.

Conventionally, when a failure occurs that requires switching of the system including the control device, that is, a system failure, the entire system is switched, the system is started up, and when the system cannot be restarted, each affected system is restarted. The system is restarted by selectively switching the control system devices, that is, by sequentially switching the system configuration t.

Such a system restart method requires time to reconfigure and restart the system after a system failure occurs, which is a problem especially in processing systems with strict real-time performance. In order to solve the above-mentioned problems, the emergency control circuit KL9, which is equipped with a faulty device identification means 1, provides a system restart system that minimizes the restart processing time of the system.

In order to achieve the above object, the present invention uses a redundant system K consisting of a control device and a main storage device and an external storage device as controlled devices, and a fault state display register that displays the fault state of the controlled device. A counter for counting the number of emergency controls is provided in the % control device, and the system is reconfigured by switching from one system system to another.
1. The controllers of other system systems use the register m* indicating register and the value KL of the counter, and the storage device of the controlled system also uses a highly reliable storage device to manage system restart processing. Things to do 1? 411 images.

Hereinafter, the present invention will be explained in more detail with reference to Examples.

The first factor is a schematic configuration of a switching system as an example of a duplex system configuration.

In @, CCsu01CCφl is the control device, MMφ0
, MVφl is the main storage device, FMφ0, FMψ0 are 7-isle memories as external storage devices, CMT is a magnetic storage device as external storage device, NW is a switching network.
．． If file name H)v, l system xfb(CC*1s
MM+1,, FMφl) is in standby state] 1
1) Become a system.

At the initial startup of the system, the processing program subscriber data etc. stored in the magnetic storage device CMTK are stored in the main storage device MMφO (MMφ1) and the file memory FMφ0 (F
Mφ11KIPL (initial program loaded) and control device ccφ0 (CCφ11# operating KLp
The system is activated. File memory FMφQ (F
Mφl) is composed of an IC memory, etc., like the main memory device MMφ0 and NAL, and is indispensable for the exchange operation! It is used as a backup for the main storage device MMφO (MM◆1) which stores call processing programs and the like. tp This file message vFM
φO(FMφIIK is initially stored in main memory (a
- Contents to be downloaded), other data such as subscriber data, data that is used less frequently, data that is slow in processing time, etc. are stored.

With the above system configuration, the conventional system configuration O switching process lt 21st
This will be explained as 1g1K. 211 shows an example of the configuration of the duplicated part in FIG. 1, and the same symbols as in FIG. 1 indicate the same parts.
If a system failure occurs during operation in φ0), control of the system by the control device CCφ0KJC becomes impossible. Conventionally, when this system failure occurs, the emergency control circuit EMK activates each of the act CCφO and standby system CCφl.
Nt- applied, control device CCφ1t act state S! To, C.C.
The main storage device and file memory are set to MMφ0. FMφ0 to MMφ1, FMφ
I was switching to l.

After that, when the system cannot be restarted due to the contents of the main memory MMφl, the program is loaded from the file memory FMφl, the system is restarted, and if the normal process can be restarted, it enters the operating state IIK as an l-based system. p1 If it is still abnormal, main memory MMφ0
, file memory 1 Keiφ0 etc! Switch and connect to reconfigure the optimal system.

However, when the above method is used, when a system failure occurs, system switching is performed to isolate the faulty device, which takes time, and is not preferable for failure recovery during online processing such as a replacement system. .

In order to minimize the processing time for restarting the system, the present invention is provided with a means for storing the status of the redundant devices, and when a system failure occurs, the status of each device is checked at the time the system is switched. By using a storage device whose contents can be guaranteed, the system can be restarted.

A specific embodiment of the present invention will be described using FIGS. 3 to 5. 311f1 An explanatory diagram illustrating the relationship between the 0 system and 1 control device of the present invention! 4 is a configuration diagram of an emergency control circuit relating to system restart of the present invention, and 5111 is a configuration diagram of a system restart processing method of the present invention.

11:3sllK&In? , each control device CC+01CC
+1 is failure fall l1 indication register PI, Tφ0, FLT
φlt-, which makes it possible to indicate the state mt* of the controlled device no matter which control device is in the actuate state. Therefore, display register FLTφ01FLTφIC)! !
The contents are always stated in one sentence. That is, the first o411 of the present invention
11 is a point KT that makes it possible to identify the failure state s' of the controlled system device from any control device.

Ru.

If a system failure occurs during operation of the act system control device CCφOK system t1, the activation IN will be changed to each control device C.
Cφ0, CC IK entered, control device CCφ0 is CCφI
When K switches, the emergency control circuit EM in the control device CCφl
(See Figure 2) operates. The 4111th emergency control circuit has a 0IIPIII configuration, and has an emergency control circuit EMfx activation detection 11D.

The circuit includes a timer TI for seven bits, a counter EMAC that counts the number of activations, a decoder DEC, a control circuit EMAL, and a third
As explained in the figure ■Consists of display register FLT, MM
fl main storage, CMT lamp magnetic storage, and FMFX file memory. The MSD is an external maintenance signal placement device that relays maintenance side signals in exchange control and also relays 0-bit information to the state II display register FLT of the present invention. MSCN is a maintenance scanning device that can read devices where a failure has occurred.

Obstacle autumn l! The display register FLT is connected to the control device CCφ.
When O is in the act state ■, O failure information KJC is received from the maintenance scanning device M8CN, and the detected failure status III is set via the maintenance signal distribution device MSDt-.

When the control device CCφl enters the act state at 70, the number of times the emergency control command EMALT has been activated [the contents of the counting counter KMA and the contents of the fault status *display register KLT
The restart control circuit EMAL determines the phase of restarting the synutemno. This phase is restarted by unconditional KO-ding of the conventional emergency control circuit (for example, two or more times according to the value 1 of the counter EMAC). Phase 2 restart, restart from a fixed address in the main memory In addition to phase 1 restart, load the program from the file FM to the main memory MV and restart. . An emergency interrupt is generated in the circuit reset timer TE, and the emergency control IMAm CφIK is sent.

Next, explain to the restart part mt of the system of C - the component pipe t of the emergency control circuit EMAL of the 5th @,
The 4th @ and the t with the same sign indicate the same thing and indicate the fault.
FOI Fl e C71t'X Main storage device M
Mφ0% main storage device MM◆l.

Indicates the status of file memory FMφ0, file memory FMφ1, and magnetic storage device CMT. For example, 10” is normal;
= 1” is displayed as there is a failure.

S] CLoe 81LS, 5ICLlFXo series
The system's controlled device selection circuit, A'NNI~MUNNS, is a judgment circuit that judges whether the *mm pit content of the display register is "11 or @O", and the result is output 41. The consideration is @111 for L side, 10” for FX
It is output to the o side. CMP is a comparison circuit that determines whether the content of the emergency control counter EMAC is greater than or equal to a certain reference value, and when it is greater than or equal to the reference value, an output signal is output to N when Y1 is negative.

18MM is the 7A's IF that restarts the individual location Lp of the main memory! Open, R8CMT loads the program from the magnetic storage to the main memory and resumes from that fixed address.7Aze2 resumes, R8FM loads the program from the file memory to the main memory and resumes from the fixed address. 7A and 2 are respectively shown, and TSD indicates that even if the system is switched, the system cannot be started up, and the so-called system is down.

Next, the fifth @O restart processing operation will be explained. When the switching operation from the 0-system system to the 1-system system occurs and the control device CCφ1 is activated, first, the selection circuit 8E is activated in the control device CCφ1.
In this embodiment, the memory trowel tMM is used to input the signal S to be performed upon selection of the state information MO, M of the main memory device inputted with LoK.
The state tube of Φl is checked, and M is input to the determination circuit. Then, when the judgment circuit NNI is activated, the comparison circuit CMPt operates when the storage device MMφl is normal, that is, Ml=00. When is smaller than the predetermined value, the fault 1F that occurred in the past is
Judging that it will not affect the entire system, the output N is removed and the 7 aids 1 is reprocessed from a fixed address in the main memory MMsu.
Do l.

If the value of the counter EMAC is greater than or equal to a predetermined value, it is assumed that the frequency of failure occurrence is high and the contents of the main storage device are determined to be unreliable.
Start N4vt. First, whether or not the contents of the file memory FMO are normal (that is, reliable or not, IVtF,
t-Acquire and judge.

When the file memory FM1 is normal, that is, y, ao, the program is loaded from the file memory FMI to the main memory MMφ1 and the process is restarted from its fixed location! l(R8
FM) t-starting first order 1' When CF1=1, the judgment circuit NN5 is activated, and the normality/failure kcMw of the magnetic storage device CMT is taken in and judged.

If it is CM2O at this time, it is determined that the magnetic storage device CMT is normal, the program is loaded into the main storage device MM2, and the restart process R8CMT starts from the fixed address 1. However, if the magnetic memory device CMT is determined to be a failure (CM = 11), the contents of the main memory device MMφ1 are not determined to be a failure because M1 = O, but the counter value EMAC is determined to be greater than the specified value. In order to keep the system alive, restart processing is performed in the main memory fMMφl (R8
MM1゜Next, when the determination circuit ANNI determines that M1=1, and the main memory device MMsu1 is considered to be at fault, the determination circuit ANN2t- is activated by the output l, and the selection circuit 5EL1'
Check the file memory FMsu1 via YtF1'
If it is normal (FI ■ 0), the process is the same as the process performed when F1=O in the previous judgment circuit 4 (file memory FMφILf) restart section IIt-do
Fault (When F1=11, K starts the judgment circuit ANN3t and performs a check of the magnetic storage device CMT Vt0MK19. Similarly to CM=00 and the output 0 of the IU judgment circuit 5, the magnetic storage device CMT!ff is restarted. No, CM-1
In this case, K outputs the output 1 of the judgment circuit 3 to indicate that the entire 1p system has failed to start up 111t and the system is down, and an alarm [is raised] in response to operator intervention.

Thus, the system reprocessing m of the present invention is based on the past failure layer @KL
fJi&% Resume from a reliable storage device can be done in a short time.Although the embodiment of the present invention has been described with respect to system switching from system 0 to system 1, the effect is the same even if the reverse is true. In the first slIll, the selection signal 5Kjp of all the selection circuits 5ELO to 8EL2Vrll- is switched so that the system 1 system is selected as the device of the controlled system, but you can learn the input conditions of the selection signal for each selection circuit. In other words, it is possible to select either the KjJ) 0 system or the I system, and it is also possible to reconfigure the system by combining the controlled device pipe combinations of the 1kTj and 2 systems.

In addition, in this embodiment, writing to the failure state display register is performed by the control device CCφ0 (CCφl) using the maintenance scanning device M.
8C) Jl? As explained in the case where the F program processing IIK is used to determine and set the reading, the fault detection signal line of the main storage device or external storage device is drawn in in a wire drastic manner, and the fault state is indicated at 111%. As explained above, the present invention 11CL is a storage device that is most likely to match the state of the storage device stored in the failure status display register and has a short program loading time. Since it is possible to restart the system from the start, compared to conventional system reconfiguration methods, the system start-up time can be greatly reduced, and the return to online status is faster, contributing to maintaining system reliability by 1%. and
The effect is great.

[Brief Description of the Drawings] No. 111 is a duplex system, a schematic configuration diagram of an exchange system as one row of the system configuration, a configuration diagram of a doubled portion of the second color lamp No. 1II, and No. 3111t'! An explanatory diagram of the relationship between the 0 system and 1 system of the control device of the present invention, Figure 4 is a configuration diagram of an emergency control circuit related to system restart of the present invention, and Figure 5 shows the system restart process of the present invention. It is a block diagram of a method. CCφO, CCsu1: Control device, MMφ0. MM◆l:
Main memory, FMφO, FMsu: File memory, C
MT: Magnetic storage device, FLT (su0, φl); failure condition m
Display register, EMAC: counter, 5ELO to 8EL
2: Selection circuit, ANN1 to ANN5; Judgment circuit, CM
P: Comparison circuit.

Claims (1)

[Scope of Claims] In a redundant system comprising a control device and a main storage device and an external S storage device as controlled devices, there is provided a fault status m display register and an emergency w to display a group letter Mt of the controlled m device. [Each control device 11c is provided to count the number of times the system is switched. The system control device has a system restart section 41111 that uses a highly reliable storage device in the storage device of the fine control system according to the values of the failure status display register and counter. 1 method.