JP4543988B2 - Fault tolerant equipment - Google Patents

Description

  The present invention relates to a fault-tolerant device such as an information processing device that prevents a system from going down, and more specifically, in a replacement operation of a failed module, a risk of erroneously pulling out an operating module that is not a replacement target and causing the system to go down. The present invention relates to a fault-tolerant device for preventing the problem.

  The configuration of a conventional fault tolerant server (ft server) will be described with reference to FIG. A fault-tolerant server has a dual circuit configuration to avoid a system failure, and while one system is operating, the other system is also operating and waiting, and if one system fails Prevents the occurrence of system down by operating the system in a standby system. During this time, the system that failed is repaired, and the duplex control is performed again to prevent the system from going down. As fault-tolerant technology, a triple structure is also known. In this case, the three systems are operated while being synchronized, and information is input / output by taking a majority vote, thereby improving the reliability of the information and preventing the occurrence of system down. During the triple control, if one of the three systems fails, two systems will control the duplex, and if two systems fail, operate the system with the remaining single system. To prevent system down.

  In FIG. 7, the fault-tolerant server 1 includes a housing 2, a backplane 3, a module 4, and a module 5. The module 4 and the module 5 are inserted into the housing 2 from the left side in the drawing, and are connected to the backplane 3 so as to be freely inserted and removed by a connector or the like. The module 4 and the module 5 are the same HW (hardware), and the duplex control circuits 4d and 5d communicate with each other via the backplane 3 to perform the duplex control as described above. There are mainly three statuses of module 4: “duplex operation in progress”, “operating in one system”, “stopped”, and duplex control circuit 4d displays status information of module 4 via signal line 4p. Notify the control circuit 4e. The LED control circuit 4e determines the lighting color of the LED (light emitting diode) 4g according to the module status, and transmits an LED control signal, which is a lighting instruction signal, to the LED 4g via the signal line 4q. The LED 4g lights a specified color according to a lighting instruction from the LED control circuit 4e. The module 5 also performs the same LED control as the module 4. When the module 4 or the module 5 stops due to a failure, the maintenance worker generally determines the stopped module based on the lighting color of the LED 4g and replaces the failed module.

  However, human beings make mistakes, and there is a possibility that a maintenance worker accidentally pulls out a module that is not a replacement target. For this reason, in Patent Document 1, when a printed circuit board (corresponding to a module) fails and is to be separated, the failure / detachment is displayed to notify the operator (maintenance worker, etc.) of replacement, and the failure If an operator touches the PCB mounting / removal bracket that is not faulty / separated so that the printed circuit board is not accidentally removed when the printed circuit board is unplugged while the power is on, this contact is detected and the error is detected. It is determined that the removal is about to be performed, and a warning signal is generated. The alarm device receives a warning signal issued from each printed circuit board, and warns the operator of contact with the printed circuit board not to be removed.

Further, in Patent Document 2, when an electronic device is operating, an electronic circuit unit is prevented in order to prevent an operator from accidentally pulling out an electronic circuit unit (corresponding to a module) connected to a back panel by a connector. As a structure to fix the unit to the frame (housing), a special lock screw with a washer is used and the fixing plate on the front of the module is fixed at a distance from the frame. By making the lever of the insertion / extraction monitoring switch in the module board movable, the insertion / extraction monitoring switch operates when this lock screw is loosened. An alarm is issued from an alarm generation circuit common to the circuit units.
Japanese Patent Laid-Open No. 06-295250 JP 09-331170 A

    However, the conventional technique has the following problems. That is, in the above conventional fault-tolerant server, when a maintenance worker replaces a failed module, the maintenance worker determines the replacement target module by checking the display state of the LED indicating the module status. It is a problem that there is a risk that the system is brought down by removing a module in operation due to a judgment error.

  Therefore, as described in Patent Document 1, when an attempt is made to replace a non-replaceable module that is not a failed / isolated module, and the module is touched, an alarm unit common to each module alerts Is ringing. In this case, firstly, since the module width is narrow and there is a conductor such as a frame, there is a problem that it is difficult to identify a contact without error for each module. Secondly, when a warning is issued with a common buzzer, the buzzer will sound at a distance, so it is difficult to determine what caused the buzzer, and it may not be possible to avoid misjudgment. There's a problem.

  As a measure for solving the first problem of Patent Document 1, a method of detecting looseness of a screw for fixing a fixing plate on the front surface of the module to the frame as described in Patent Document 2 can be considered. However, since this fixing screw is generally fixed to a tap or the like in one or both of the upper and lower frames of the frame in which the module is inserted, it is located outside the board constituting the module. It is difficult to detect looseness in the frame after inserting the. In Patent Document 2, a special fixing screw with a washer is used, and the fixing plate on the front of the module is fixed in a floating state with a gap from the frame. The lever can be moved and the looseness of the fixing screw is detected. As a general module fixing structure, a fixing plate on the front surface of the module is fixed in close contact with the frame. However, in the technique of Patent Document 2, a special fixing screw or fixing plate is spaced from the frame and floated. For this reason, there is a problem that the fixing structure of the module becomes special because the ejector or the like cannot be easily attached, and a general fixing structure of the module cannot be taken.

  The present invention has been made in view of the above-described circumstances, and can adopt a general fixing structure in which a fixing plate on the front surface of a module of a maintenance replacement unit is closely fixed to a housing. To provide a fault-tolerant device that can reliably notify the maintenance operator of an erroneous operation when the maintenance operator tries to pull out an operating module that is not a replacement target when It is aimed.

In order to solve the above-mentioned problems, the invention according to claim 1 relates to a fault-tolerant device, and relates to a fault-tolerant device in which modules of exchange units that are n-folded (n is a natural number of 2 or more) are connected, module, and the module substrate, and a transmitting member for transmitting the fixed plate that will be fixed by a fixing member so disposed on the front surface of the module substrate into close contact with the casing, the displacement of the fixing member on the module substrate, wherein the module substrate Detecting means mounted on the side for detecting looseness of the fixing member by displacement of the transmission member, display control means for displaying a status indicating whether the module is operating or stopped, and the detecting means at the time of the operating display provided in a notification control means for outputting a notification signal of the erroneous operation in the case where the detection was made, and informing means for informing an operation error by the broadcast signal In addition, the transmission member holds the head of the fixing member on the back side opposite to the surface of the fixing plate facing the module substrate, so that the fixing plate is fixed even when the fixing plate is stationary. A lever member that rotates in conjunction with tightening or loosening of the fixing member with a portion near the center of the back surface of the plate as a fulcrum, and a hole provided in the middle of the lever member and formed in the fixing plate. And a protrusion that transmits the displacement of the fixing member to the detection means .

The invention of claim 2 wherein relates to fault-tolerant device according to claim 1, wherein the detecting means, taking the fixing member by displacement in the direction away of the protrusion provided in the middle of the lever member It is characterized by detecting that it is going to be removed.

  The invention according to claim 3 relates to the fault tolerant device according to claim 1 or 2, characterized in that the module of the n-folded exchange unit is a doubled or tripled module.

  The invention according to claim 4 relates to the fault tolerant device according to claim 3, characterized in that the operation is any one of a triple operation, a dual operation, and a single system operation. .

  According to a fifth aspect of the present invention, in the fault tolerant device according to any one of the first to fourth aspects, the notification means notifies an erroneous operation by a voice message.

  As described above, according to the configuration of the present invention, as a first effect, when a maintenance worker tries to pull out an operating module that is not a replacement target and tries to remove the fixing member of the fixing plate, Since this is detected using the detecting means and is notified by the notifying means in the module, the maintenance worker can be surely notified of the erroneous operation, and the maintenance worker is erroneously operating. It is possible to prevent the risk of pulling out the module and causing the system to go down. As a second effect, the mounting state of the fixing member for tightly fixing the fixing plate on the front surface of the module to the housing can be sensed in the module substrate by a transmission member such as a lever member. It can be applied to a module having a fixed structure.

The present invention can adopt a general fixing structure in which the fixing plate on the front surface of the module of the maintenance replacement unit is closely fixed to the housing, and when this module is stopped due to a failure or the like, the maintenance work is performed. The purpose of providing a fault-tolerant device that can reliably notify a maintenance operator of an erroneous operation when an operator tries to pull out an operating module that is not a replacement target is to provide a buzzer for each module in the maintenance replacement unit. A buzzer control circuit that controls the sounding of this buzzer and a switch that detects the mounting state of the screws that are in close contact with the housing and fix the fixing plate on the front of the module within the module board using the lever plate are stopped. If the display lighting command is issued, even if the switch detects the looseness of the fixing screw, the buzzer control circuit determines that the operation is normal and the buzzer If the screw that fixes the active module is loosened without trying to pull out the module that is not to be replaced, the buzzer control circuit detects the looseness of this screw and the buzzer control circuit This was realized by sounding the module's buzzer by judging that it was an erroneous operation and notifying the maintenance operator of the erroneous operation.
Embodiments of the present invention will be described below with reference to the drawings. The description will be made specifically using examples.

  First, a first embodiment of the present invention will be described. FIG. 1 is a block diagram showing an electrical configuration of an information processing apparatus such as a server according to this embodiment. In FIG. 1, an information processing apparatus 1 such as a server includes a housing 2, a backplane 3, a module 4, and a module 5. The module 4 and the module 5 are composed of module boards 4a and 5a on which circuits are mounted, and fixed plates 4b and 5b are attached to the front surfaces thereof. The module 4 and the module 5 are inserted into the housing 2 from the left side in the figure, and are electrically connected to the backplane 3 with a connector or the like. The fixing plates 4b and 5b are screws 4c and 5c, and the module insertion port of the housing 2 It is fixed in close contact with the upper or lower edge. When removing the module 4 and the module 5, the screws 4c and 5c can be loosened and removed in the left direction in the figure.

  The module 4 and the module 5 are the same HW (hardware), and the duplex control circuits 4d and 5d, the LED control circuits 4e and 5e, and the buzzer control circuits 4f and 5f are mounted on the module boards 4a and 5a. . The LEDs 4g and 5g and the buzzers 4h and 5h are attached to the fixed plates 4b and 5b or mounted on the module substrates 4a and 5a so that they can be seen and heard from the fixed plates 4b and 5b. The switches 4i and 5i detect the removal of the screws 4c and 5c with a transmission member (insulator plates 4j and 5j) described later, but are schematically illustrated in the drawing. As the switches 4i and 5i, various detection means for detecting displacement such as a microswitch or a photosensor can be used.

  Details of the fixing structure for inserting and fixing the module to the housing will be described with reference to the side view of FIG. 2 and the front view of FIG. Here, the configuration of the module 4 will be described, but the configuration of the module 5 is also the same (the reference numeral 4 is replaced with 5). One end of a lever plate 4j that is foldable with resin or the like from the center of the front surface is fixed to the fixing plate 4b. As a result, the insulator plate 4j can rotate with the portion closer to the center of the fixed plate 4b as a fulcrum. At the other end of the insulator plate 4j, a screw 4c is positioned where the screw 4c is fixed to the lower edge of the module insertion slot of the housing 2, and a hole 4k through which the screw 4c is inserted is formed. The hole 4k is attached so that the head of the screw 4c does not fall off due to a washer or the like. A protrusion 4m is provided in the middle of the insulator plate 4j toward the module substrate 4a. The fixing plate 4b is provided with a hole 4n through which the protrusion 4m is inserted, and the switch 4i for detecting the displacement of the protrusion 4m when the insulator plate 4j is in close contact with the fixing plate 4b on the module board 4a side at that position. Place. Note that this fixing structure may be provided on the upper end side of the fixing plate 4j contrary to the illustrated example, or may be provided on both the upper end side and the lower end side. On one end side where the fixing structure is not provided (upper end side in the illustrated example), normal screw fixing may be performed or may be omitted depending on circumstances.

  Next, an operation example of this embodiment will be described. The duplex control circuits 4d and 5d mounted on the module boards 4a and 5a communicate with each other via the back plane 3 to control the duplex. There are mainly three statuses (operation states) of the module 4: “duplexing operation in progress”, “operating in one system”, and “stopping”, and the duplexing control circuit 4d is connected to the module 4 via the signal line 4p. The status information is transmitted to the LED control circuit 4e. The LED control circuit 4e determines the lighting color of the LED 4g according to the module status, and transmits an LED control signal, which is a lighting instruction signal, to the LED 4g and the buzzer control circuit 4f via the signal line 4q. Examples of LED lighting colors are green when "duplexing" is active, amber (dark blue) when "single system is operating", and red when "stopping". The LED 4g lights up a color representing a status designated by the LED control signal from the LED control circuit 4e.

  The switch 4i senses the attachment state of the screw 4c to the housing 2 by the displacement of the protrusion 4m provided on the insulator plate 4j, and transmits a switch control signal to the buzzer control circuit 4f via the signal line 4r. The buzzer control circuit 4f determines the module status based on the LED control signal, and simultaneously issues a buzzer sounding signal command to the buzzer 4h via the signal line 4s based on the switch control signal. Redundant control circuit 5d, LED control circuit 5e, LED 5g, buzzer 5h, buzzer control circuit 5f, switch 5i, screw 5c, insulator plate 5j, protrusion 5m, signal line 5p, signal line 5q, signal line 5r, signal line 5s Each has the same function as each component of the code | symbol corresponding to the module 4 mentioned above.

  Next, a case where a failure occurs in the module 5 and the module 5 is replaced will be described with reference to FIGS. 4 and 5. When a failure occurs in the module 5 and the module 5 is logically disconnected from the apparatus, the LED 5g is lit red (indication of stoppage). At this time, since the module 4 is operating in one system, the LED 4g is amber-lit. Here, when the screw 5c is loosened to replace the module 5, the switch 5i senses it and transmits a switch control signal to the buzzer control circuit 5f via the signal line 5r. However, since the LED 5g is in a red lighting state (stopped display), the buzzer control circuit 5f does not issue a buzzer sounding instruction to the buzzer 5h, so that the module 5 can be removed and replaced.

  Here, when the maintenance worker unintentionally loosens the screw 4c in an attempt to replace the module 4 that is operating (operating in one system), the protrusion 4m provided on the insulator plate 4j causes the screw 4c to loosen. Accordingly, the switch 4i senses this displacement because it moves away from the module substrate 4a and displaces, and the switch 4i transmits a switch control signal to the buzzer control circuit 4f via the signal line 4r. Since the LED 4g is in the amber lighting state (display indicating that one system is operating), the buzzer control circuit 4f issues a buzzer sounding instruction to the buzzer 4h in the module 4, and the buzzer 4h sounds. For this reason, the maintenance worker is surely aware of an error trying to replace a module that is not a replacement target. Therefore, when replacing the failed module, it is possible to reliably prevent the danger that the maintenance operator will accidentally pull out the operating module and bring down the system.

  Next explained is the second embodiment of the invention. FIG. 6 is a block diagram showing an electrical configuration of an information processing apparatus such as a server according to this embodiment. In this embodiment, the present invention is applied to an information processing apparatus having a triple module configuration. This embodiment is different from the first embodiment in that three modules perform a triple operation. Others are almost the same as those in the first embodiment, and can be easily analogized. Therefore, the differences will be mainly described, and the description of the same parts as those in the first embodiment will be omitted.

  In FIG. 4, the information processing apparatus 1 includes a module 6 having the same HW (hardware) configuration as the modules 4 and 5, in addition to the housing 2, the backplane 3, the module 4, and the module 5. Similar to the first embodiment, these modules 4, 5, and 6 are inserted into the housing 2 from the left side in the figure, and the rear portions of the module boards 4a, 5a, and 6a are electrically connected to the backplane 3 by connectors or the like. Then, the fixing plates 4b, 5b, 6b on the front face of the modules 4, 5, 6 are fixed to the upper edge or the lower edge of the module insertion opening of the housing 2 by screws 4c, 5c, 6c. When removing the modules 4, 5 and 6, they can be removed by loosening the screws 4c, 5c and 6c and pulling them out in the left direction in the figure.

  The modules 4, 5 and 6 of this embodiment are different from the modules 4 and 5 of the first embodiment in that the triple control circuits 4t, 5t and 6t are mounted on the module substrates 4a, 5a and 6a instead of the double control circuit. It is a point that has been. The triple control circuits 4t, 5t, and 6t communicate with each other via the backplane 3 to control the triple, and the status of each module is mainly “tripling in operation” and “duplexing in operation”. The four types of status information “operating in one system” and “stopped” are notified to the LED control circuits 4e, 5e, 6e of the respective modules via the signal lines 4p, 5p, 6p.

  The LED control circuits 4e, 5e, and 6e determine the lighting colors of the LEDs 4g, 5g, and 6g according to the module status, and the signal lines 4q, 5q, and 6q are connected to the LEDs 4g, 5g, and 6g and the buzzer control circuits 4f, 5f, and 6f. Then, an LED control signal that is a lighting instruction signal is transmitted. Examples of LED lighting colors are green when “triple operation” and “duplex operation”, amber (orange) when “single system operation”, and red when “stop” Three colors can be turned on, and the LEDs 4g, 5g, and 6g are turned on to indicate the statuses designated by the LED control signals from the LED control circuits 4e, 5e, and 6e.

  The fixing structure in which the module is inserted and fixed to the casing is a general fixing structure in which the fixing plate is fixed in close contact with the casing as in the first embodiment. The switches 4i, 5i, 6i sense the attachment state of the screws 4c, 5c, 6c to the housing 2 by the displacement of the protrusions 4m, 5m, 6m provided on the insulator plates 4j, 5j, 6j, and signal lines 4r, 5r. , 6r, the switch control signal is transmitted to the buzzer control circuits 4f, 5f, 6f. When the buzzer control circuits 4f, 5f, and 6f determine that the module status is “stopped” based on the LED control signal, the buzzer sounding signal is not issued even when the switch control signal is received. If it is determined that the module status is in operation other than “stopped” (“triple operating”, “redundant”, “operating in one system”), the signal line 4s is switched by the switch control signal. , 5s, 6s, a buzzer sounding signal command is issued to the buzzers 4h, 5h, 6h.

  In this embodiment, as in the case of the information processing apparatus of the first embodiment in which the module is duplicated, a general fixing structure in which the fixing plate is fixed in close contact with the housing can be taken, and the status is “stopped” Only when the screw is loosened to remove the module, do not sound the buzzer, and remove the module that is in operation ("triple operation", "duplex operation", "single system") When the screw is loosened, the buzzer in the module is sounded, so that it is possible to reliably prevent the operating module from being removed and the system going down when replacing the failed module.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and the present invention can be changed even if there is a design change or the like without departing from the gist of the present invention. include. For example, in the above-described embodiment, an example in which the present invention is applied to an information processing apparatus such as a server has been described.

In each embodiment, the case where there is one module of each system is illustrated, but a plurality of modules of each system may be configured. In each embodiment, an ejector that facilitates removal of the module is omitted. However, since the present invention is applied to a general fixing structure, the ejector can be easily provided. Of course, it does not matter. Also, instead of fixing screws, for example, a fixing member made of resin or the like that, when inserted through a hole on the housing side, becomes large in diameter at the back and is fixed unless it is pulled out by applying force It may be. Further, in each embodiment, a buzzer is provided in the module as a notification means to notify an erroneous operation, but if a notification is given by a voice message such as "I am trying to remove a module in operation", It is possible to reliably prevent erroneous operations.
In the above-described embodiment, the case of the duplex module and the triple module has been described. However, the present invention is not limited to these, and may be quadruple or more as necessary. .

  The present invention can be suitably used for various control devices that require reliability in addition to information processing devices such as servers.

It is a block diagram which shows the electrical constitution of the information processing apparatus comprised with the duplication module which is 1st Example of this invention. It is a side view which shows the example of fixing structure of the module. It is a front view which shows the example of fixing structure of the module. It is a figure explaining the example of operation at the time of exchange operation of the module. It is a figure which shows the operation example of the module. It is a block diagram which shows the electric constitution of the information processing apparatus comprised with the triple module which is 2nd Example of this invention. It is a block diagram explaining the electrical constitution of the fault tolerant server comprised by the duplication module which is a prior art example.

Explanation of symbols

1 Information processing equipment (fault tolerant equipment)
2 Housing 4, 5, 6 Module 4a, 5a, 6a Module substrate 4b, 5b, 6b Fixing plate 4c, 5c, 6c Screw (fixing member)
4e, 5e, 6e LED control circuit (display control means)
4f, 5f, 6f buzzer control circuit (notification control means)
4h, 5h, 6h Buzzer (notification means)
4i, 5i, 6i switch (detection means)
4j, 5j, 6j Insulator plate (transmission member, insulator member)
4m, 5m, 6m protrusion

Claims (5)

A fault-tolerant device in which modules of exchange units that are n-folded (n is a natural number of 2 or more) are connected,
Each module, and the module substrate, and a transmitting member for transmitting the fixed plate that will be fixed by a fixing member so disposed on the front surface of the module substrate into close contact with the casing, the displacement of the fixing member on the module substrate, wherein the module A detection means mounted on the substrate side for detecting looseness of the fixing member by displacement of the transmission member; a display control means for displaying a status indicating whether the module is operating or stopped; and the detection at the time of display during operation. and notification control means for outputting a notification signal of the erroneous operation in the case where the detection was made by means and comprising a notifying means for notifying the erroneous operation by the notification signal,
The transmission member is
By holding the head of the fixing member on the back side opposite to the surface facing the module substrate of the fixing plate, even when the fixing plate is stationary, the portion near the center of the back surface of the fixing plate As a fulcrum, a lever member that rotates in conjunction with the tightening or loosening of the fixing member, and a displacement of the fixing member that is provided in the middle of the lever member and is formed in the fixing plate. A fault-tolerant device comprising a projection for transmitting to the detection means . The detection means, according to claim 1, characterized in that to detect that an attempt to remove the stationary member by displacement in the direction away of the protrusion provided in the middle of the lever member Fault tolerant equipment.   3. The fault tolerant device according to claim 1, wherein the module of the n-folded exchange unit is a duplexed or tripled module.   4. The fault tolerant device according to claim 3, wherein the operation is any one of a triple operation, a dual operation, and a single system operation.   The fault-tolerant device according to claim 1, wherein the notification unit notifies an erroneous operation by a voice message.

Images