JPH1139250A - Peripheral device connection confirmation method for cluster system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To early detect a connection mistake and to shorten the time needed for a site adjustment job by integrating the information written in an inherent information storage part and producing a connection information table of peripheral devices as a cluster system. SOLUTION: The file devices 311, 312, 322, 331 and 332 which are controlled by the file controllers 31 to 33 via the LC (control channels) have the inherent information storage parts 411, 412, 421, 422, 431 and 432 which store the inherent information consisting of the node numbers and the channel numbers consisting of the physical and logical channels numbers. The information which are written in the storage parts 411 to 432 are integrated and a connection information table of peripheral devices is produced as a cluster system. Thus, the connection states of peripheral devices are displayed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for confirming connection of a connection cable between a node and a peripheral device in a cluster system, and more particularly to a method for confirming connection when the number of connection cables is increased, which is suitable for a large-scale system. .

[0002]

2. Description of the Related Art As a method of creating I / O configuration information in a computer system, for example, Japanese Patent Application Laid-Open No. 3-269754
In the publication, device-specific information (I /
O based on a database of control information, interface information, etc. for O processing, and connection information of peripheral devices used in the system of the specific user, the connection information of each peripheral device used in the system of the specific user, and the device. A method of automatically creating a configuration information file including unique information has been proposed. This method realizes the automation of the setting information of the peripheral devices regardless of how the peripheral devices are actually connected.

[0003] Also, Japanese Patent Application Laid-Open No. 1-187663 discloses that
Connection means for connecting a peripheral device and a peripheral device adapter and having a dedicated signal line for transmitting a type identification signal in a cable having at least the peripheral device as a connector, and a type identification signal of the peripheral device provided in the peripheral device And a type identification signal generating means for outputting a signal to a dedicated terminal of the connector of the connection means, and a type identification signal provided from the connection means provided in the peripheral device adapter to identify the type of the connected peripheral device and to the higher-level device. There has been proposed a device type recognition system having type identification means for transmitting the type of the peripheral device. In this method, the cable connecting the peripheral device and the peripheral control device has a dedicated signal line for identifying the type of the peripheral device, so that which type of peripheral device is connected is determined by the higher rank. This is a method that the device recognizes, and detects only the type of the connected peripheral device.

[0004]

In the above-mentioned conventional system, after the connection work between a plurality of higher-level devices and the peripheral devices has been completed, it is not known how the connections were actually made. There is a problem that man-hours increase.

The reason is as follows. In recent years, with the progress of clustering technology, the scale of a cluster system has increased, and the number of shared peripheral devices has also increased accordingly. In addition, the number of cables for connecting them is inevitably increased, and it is very difficult to visually confirm the connection.

In the above-mentioned conventional method, the actual connection relationship between a specific node and a specific peripheral device is not known. Therefore, if a cable connection error occurs, it is not possible to point out a specific location. Work time for adjustment increases.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to detect a connection error of a cable between a node and a peripheral device at an early stage, thereby shortening the time required for on-site adjustment. It is an object of the present invention to provide a method for confirming connection of peripheral devices of a cluster system.

[0008]

In order to achieve the above object, a peripheral device connection confirmation system for a cluster system according to the present invention comprises a plurality of central processing units (referred to as "nodes") connected to the respective nodes. A plurality of peripheral devices,
A cluster system comprising a unique information storage unit in which each of the peripheral devices writes information unique to each of the peripheral devices in the system from each of the nodes, and integrates the information written in the unique information storage unit. By creating a connection information table of the peripheral devices as the above, the connection status of the peripheral devices is displayed.

[0009]

Embodiments of the present invention will be described below. In a preferred embodiment of the present invention, a plurality of nodes (21 and 22 in FIG. 1) and a plurality of peripheral devices (311 to 31 in FIG. 1) connected to each node are provided.
332), a unique information storage unit (41 in FIG. 1) capable of writing information unique to each peripheral device in each system from each node to each peripheral device.
1 to 432), the information written in the unique information storage unit is integrated, and a connection information table of peripheral devices as a cluster system is created.

More specifically, unique information clearing means (5 in FIG. 2) for erasing the contents of the unique information storage unit in each node.
1), a unique information reading means (52 in FIG. 2) for reading and checking the content of the unique information storage unit, and a unique information writing means (FIG. 2) for writing the unique information when the content of the unique information storage unit is cleared. 2) 53) and an in-node unique information table creating means (54 in FIG. 2) for creating an in-node unique information table including unique information and a channel number of a peripheral device corresponding to the unique information. ,
A unique information clear instructing means (61 in FIG. 3) for sequentially issuing an instruction to clear unique information to all nodes in the system, and a unique information in the node for sequentially creating and transmitting a unique information table to all nodes Information table creation instructing means (62 in FIG. 3) and system unique information table creating means (63 in FIG. 3) for creating and displaying a system unique information table based on the in-node unique information tables collected from all nodes. Have.

In an embodiment of the present invention, in a cluster system in which a plurality of nodes are connected while sharing a plurality of peripheral devices, all nodes have cleared unique information in all connected peripheral devices. Thereafter, the nodes in the system perform the following processing sequentially from the weakest node.

Each node reads the contents of the unique information storage section existing in the peripheral device in order from the peripheral device of the weakest channel with respect to the peripheral devices of all channels connected to the own node, If the content is empty, the unique information including the own node number and the channel number is written in the unique information storage unit, and if the content already exists, it is not written.

Further, a unique information table in the node is created using the unique information in the peripheral device and the channel number corresponding to the peripheral device as one piece of row information.

When the above processing is completed in all the nodes, the node number and the channel number corresponding to one piece of unique information are set to one based on the in-node unique information table created by each node and each node number. After creating a system-specific information table as line information, it is displayed on the console and notified to the operator.

[0015]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of the present invention;

FIG. 1 is a diagram for explaining a peripheral device connection confirmation method according to an embodiment of the present invention, and is a block diagram showing a configuration of a cluster system. In the following, a case will be described in which the cluster system includes two nodes, but the present invention is not limited to such a configuration.

Referring to FIG. 1, in this embodiment, the cluster system includes nodes 21 and 2 serving as a central processing unit.
2, an integrated service processor (hereinafter, referred to as an “integrated SVP”) 1 that controls all the nodes 21 and 22, a console terminal 11 connected to the integrated SVP 1, and a physical channel (hereinafter, referred to as an “integrated SVP”). , "P
C) connected via the file control device 3).
1, 32, and 33, and the file devices 311, 312, 321, and 3 controlled by the file control devices 31 to 33 via logical channels (hereinafter, referred to as “LC”), respectively.
22, 331, and 332.

As shown in FIG. 5, each of the file devices 311 to 332 stores unique information composed of a node number and a channel number including a physical channel number and a logical channel number. Storage unit 41
1, 412, 421, 422, 431 and 432 are provided. FIG. 5 is a diagram showing an example of a format of unique information stored in a peripheral device according to an embodiment of the present invention.

FIG. 2 shows a node 21 according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating a configuration of a second embodiment. Referring to FIG. 2, nodes 21 and 22 include a unique information clear unit 51 that erases the contents of unique information storage units 411 to 432, and a unique information storage unit 41.
Specific information reading unit 52 that reads the contents of items 1 to 432
And a unique information writing unit 53 that writes unique information into the unique information storage units 411 to 432, and an in-node unique information that creates an in-node unique information table including unique information and a channel number corresponding to the unique information illustrated in FIG. And a table creation unit 54.

FIG. 3 shows the overall S in one embodiment of the present invention.
FIG. 3 is a diagram illustrating a configuration of a VP. Referring to FIG.
1 is a unique information storage unit 411-4 stored in each of the nodes 21 and 22.
A unique information clear instructing unit 61 for giving an instruction to delete the contents of the node 32; an in-node unique information table creation instructing unit 62 for giving an instruction to each node 21 and 22 to create an in-node unique information table; A system unique information table creating unit 63 for creating and displaying a system unique information table based on the in-node unique information tables created by the nodes 21 and 22.

FIG. 4 is a flow chart for explaining the operation of one embodiment of the present invention. A file device connection confirmation operation according to the present embodiment will be described with reference to FIGS.

Specific information clear instructing section 61 of integrated SVP 1
Issues an instruction to clear the unique information to the weakest node 21 of the cluster system (steps 1001 and 100).
2).

When the node 21 receives the instruction to clear the unique information, the unique information clear unit 51 of the node 21
The I / O for clearing the unique information storage unit 411 is executed for the file device 311 connected to PC = 1, LC = 1, which is the weakest channel number in No. 1 (steps 2001 and 2002). Continue with the unique information clearing section 5
1 is obtained by adding one channel number and PC = 1, LC = 2
The I / O for clearing the unique information storage unit 412 is executed for the file device 312 connected to the server (steps 2003, 2004, 2002). Similarly, the unique information clearing unit 51 sets the maximum channel number PC = 2,
After executing I / O for clearing the unique information 432 on the file device 332 connected to LC = 2,
Integrate notification of end of unique information clear processing as a node S
Send to VP1 (steps 2002 and 2003).

The unique information clear instructing unit 61 of the integrated SVP 1
Receives the notification of the end of the unique information clearing process from the node 21, and issues a unique information clear instruction to the node 22 having the next lowest node number after the node 21 (step 10).
02, 1003, 1004).

When receiving the instruction to clear the unique information, the node 22 performs the same processing as that performed by the unique information clear unit 51 of the node 21 and notifies the integrated SVP 1 of the end of the unique information clear processing as the node. Is sent (steps 2001 to 2004).

The specific information clear instructing section 61 of the integrated SVP 1
When receiving the notification of the end of the unique information clearing process from the node 22, since the node number of the node 22 is the maximum node number of the cluster system, the process proceeds to the next process (steps 1002 and 1003). At this point, the unique information storage units 411 to 432 of all the file devices 311 to 332 existing in the cluster system have been cleared.

Next, the in-node unique information table creation instructing unit 62 of the integrated SVP 1 issues an instruction to create the in-node unique information table to the weakest node 21 of the cluster system (steps 1005 and 1006).

When the node 21 receives the instruction to create the intra-node unique information table, the unique information reading unit 52 of the node 21 sends the PC which is the weakest channel number of the node 21 to the PC.
= 1, I / O for reading the unique information storage unit 411 is executed for the file device 311 connected to LC = 1 (steps 2010 and 2011). When the unique information read by this I / O is referred to, the content is cleared, so that the process is passed to the unique information writing unit 53 of the node 21 (step 2012).

The unique information writing unit 53
The unique information is created on the basis of the node number = 1 and the channel numbers PC = 1 and LC = 1 executed by the unique information reading unit 52. Next, I / O for writing the created unique information to the unique information storage unit 411 of the file device 311 corresponding to the channel number executed by the unique information reading unit 52 is executed, and the created unique information is stored in the node 2.
1 (step 2013).

The in-node unique information table creation unit 54
The unique information received from the unique information writing unit 53 and the channel number P
As shown in FIG. 6, C = 1 and LC = 1 are registered in the in-node unique information table as one piece of row information (step 2014). FIG. 6 is a diagram illustrating an example of a format of the intra-node unique information table created by the intra-node unique information table creation unit of the node 21 according to the embodiment of the present invention.

Subsequently, the unique information reading unit 52 adds the channel number by one, and sends an I / O for reading the unique information storage unit 412 to the file device 312 connected to PC = 1 and LC = 2. Execute (Step 201)
5, 2016, 2011). Hereinafter, similar to the processing performed by the file device 311 described above, the processing is performed up to the file device 332 connected to the maximum channel number, that is, the channel number PC = 2 and LC = 2 (step 2011 to 2011).
2016).

The unique information reading unit 52 is provided for the own node 21.
When the processing up to the maximum channel number PC = 2 and LC = 2 is completed, the in-node unique information table is sent to the integrated SVP1, and the processing as the node is completed (step 201).
5, 2017).

The in-node unique information table creation instructing unit 62 of the integrated SVP 1 holds the in-node unique information table sent from the node 21 and the node number = 1 of the source node 21, and An instruction to create an in-node unique information table is issued to the node 22 having the young node number (steps 1007, 1008, 1009,
1006).

When the node 22 receives the instruction to create the intra-node unique information table, the same processing as that performed by the above-described unique information reading unit 52, unique information writing unit 53, and intra-node unique information table creating unit 54 of the node 21 is performed. Processing of the file devices 321 and 32 under the file control device 32
2 (steps 2010 to 2016). Note that the channel numbers set at the end of this process are PC = 1 and LC = 2.

Next, the unique information reading unit 52 adds one channel number to the file device 331 connected to PC = 2 and LC = 1 and stores the unique information in the unique information storage unit 431.
Is executed (steps 2015 and 20).
16, 2011). By referring to the unique information that was read,
Since the unique information at the time of processing by the node 21 is already stored, the intra-node unique information table creation unit 54
The read unique information and the unique information
The channel number PC = 2 and LC = 1 for which O was executed is taken as one piece of line information. As shown in FIG. 7, the information is registered in the unique value information table in the node (steps 2012 and 201).
4). FIG. 7 is a diagram showing a format of the intra-node unique information table created by the intra-node unique information table creation unit of the node 22 in one embodiment of the present invention.

Further, the unique information reading unit 52 adds one channel number to the file device 332 connected to PC = 2 and LC = 2, and
The same processing as the processing performed in is performed (step 2015,
2016, 2011, 2012, 2014).

The unique information reading unit 52 is a
When the processing up to the maximum channel number PC = 2 and LC = 2 is completed, the intra-node specific information table is sent to the integrated SVP1, and the processing as a node is completed. (Step 201
5, 2017).

The in-node unique information table creation instructing unit 62 of the integrated SVP 1 holds the in-node unique information table sent from the node 22 and the node number = 2 of the source node 22. Since the node number of the node 22 is the maximum node number of the cluster system, the process is taken over to the system unique information table creation unit 63 (step 10).
07, 1008).

The system-specific information table creation unit 63 of the integrated SVP 1 of the integrated SVP 1 first sets a node number to the node 21 which is the weakest node (step 101).
0).

Next, the line number of the intra-node unique information table is set to 1 (step 1011). The unique information in the first row of the unique information table within the node and the unique information area of the unique information table in the system are searched for the same unique information. In this case, since the same unique information does not exist,
The unique information, the node number of the node 21 = 1, and the channel number corresponding to the unique information in the intra-node unique information table are registered as one piece of line information in the system unique information table as shown in FIG. (Step 1012,
1013). FIG. 8 shows the integrated S in one embodiment of the present invention.
FIG. 4 is a diagram illustrating an example of a format of a system unique information table created in a VP.

The second and subsequent rows of the intra-node specific information table are processed up to the last row in the same manner as the first row, and the processing of the intra-node specific information table sent from the node 21 is completed (steps 1015, 1016, 1012). , 101
3).

The node number is set to the next node 22 after the node 21.
Is set to node = 2 (step 1018).

Next, the line number of the intra-node unique information table is set to 1 (step 1011). The unique information in the first row of the unique information table in the node and the unique information area of the unique information table in the system are searched for the same unique information. Since the same unique information does not exist, the unique information and the node number of the node 22 are found. = 2 and a channel number corresponding to the unique information in the unique information table within the node,
It is registered as one piece of line information in the system unique information table (steps 1012 and 1013).

The second line of the in-node specific information table performs the same processing as the first line (steps 1015, 1016,
1012, 1013). When searching for the same unique information in the unique information area of the third line of the unique information table in the node and the unique information area of the unique information table in the system,
Since the same unique information exists, the node number = 2 of the node 22 and the channel number corresponding to the unique information in the unique information table in the node are added to the line of the unique information already in the unique information table within the system ( Step 101
2, 1014).

The fourth line of the in-node unique information table performs the same processing as the third line (steps 1015, 1016,
1012, 1014). Since the fourth row is the last row, the processing of the in-node unique information table sent from the node 22 ends (step 1015).

As described above, the processing of the in-node unique information table up to the maximum node number of the system has been completed.
(Steps 1017 and 1019).

The unique information in the system unique information table indicates a specific file device, and the node number and the channel number corresponding to the unique information indicate the node connected to the file device and the channel number of the node. Therefore, by comparing the system specific information table displayed on the console 11 with a system configuration diagram prepared in advance, it is possible to know whether there is a connection error and what kind of connection error there is. For this reason, the on-site adjustment work time is reduced.

[0048]

As described above, according to the present invention,
Even if the number of cables to be connected increases due to the increase in the number of nodes and peripheral devices, it is possible to detect cable connection mistakes at the time of on-site adjustment at an early stage, thereby reducing the time required for on-site adjustment. Play.

The reason is that, in the present invention, unique unique information is written to all peripheral devices existing in the system, and a list of connection information corresponding to the unique information is displayed, thereby preparing in advance. This is because a connection error can be reliably found and corrected by comparing it with the existing system configuration diagram.

[Brief description of the drawings]

FIG. 1 is a block diagram for explaining a peripheral device connection confirmation method according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a node according to an embodiment of the present invention.

FIG. 3 is a block diagram illustrating a configuration of an integrated SVP according to an embodiment of the present invention.

FIG. 4 is a flowchart for explaining the operation of one embodiment of the present invention.

FIG. 5 is a diagram showing an example of a format of unique information stored in a peripheral device according to an embodiment of the present invention.

FIG. 6 is a diagram illustrating an example of a format of an intra-node unique information table created by an intra-node unique information table creation unit of the node 21 according to an embodiment of the present invention.

FIG. 7 is a diagram showing a format of an intra-node unique information table created by an intra-node unique information table creation unit of the node 22 according to an embodiment of the present invention.

FIG. 8 is a diagram showing an example of a format of a system unique information table created in an integrated SVP according to an embodiment of the present invention.

[Explanation of symbols]

1 Integrated Service Processor (Integrated SVP) 11 Console 21, 22 Central Processing Unit (Node) 31, 32, 33 File Control Device 311, 312, 321, 322, 331, 332 File Device 411, 412, 421, 422, 431, 432 Unique information storage unit 51 Unique information clear unit 52 Unique information reading unit 53 Unique information writing unit 54 Intra-node unique information table creation unit 61 Unique information clear instructing unit 62 Intra-node unique information table creation instructing unit 63 System unique information table creating unit

Claims (4)

[Claims] 1. A plurality of central processing units (referred to as "nodes").
And a plurality of peripheral devices respectively connected to each of the nodes, wherein each of the peripheral devices writes information unique to each of the peripheral devices in the system from each of the nodes. Displaying the connection status of the peripheral devices by integrating the information written in the unique information storage unit and creating a connection information table of the peripheral devices as a cluster system;
A method for confirming connection of peripheral devices of a cluster system. 2. A plurality of central processing units (referred to as "nodes").
Is connected in a form sharing a plurality of peripheral devices, wherein each of the peripheral devices includes a unique information storage unit that stores information unique to each peripheral device from each of the nodes, wherein each of the nodes includes: After clearing the unique information storage units in all the connected peripheral devices, the peripheral devices of all the channels connected to the own node are sequentially turned on for each node in the system. Read the contents of the unique information storage section that exists in the
If the content is empty, the unique information including the own node number and the channel number is written in the unique information storage unit, and the unique information in the peripheral device, the channel number corresponding to the peripheral device, and one entry. A unique information table within a node is created as information, and a node number and a channel number corresponding to one piece of unique information are entered into one entry based on the unique information table within a node created by each node and each node number. Create a system-specific information table to be displayed on the display terminal,
A method for confirming connection of peripheral devices of a cluster system. 3. The unique information clear means for erasing the contents of the unique information storage unit of the peripheral device, the unique information reading means for reading and checking the contents of the unique information storage unit, and the unique information Unique information writing means for writing unique information when the contents of the storage unit are cleared, and in-node unique information for creating an in-node unique information table comprising unique information and a channel number of a peripheral device corresponding to the unique information 3. The system for confirming connection of peripheral devices of a cluster system according to claim 1, further comprising: table creation means. 4. A unique information clear instructing means for issuing an instruction to sequentially clear unique information to all nodes in the system, and a unique information table for sequentially creating and transmitting a unique information table to all the nodes. 2. A unique information table creation instructing means, and a unique system information table creating means for creating and displaying a unique system information table based on the in-node unique information tables collected from all the nodes. Or the peripheral device connection confirmation method of the cluster system according to 2.