JP3270400B2 - Print processing cluster system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a print processing cluster system, and more particularly to a print processing cluster system that can take over processing without the intervention of a manager when a failure occurs in the cluster system.

[0002]

2. Description of the Related Art A cluster system is a combination of a plurality of computers, and when a computer becomes unable to continue processing due to some kind of failure, another computer takes over the interrupted processing instead. .

One example of such a cluster system is disclosed in Japanese Patent Application Laid-Open No. 9-34852 ("Cluster System"), as shown in FIG.

Referring to FIG. 6, a conventional cluster system includes a node 10-1 and a node 10 (not shown).
−2, 10−3, etc., and these nodes are LAN
They are interconnected by 220-1 and 20-2. Then, in response to the detection of the failure, each node performs a dedicated service processor 4 for performing a lifetime recovery process and a reconfiguration process.
4 is provided. Then, the service processor 44
When a failure occurs in its own computer and it is determined that continuation of processing is impossible, the service processor is notified of the fact. And the service processor:
A start-up process, a configuration control process, a lifetime process, a remote diagnosis support, and the like can be performed for each computer, and another computer can take over the process based on a failure occurrence signal notified when a lifetime occurs.

[0005]

However, in the above-mentioned conventional cluster system, a manager system for recovery from a failure or the like is not required. However, if a failure occurs in a service processor, other nodes are notified of the failure and printed. Cannot receive instructions to take over processing.

Accordingly, an object of the present invention is to take over print data of a failed node in a print processing cluster system without requiring a manager system or a service processor.

[0007]

A system according to the present invention for solving the above-mentioned problems includes a node switching control unit, a print takeover processing unit that takes over a printing process based on an instruction from the node switching control unit, A plurality of nodes each including a spool control / print output unit that outputs print data based on an instruction from the takeover processing unit; and each of the nodes includes a disk device having a print spool area that stores print data. A print processing cluster system in which the node and a plurality of printers are connected to a common bus line, wherein a print processing failure in one of the nodes is detected by another node via a bus line, The disk device connected to the other node is connected to another node, and the printing process of one node is taken over by the other node. It is as to cause. In addition, the present invention
A node switching control unit, and the node switching control
Print takeover processing that takes over print processing based on the instructions of the department
And print data based on an instruction from the print takeover processing section.
With a spool control / print output unit for outputting data
Each node has print data.
A disk device having a print spool area for storing
A common bus line for the node and a plurality of printers
A cluster system connected to the
The unit stores the print data file in the spool control /
Once stored in the print spool area via the print output unit
Means for storing the data and the printer device being usable.
Wait, the print data file is sent to the spool control
Means for outputting to the printer device via a print output unit
And the printer finishes printing the print data.
After that, the print information file is stored in the print spool area.
From the print data file.
Is stored in the print spool area,
If a failure occurs in one of the nodes,
The takeover processing unit of another node detects the failure
And refer to the print spool management table and connect
One of the connection points,
The print data information file of the node
And wait for the printer device to become available
Storing the print data file in the spool control /
Output to the printer device via a print output unit,
After the printer finishes printing the print data,
Delete a print information file from the print spool area
Like that. Further, the present invention provides a node switching system.
Control unit and a mark based on the instruction of the node switching control unit.
A print takeover processing unit that takes over the printing process;
Spool that outputs print data based on instructions from the processing unit
A plurality of nodes each having a control / print output unit,
Is a print spool area for storing print data.
A disk device having an area, wherein the node and a plurality of
A printing process that connects a printer to a common bus line.
A raster system, wherein the cluster system
A node switching control unit is further provided on the bus line.
The print process is taken over based on the instruction of the mode switching control unit.
The print takeover processing unit and the instruction of the print takeover processing unit.
Spool control / print output to output print data based on
And the takeover processing unit includes a print data file.
Files via the spool control / print output unit.
Means for storing in the print spool area;
Waiting for the printer to become available,
The file is forwarded via the spool control / print output unit.
Means for outputting to the printer device;
After the printing of the print data is completed, the print information file
Means for deleting a file from the print spool area.
The print data file is stored in the print spool area.
In one of the nodes,
In the event of a failure, the takeover of the standby node
The failure processing unit detects the failure, and detects the fault where the failure occurred.
Handover of the print data information file of the
Wait until the printer device becomes available and
Print data file to the spool control / print output unit.
Output to the printer device via
After the printing of the print data is completed, the print information file
File from the print spool area
I have.

Another system according to the present invention is a cluster system including a node and a standby node, wherein the standby node does not have a disk device having a print spool area, and a failure occurs in the node. , The print processing is taken over by using the print spool area of the failed node as it is.

Further, according to a method of the present invention for solving the above-mentioned problems, there is provided a node switching control unit, a print takeover processing unit that takes over a printing process based on an instruction from the node switching control unit, A plurality of nodes each including a spool control / print output unit that outputs print data based on an instruction; each of the nodes includes a disk device having a print spool area for storing print data; A plurality of printer devices are connected to a common bus line, and the disk device stores a print spool management table storing a node name and a connection point at which another node can be connected to a disk device included in the node having the node name. a printing processing method using a cluster system, a single print data file including the print data of the node Type on the printing requesting unit, the print data file via the spool control / print output unit, temporarily stored in the printing spool, waiting for the printer device can be used, the print data Outputting a file from the print spool area to the printer via the spool control / print output unit, and deleting the print information file from the print spool area after the printer has finished printing the print data; Is stored in the print spool area, if a failure occurs in the one node,
The takeover processing unit of the other node detects the failure, refers to the print spool management table, finds the connectable connection point, and retrieves the print data information file of one of the failed nodes. The print data file is output to the printer device via the spool control / print output unit after taking over to the own node and waiting for the printer device to become usable, and the printer device prints the print data. Is completed, the print information file is deleted from the print spool area.

Another print processing method according to the present invention is a cluster system including a node and a standby node, wherein the standby node uses a cluster system that does not include a disk device having a print spool area. When a failure occurs, the print processing is taken over by using the print spool area of the failed node as it is.

[0011]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a print processing cluster system according to the present invention. In FIG. 1, only two nodes are shown for simplicity. As shown in FIG. 1, the print processing cluster system according to the present invention includes nodes 100 and 200.
And the shared disk devices 150 and 250 shared by them.
And printer devices 300 and 301. Then, the nodes 100 and 200 and the printer devices 300 and 3
01 is connected to the same bus line.

More specifically, a shared disk 150 is connected to the node 100, and a shared disk 250 is connected to the node 200.

The node 100 has a node switching control section 101, a print processing takeover processing section 102, a print request section 103, and a spool control / print output section 104. Further, the shared disk 150 has a spool area 151 for holding print data requested by the print request unit 103.

Similarly, the node 200 has a node switching control section 201, a print processing takeover processing section 202, a print request section 203, and a spool control / print output section 204, and the shared disk 250 has a spool area 251.

The node switching control units 101 and 201
The failure of the node is detected and it is determined whether or not to take over at the own node. When taking over, the shared disks 150 and 250 are connected to the own node and the print processing takeover processing unit is started.

In the node 100, the node switching control unit 101 detects a failure of the node 200 and connects the shared disk 250 of the failed node 200 to the own node 100. Also, the print takeover processing unit 102 for taking over the print process being executed by the failed node at the own node 100 is started.

The print processing takeover processing unit 102 determines from the means for referring to the spool management table 500 and the shared disk connection point 502-x of the spool management table 500 that the shared disk connection point 502-x can be referenced from its own node. Means for moving the print data on the shared disk to the spool area of the own node, and means for activating the spool control / print output unit.

In the spool management table 500, node names 501-1 to 501-1 having print processing in the cluster system are stored.
N and a shared disk connection point 50 on a shared disk which is spooled by the node corresponding to the node name
2-1 to N are held.

The print takeover processing unit 102
Operates by program control. Print takeover processing unit 10
In step 2, reference is made to the spool management table holding the allocation locations of the print spool areas 151 and 251 used by each node in the cluster system, and whether a shared disk connection point of a spool other than the own node exists in the own node. Check. If it is connected, the spool executes the operation of transferring the print data being spooled to the spool of the own node 100.

Therefore, the print processing system is taken over without a failure notification or spool takeover request from the print management manager.

Hereinafter, the operation of the present embodiment will be described in detail.

First, when a node 200 constituting a cluster fails, another node 100 within the cluster
Detects a failure of the node 200. When the node 100 takes over the processing of the node 200, the processing node is switched by the node switching control unit 101.

The node switching control unit 101 connects the shared disk 250 to the node 100 and activates a node switching process corresponding to each process being executed in the node 200. In the present invention, the print data is taken over in the takeover process related to the print process.

FIG. 2 shows the sharing of the node names 501-1 to 501-N of the nodes constituting the cluster system and the print spool area used as the print data storage area by the print processing system of each node. It is a spool management table having connection points 502-1 to 502-N to a disk.

The spool management table 500 stores information on the node 5
Each of 01-1 to 501-N holds the same content as one of the print environment setting information.

FIG. 3 is a flowchart for explaining the processing in the print takeover processing unit 102.

In step S401, it is determined whether the spool area of the shared disk connection point 502 other than the own node can be referenced by referring to the spool management table 500. If the spool can be referred to, it is checked in step S402 whether print data exists in the spool area. In steps S403 and S404, the print processing of the own node is started in order to transfer the print request of step S405 from the spool area to the own spool area.

Next, a specific example will be described. The operator or the business application inputs an instruction to request the print data from the printer 300 to the print request unit 203.

The print request unit 203 transmits a print data file containing print data and a print data information file containing information such as an output printer relating to the print data and attributes of the print data to the spool control / print output unit 204. I do.

The spool control / print output unit 204 temporarily stores the received print request in the spool area 25 on the shared disk.
Save to 1. When the printer is ready for printing, the spool control / print output unit 204 takes out a print request that can be output by the printer from the spool area and sends it to the printer. The print processing system performs processing of transmitting print data to the printer and deleting the print data from the spool area when output is completed.

If a failure occurs in the node 200 before the print request is transmitted to the printer or before the printing is completed, that is, in a state where the print data exists in the spool area 251, the node switching control of the node 100 is performed. The unit 101 detects a failure of the node 200. Node 100
However, the node switching control unit 101 detects a failure of the node 200 and executes the print takeover process registered in advance so that the process performed by the node 200 can be taken over by the node 100.

The print takeover processing unit 102 checks whether the node name 501-x in the spool management table is its own node name. If not the local node, shared disk connection point 5
It is checked whether 02-x can be referred to (step S401).
If the shared disk connection point 502-x can be referred to, it is checked whether a print request exists in the spool. It refers to the contents of the print data information file of the print request and takes over to the spool of its own node in consideration of the output printer, output priority, and the like. Since the contents of these print data information files are generally known in a print processing system having a print data transfer function, detailed description will be omitted.

While the embodiment of the present invention has been described above, a print processing cluster system that does not use a spool management table is also conceivable.

That is, since the print processing system does not need to be activated on a plurality of nodes constituting the cluster, the same spool area on the shared disk is shared between the active print processing system and the standby print processing system. Just use it.

FIG. 4 is a block diagram of a print processing cluster system sharing a spool area. As shown in FIG. 4, the shared disk is connected to a node that operates the print processing system. If a failure occurs in the operating node, it determines the takeover, connects the shared disk 700 to one of the standby nodes, and executes the print takeover processing unit 602.
Is started.

FIG. 5 is a flowchart for explaining the operation of the print processing cluster system sharing the spool area.

In the case of this embodiment, since the spool area uses the same area in the print processing system in the cluster system, it is confirmed that the spool area is accessible from its own node in the print takeover processing. Then, if access is possible, the spool control / print output unit is activated (step S502).

[0039]

According to the present invention described above, in a print processing system of a cluster system composed of a plurality of nodes, when a node is switched, the location of the takeover spool area and the corresponding node name are recorded. Is provided, and a print takeover process for taking over the print by referring to the information in the spool management table enables the takeover between the nodes of the print processing system without the presence of the print process management manager. .

According to the present invention, since the same spool area is used, it is not necessary to move the print data.
Further, since the printing system does not need to be started on a plurality of nodes in the cluster system, the CPU load on the data processing device is reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram of a print processing cluster system according to the present invention.

FIG. 2 is a diagram showing an example of a spool management table.

FIG. 3 is a time chart for explaining the operation of the print processing cluster system of the present invention.

FIG. 4 is a block diagram of another print processing cluster system according to the present invention.

FIG. 5 is a time chart for explaining the operation of another print processing cluster system of the present invention.

FIG. 6 is a block diagram of a conventional cluster system.

[Explanation of symbols]

 100, 200 nodes 101, 201 node switching control unit 102, 202 print takeover processing unit 103, 203 print request unit 104, 204 spool control / print output unit 150, 250 shared disk device 151, 251 spool area 300, 301 printer device 500 Spool management table

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-9-293059 (JP, A) JP-A-9-34852 (JP, A) Table 9-9506731 (JP, A) Yoshiya Mori et al. Aiming to Improve Server Cluster Availability ”Information Processing Vol. 39, Masaru Sakai, Naoki Izumida, Kazuhiro Shimada“ UN IX System DS / 90 Cluster System ”FUJITSU (58) Int.Cl. ⁷ , DB name) G06F 3/12

Claims (5)

(57) [Claims] 1. A node switching control unit, a print takeover processing unit that takes over print processing based on an instruction from the node switching control unit, and a spool control / printing that outputs print data based on an instruction from the print takeover processing unit A cluster including a plurality of nodes each including an output unit, each of the nodes including a disk device having a print spool area for storing print data, and connecting the node and a plurality of printer devices to a common bus line The system, wherein the takeover processing unit stores the print data file in the
Once via the spool control / print output unit, the print spool
Means for storing in a pool area, used by the printer device
Wait until it is possible to copy the print data file
The printer device via a spool control / print output unit
Means for outputting to the printer device the print data
After printing is completed, the print information file is
Means for deleting from the print spool area, wherein the print data file is stored in the print spool area.
In the one of the nodes,
If harm occurs, the takeover process of another node
The control unit detects the failure , refers to the print spool management table,
Find the connection point, and the one before the failure occurred
Node's print data information file is taken over to its own node
Technique, waiting for the printer device can be used, the indicia
Print data file to the spool control / print output unit.
Output to the printer device via the printer device, after the printer device finishes printing the print data.
The print information file from the print spool area.
A cluster system characterized by being deleted. 2. The print processing cluster system according to claim 1, wherein the disk device stores a print spool management table in which names of nodes that can be connected to each other are described. 3. A node switching control unit, a print takeover processing unit that takes over print processing based on an instruction from the node switching control unit, and a spool control / printing unit that outputs print data based on an instruction from the print takeover processing unit. A plurality of nodes each including an output unit, each of the nodes includes a disk device having a print spool area for storing print data, and the node and a plurality of printer devices are connected to a common bus line. A processing cluster system, further comprising a node on the bus line of the cluster system.
Node switching control unit, and instructions from the node switching control unit
A print takeover processing unit that takes over print processing based on
Print data is output based on the instruction of the print takeover processing unit
A spool control / print output unit that performs the transfer.
Once via the spool control / print output unit, the print spool
Means for storing in a pool area, used by the printer device
Wait until it is possible to copy the print data file
The printer device via a spool control / print output unit
Means for outputting to the printer device the print data
After printing is completed, the print information file is
Means for deleting from the print spool area, wherein the print data file is stored in the print spool area.
In the one of the nodes,
If harm occurs, the takeover process of the standby node
The control unit detects the failure and stores the print data information file of the failed node.
The node takes over, waits for the printer device to become available, and
Print data file to the spool control / print output unit.
Output to the printer device via the printer device, after the printer device finishes printing the print data.
The print information file from the print spool area.
A cluster system characterized by being deleted. 4. A node switching control unit, a print takeover processing unit that takes over a printing process based on an instruction from the node switching control unit, and a spool control / printing unit that outputs print data based on an instruction from the print takeover processing unit. An output unit and a plurality of nodes, each of the nodes includes a disk device having a print spool area for storing print data, connecting the node and a plurality of printer devices to a common bus line, The disk device is a print processing method using a cluster system that stores a print spool management table in which node names that can be connected to each other are described, wherein a print data file containing print data is stored in one of the nodes.
And temporarily stores the print data file in the print spool area via the spool control / print output unit, and waits until the printer device becomes usable. Outputting a print data file to the printer via the spool control / print output unit, after the printer has finished printing the print data, deleting the print information file from the print spool area; When a failure occurs in the one of the nodes while the print data file is stored in the print spool area, the takeover processing unit of another node detects the failure and performs the printing. Referring to the spool management table, the connectable connection point is found, and the print data information of one of the failed nodes Taking over the file to its own node, waiting for the printer device to become available, outputting the print data file to the printer device via the spool control / print output unit, wherein the printer device A printing processing method using a cluster system, wherein the print information file is deleted from the print spool area after printing of the print information is completed. 5. A node switching control unit, a print takeover processing unit that takes over a printing process based on an instruction from the node switching control unit, and a spool control / printing that outputs print data based on an instruction from the print takeover processing unit. A cluster including a plurality of nodes each including an output unit, each of the nodes including a disk device having a print spool area for storing print data, and connecting the node and a plurality of printer devices to a common bus line The system further comprises a node switching control unit, a print takeover processing unit that takes over a print process based on an instruction from the node switching control unit, and an instruction from the print takeover processing unit on the bus line of the cluster system. One or more standby nodes having a spool control / print output unit for outputting print data based on the A print processing method using a raster system, wherein a print data file including print data is stored in one of the nodes.
And temporarily stores the print data file in the print spool area via the spool control / print output unit, and waits until the printer device becomes usable. Outputting a print data file to the printer via the spool control / print output unit, after the printer has finished printing the print data, deleting the print information file from the print spool area; When a failure occurs in the one of the nodes while a print data file is stored in the print spool area, the takeover processing unit of the standby node detects the failure, and The node takes over the print data information file of the generated node to its own node, waits until the printer becomes available, Outputting a data file to the printer via the spool control / print output unit, and deleting the print information file from the print spool area after the printer has finished printing the print data. A print processing method using a cluster system.