JP3394822B2 - Network 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 network system composed of a management host and a large number of nodes, and in particular, the management host is connected to each node and each node without affecting other programs. The present invention relates to a network system capable of collectively collecting fault information of terminal devices at high speed with an optimum load.

[0002]

2. Description of the Related Art FIG. 9 is a block diagram showing a schematic configuration of a network system including a conventional management host and a large number of nodes.

In FIG. 9, reference numeral 900 denotes a management host, and 92
Reference numerals 0 and 950 denote a plurality of nodes, and 910 denotes the management host 9
00 and a plurality of nodes 920 and 950 are communication lines.

A host side terminal device 901 and a host side external recording device are connected to the management host 900, and a collection file 902 is provided in the host side external recording device.

A node side terminal device 922, 923, 952, 953 and a node side external recording device are connected to the plurality of nodes 920 and 950, and failure information files 921 and 951 are provided in the node side external recording device.

In the network system shown in FIG. 9, the management host 900 manages the failure status of the nodes 920 and 950 and the node side terminal devices 922, 923, 952 and 953 connected to the nodes 920 and 950. There is.

The nodes 920 and 950 are connected to the nodes 920 and 950 and the nodes 920 and 95, respectively.
0-side terminal devices 922, 92 connected to 0
3, 952, 953, when a failure occurs, failure information is stored in the failure information file 921, 95 of the external recording device.
Record at 1.

In the above network system, when the failure information recorded in the failure information files 921 and 951 of the external recording devices of the nodes 920 and 950 is collectively collected from the management host 900, for example, Japanese Patent Application Laid-Open No. No. 5-205074 “Data collection method of information processing system” or “Data collection device” of Japanese Patent Laid-Open No. 5-210749, one node at a time, or at the same time from all nodes collectively. I was collecting fault information.

[0009]

However, the method of collecting the failure information for each node has a problem that it takes a lot of time to complete the collection of the failure information from all the nodes.

Further, in the method of transferring the fault information file from each node to the management host, the fault information file is transferred for each node and the fault information required by the management host cannot be selected at each node. There is a problem that a large amount of failure information is required and it takes a lot of time to complete collection of failure information of all nodes.

Further, in the method of collectively collecting the failure information from all the nodes at the same time, in order to reduce the line load, CPU load and I / O load on the management host side, the failure information is transferred from each node to the management host. A time interval is provided, that is, failure information is transferred to the management host at regular intervals.

However, even when the failure information is transferred from each node to the management host with a time interval, a load such as a line load, a CPU load, an I / O load and a memory load is applied to the management host side. However, there is a problem that it may cause a slowdown of the management host or the like and may adversely affect other programs.

The present invention has been made to solve the above-mentioned problems of the prior art, and an object of the present invention is to collectively collect fault information recorded in each node from a management host in a network system. An object of the present invention is to provide a technique capable of collecting data at a high speed and with an optimum load without affecting other programs.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[0015]

Among the inventions disclosed in the present application, a brief description will be given to the outline of typical ones.
It is as follows.

(1) A management host, a plurality of nodes, and a communication line connecting the management host and a plurality of nodes, and when each node has a failure, failure information is recorded in a failure information file. In the network system configured to do so, the management host sends a failure information collection request for collecting failure information recorded in the failure information file of each node to a part of designated nodes of the plurality of nodes. And a first means for issuing the failure information, and when the failure information collection for one node among the specified nodes is completed, the failure information collection request is issued from one of the nodes other than the specified node. A second buffer for issuing to a node, and a reception buffer for receiving and temporarily storing failure information from each node,
Each of the nodes includes means for receiving a failure information collection request from the management host and transferring the failure information to the management host.

(2) A management host, a plurality of nodes, and a communication line connecting the management host and the plurality of nodes. When a failure occurs in each node, failure information is recorded in a failure information file. In such a network system, the number of times that the management host can continuously transmit the failure information collection request for collecting the failure information recorded in the failure information file of each node from each node By specifying the first means for issuing to some of the specified nodes of the plurality of nodes, and when the failure information collection for one node of the specified nodes is completed, One of the fault information collection requests is designated from the nodes other than the designated node by designating the number of times that the fault information can be continuously transmitted from the node. Second means for issuing to the node, a reception buffer for receiving and temporarily storing the failure information from each node, and after processing the failure information from each node stored in the reception buffer, Thirdly, the transmission restart request is issued to the designated node by designating the number of times that the failure information can be continuously transmitted from each of the nodes.
Each of the nodes receives the failure information collection request from the management host or the transmission restart request, and receives the failure information for the number of transmissions designated by the failure information collection request from the management host. It is characterized by comprising means for transferring to the management host.

[0018]

According to the above-mentioned means (1), the management host issues a failure information collection request for collecting failure information to a designated node which simultaneously collects failure information.

Further, when the collection of the failure information for one of the designated nodes is completed, the failure information collection request is issued to one node other than the designated node.

Each node designated by the management host transfers fault information to the management host.

The management host processes the fault information transferred from each node and stored in the reception buffer.

That is, by designating the number of nodes (multiplicity) at the same time that the management host collects the failure information, the number of nodes at the same time the failure information is collected can be limited, and the CPU load, I / O load and The memory load can be reduced.

According to the above-mentioned means (2), the management host designates the failure information collection request for collecting failure information by designating the number of transmissions (continuous transmissions) at which the failure information can be continuously transmitted from each node. , Issue to the specified node that collects the failure information at the same time.

Further, when the collection of the fault information for one of the designated nodes is completed, the fault information collection request can be transmitted continuously from each node in the fault information (the number of consecutive transmissions). ) Is specified to issue to one node other than the specified node.

Further, each designated node waits for a transmission restart request from the management host after transmitting the failure information the number of times of continuous transmission.

After the management host processes the fault information transferred from each node and stored in the reception buffer, the management host designates the number of transmissions (continuous transmission number) at which the fault information can be transmitted continuously from each node. Each of the designated nodes restarts the transmission of the failure information.

That is, since the transmission restart request from the management host to each node is suspended before the management host is overloaded, this prevents the management host from being overloaded.

Further, by designating the number of nodes (multiplicity) for collecting the failure information at the same time, the number of nodes for collecting the failure information at the same time can be limited, and the CPU load of the management host,
The I / O load and the memory load can be reduced.

According to the means (1) or (2), the memory capacity of the reception buffer of the management host is calculated from the multiplicity and the number of continuous transmissions. It is possible to estimate the virtual memory or the real memory in advance.

Further, it is possible to set the values of the multiplicity and the number of continuous transmissions so that the virtual memory or the real memory is not overloaded.

As a result, the memory is not overloaded, the I / O load is prevented from increasing due to frequent paging, and the memory is used indefinitely to prevent pressure on the memory usage of other programs. It becomes possible to do.

[0032]

Embodiments of the present invention will now be described in detail with reference to the drawings.

In all the drawings for explaining the embodiments, parts having the same function are designated by the same reference numerals, and the repeated description thereof will be omitted.

FIG. 1 is a block diagram showing a schematic configuration of a network system which is an embodiment of the present invention.

In FIG. 1, 100 is a management host and 10 is a management host.
5 is a terminal device on the host side, 103 is a collection file provided in the external recording device on the host side, 110 and 120 are nodes,
201 to 204 are terminal devices on the node side, 113 and 123 are failure information files provided on the external recording devices on the node side,
130 is a communication line.

The management host 100 is provided with a host-side communication control unit 101, a collection unit 102, a reception buffer 104 existing in the memory for queuing fault information, and a node table 106.

The nodes 110 and 120 are provided with node side communication control units 111 and 121 and reading units 112 and 122, respectively, and the fault information file 11 of the node side external recording device is provided.
A failure record is recorded in 3,123 as failure information when a failure occurs in the node.

FIG. 2 shows the fault information file 11 shown in FIG.
It is a figure which shows the format of the failure record recorded on 3,123.

As shown in FIG. 2, the fault information file 11
The failure record recorded in 3,123 includes a failure occurrence date, a terminal ID, and a failure type (failure code).

The management host 100 centrally manages the network system and requests collection of fault information.

The host-side communication control unit 101 connects the management host 100 and the nodes 110 and 120 via the communication line 130, and controls transmission / reception such as queuing received fault information in the reception buffer 104.

The collection unit 102 is operable in parallel with the communication control unit 101, takes out the failure information queued in the reception buffer 104 by the FIFO method, and records it in the collection file 103 of the host side external recording device.

The management host 100 also manages each node 11
After the session is established with 0 and 120, the node name is registered in the node table 106.

The node side communication control unit 111 controls transmission / reception with the communication control unit 101 of the management host 100.

The reading section 112 reads out failure information (failure record) from the failure information files 113 and 123 of the node side external recording device, and makes a transmission request to the communication control section 111.

The operation at the time of batch collection is to collect the failure information of all the nodes 110 and 120 by inputting a failure information collection command from the host side terminal device 105, and to collect the host side external recording device. When the failure information is recorded in the file 103 and the failure information is collected from all the nodes 110 and 120, a collection completion message is sent to the host side terminal device 1.
In the network system of the present embodiment, the management host 100 designates the number of nodes (multiplicity) at which the failure information is to be collected at the same time, and collects the failure information of each of the nodes 110 to 130 at once. It is a figure for demonstrating a case.

In FIG. 3, 100 is a management host and 10 is a management host.
Reference numeral 5 is a terminal device on the host side, 103 is a collection file provided in the external recording device on the host side, and 110, 120, and 130 are nodes.

In the figure, each node side terminal device 20
1 to 204 and the failure information files 113 and 123 provided in the node side external recording device are omitted.

Referring to FIG. 3, in the network system of this embodiment, the management host 100 designates the number of nodes (multiplicity) at which the failure information is collected at the same time, and each node 1
A case will be described in which the failure information of 10,130 is collectively collected.

First, in the network system of this embodiment, the number of nodes (multiplicity) at which the management host 100 collects fault information at the same time is m.

When a failure information collection command is input from the host-side terminal device 105, the management host 100 will
For example, referring to the node table 106 shown in FIG. 1, the failure information collection requesting the upper m nodes 110 and 120 out of the n nodes registered in the node table 106 to collect failure information at the same time. Make a request.

At each of the requested nodes 110 and 120,
Fault information files 113 and 12 of the node side external recording device
The failure information (failure record) is read from the node 3, and one failure information is transmitted at one time and the node side communication control unit 1
It is transmitted to the host side communication control unit 101 of the management host 100 via 11, 121.

Host-side communication control unit 1 of management host 100
In 01, when the failure information transmitted from each of the nodes 110 and 120 is received, it is queued in the reception buffer 104.

The collection unit 102 of the management host 100 extracts the failure information queued in the reception buffer 104 one by one and records it in the collection file 103 of the management host side external recording device.

When the collection of fault information from one of the m nodes 110 and 120 is completed, the management host 100 refers to the node table 106,
The above procedure is repeated for the node next to the top m nodes 110 and 120 registered in the node table 106.

FIG. 5 shows the number of nodes (multiplicity) at which the management host 100 collects fault information at the same time in the network system of this embodiment, and the number of transmissions that can be continuously transmitted from each node (the number of consecutive transmissions). Is a diagram for explaining a case where the failure information of each of the nodes 110 to 130 is collectively collected by designating, and is a diagram showing a sequence between the management host 100 and one node 110.

The sequence between the management host 100 and one node 110 shown in FIG. 5 specifies the number of nodes (multiplicity) at the same time shown in FIG.
This is performed for each of the nodes 110 and 120 when collectively collecting the failure information of each of the nodes 110 and 130.

Referring to FIG. 5, in the network system of this embodiment, the management host 100 simultaneously collects failure information from nodes (multiplicity) and the number of transmissions that can be transmitted continuously from each node (the number of continuous transmissions). ) Is specified to collectively collect the failure information of the nodes 110 to 130.

Let w be the number of transmissions (the number of continuous transmissions) that can be continuously transmitted from each of the nodes 110 to 130.

When the failure information batch collection command is input from the host-side terminal device 105, the management host 100, as shown in FIG.
06, for example, the upper m nodes 11 from the n nodes registered in the node table 106
A failure information collection request for simultaneously requesting the failure information collection is issued to 0 and 120.

In this case, the collection unit 1 of the management host 100
02 is the reading unit 11 of each of the m nodes 110 and 120.
2 to the collection request 301.

FIG. 6 is a diagram showing a transfer format of the collection request 301 shown in FIG.

As shown in FIG. 6, the collection request 301 is composed of collection request ID, partner node name, number of continuous transmissions, and failure information selection areas. Here, the failure information selection area is the management host 100. This is an area provided when the failure information is designated to collect the failure information.

When the failure information is specified from the management host 100 and the failure information is collected, for example, there are a failure occurrence period, a failure type (failure code) and a specific terminal device, and the failure information is correspondingly provided. In the area of choice,
Areas for failure occurrence date selection period, failure type (fault code) and terminal device ID are provided.

The reading section 112 of each of the m nodes 110 and 120 receives the collection request 301 and then transmits the collection response 302, and the failure information file 11 of the node-side external recording device.
The failure information (failure record) specified in the failure information selection area of the collection request 301 is read from
Is transmitted to the host-side communication control unit 101 of the management host 100 via the node-side communication control unit 111 w times.

FIG. 7 is a diagram showing a transfer format of the fault information 304 shown in FIG.

As shown in FIG. 7, the failure information 304 includes failure record, own node name, transmission NO. , Date of failure,
Each area is composed of a terminal ID and a fault type (fault code).

If the failure information transmitted from each of the m nodes 110 and 120 to the host side communication control unit 101 of the management host 100 is the last failure information, the transmission NO. Is added to the last fault information, for example, a flag and transmitted.

Host side communication control unit 1 of management host 100
In 01, when the failure information transmitted from each of the m nodes 110 and 120 is received, it is queued in the reception buffer 104.

The collection unit 102 of the management host 100 stores the failure information queued in the reception buffer 104 in the FIFO.
It is taken out by the O method and the failure information is recorded in the collection file 103 of the host side external recording device.

On the other hand, each of the m nodes 110, 120
The reading unit 112 is waiting for the transmission restart request 303 from the collection unit 102 after the completion of writing the collection file.

When the writing of the failure information is completed in the collection file 103 of the host-side external recording device, the collection unit 102 of the management host 100 issues a transmission restart request 303 for continuing collection to the m nodes 110 and 120. To send.

FIG. 8 is a diagram showing a transfer format of the transmission restart request 303 shown in FIG.

As shown in FIG. 8, the transmission restart request 303
Is the transmission restart request ID, partner node name, and continue collection /
It is composed of each area of the end ID.

Upon receiving the transmission restart request 303, the reading unit 112 of each of the m nodes 110 and 120 reads the fault information (fault record) from the fault information file 113 of the node side external recording device again, and the fault information 304
Is transmitted to the host-side communication control unit 101 of the management host 100 via the node-side communication control unit 111 w times.

When the communication control unit 101 of the management host 100 receives the failure information transmitted from each of the m nodes 110 and 120, the failure information is queued in the reception buffer 104.

The collection unit 102 of the management host 100 takes out the failure information queued in the reception buffer 104 and records the failure information in the collection file 103 of the host side external recording device.

When the collection of the fault information from one of the m nodes 110 and 120 is completed, that is, the transmission from the m nodes 110 and 120 to the host side communication control unit 101 of the management host 100 is performed. No. of the transmitted failure information 304. If the last failure information is added to the message, a transmission restart request 303 indicating the end of collection is sent to each m.
It transmits to the individual nodes 110 and 120.

After that, the management host 100 refers to the node table 106 and refers to, for example, the node table 106.
The above procedure is repeated for the node next to the top m nodes 110 and 120 registered in.

In the network system of this embodiment, the memory requirement of the reception buffer 104 is calculated by (multiplicity) × (number of continuous transmissions) × (maximum length of one transmission unit), and is virtualized in advance at the start of collective collection. It can be secured in memory or real memory.

In this case, as the receiving buffer 104,
Actual memory size = (multiplicity) x (number of consecutive transmissions) x (1
By securing the real memory which is the maximum length of the transmission unit), it becomes possible to eliminate the number of I / Os due to paging, and it is possible to collect data collectively at high speed.

Although the present invention has been specifically described based on the embodiments, it is needless to say that the present invention is not limited to the embodiments and various modifications can be made without departing from the scope of the invention.

[0083]

The effects obtained by the typical ones of the inventions disclosed in the present application will be briefly described as follows.

(1) By specifying the number of nodes (multiplicity) at which the management host collects failure information at the same time, the number of nodes that collect failure information at the same time can be limited, and the CPU load and I / O load of the management host can be limited. Also, the memory load can be reduced.

(2) The number of transmissions by which the management host can continuously transmit the failure information from each node (the number of continuous transmissions),
At the same time, by specifying the number of nodes (multiplicity) to collect failure information, the management host can hold a request to restart transmission from each node before the management host becomes overloaded. There is no overload, and the CPU load, I / O load, and memory load of the management host can be reduced.

(3) Since the memory amount of the reception buffer of the management host is calculated from the multiplicity and the number of continuous transmissions, it is possible to estimate the calculated memory amount in advance in the virtual memory or the real memory. Further, it is possible to set the values of the multiplicity and the number of continuous transmissions so that the virtual memory or the real memory is not overloaded.

As a result, the memory is not overloaded, the I / O load is prevented from increasing due to frequent paging, and the memory is used indefinitely to prevent the memory usage of other programs from being suppressed. It becomes possible to do.

Further, the actual memory size = (multiplicity) ×
By securing an actual memory of (number of continuous transmissions) × (maximum length of one transmission unit), it is possible to eliminate the number of I / Os due to paging, and it is possible to collect data at once at a high speed.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a network system that is an embodiment of the present invention.

FIG. 2 is a diagram showing a format of a fault record recorded in a fault information file shown in FIG.

FIG. 3 shows the network system of the present embodiment.
It is a figure for demonstrating the case where a management host designates the number of nodes (multiplicity) which collect failure information simultaneously, and collects the failure information of each node collectively.

FIG. 4 is a diagram showing a node table shown in FIG.

FIG. 5 shows the network system of the present embodiment.
When the management host specifies the number of nodes (multiplicity) that collects fault information at the same time and the number of transmissions that can be transmitted continuously from each node (number of consecutive transmissions), collects fault information of each node at once. It is a figure for explaining.

6 is a diagram showing a transfer format of the collection request shown in FIG.

FIG. 7 is a diagram showing a transfer format of the fault information shown in FIG.

8 is a diagram showing a transfer format of the transmission restart request shown in FIG.

FIG. 9 is a block diagram showing a schematic configuration of a network system including a conventional management host and a large number of nodes.

[Explanation of symbols]

100, 900 ... Management host, 101 ... Host side communication control section, 102 ... Collection section, 103, 902 ... Collection file, 104 ... Reception buffer, 105, 901 ... Host side terminal apparatus, 106 ... Node table, 110, 120,
130, 920, 950 ... Node, 111, 121 ... Node-side communication control unit, 112, 122 ... Read-out unit, 11
3,123,921,951 ... Fault information file, 13
0,910 ... communication line, 201-204,922,92
3, 952, 953 ... Node side terminal device, 301 ... Collection request, 302 ... Collection response, 303 ... Transmission restart request, 30
4 ... Fault information.

─────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-4-286238 (JP, A) JP-A-57-52938 (JP, A) JP-A-6-68273 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) H04L 12/24

Claims (3)

(57) [Claims] 1. A management host, a plurality of nodes, and a communication line connecting the management host and the plurality of nodes. When failure occurs in each node, failure information is recorded in a failure information file. In the network system configured as described above, the management host issues a failure information collection request for collecting failure information recorded in the failure information file of each node to some of the specified nodes of the plurality of nodes. And a first means for issuing a failure information collection request when collection of failure information for one node of the specified nodes is completed. And a reception buffer for temporarily storing the failure information from each node and storing the failure information from each node. A network system comprising means for transferring fault information to the management host in response to a fault information collection request from the management host. 2. A management host, a plurality of nodes, and a communication line connecting the management host and the plurality of nodes, wherein failure information is recorded in a failure information file when a failure occurs in each node. In the network system configured as described above, the number of times that the management host can continuously transmit the failure information collecting request for collecting the failure information recorded in the failure information file of each node from each node is set. A first means for designating and issuing to some designated nodes of the plurality of nodes; and, when collection of fault information for one of the designated nodes is completed, Specifying the number of times that the failure information collection request can continuously send failure information from the node, and selecting one from the nodes other than the specified node. Second means to issue to
A reception buffer that receives and temporarily stores the failure information from each node, and a failure buffer from each of the nodes to the specified node after processing the failure information from each node stored in the reception buffer. And a third means for issuing a transmission restart request by designating the number of times that information can be continuously transmitted, wherein each node receives a failure information collection request or a transmission restart request from the management host. hand,
A network system comprising means for transferring fault information to the management host for the number of transmissions designated by a fault information collection request from the management host. 3. The network system according to claim 1, wherein the first means of the management host further specifies failure information to be collected and specifies the failure information collection request. A network system characterized by issuing to nodes.