JP2001034510A - Device and method for crash dump management - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make analyzable an abnormality of a diskless computer without causing any futility in terms of system configuration by providing a transfer means which transfers a dump file processed by a file processing means to a storage medium of a computer having an IP address that is acquired by an an abnormal processing means. SOLUTION: When a diskless computer has an abnormality and its kernel stops, an abnormal processing module 13 instructs a transfer IO module 14 to acquire an IP address of a server computer 2 and a file name of the transfer destination. An image stored in a memory 11 is processed into a crash dump file to which the file name of the transfer destination is added. The module 14 acquires the IP address of the computer 2 and the file name of the transfer destination by means of a TFTP 15, i.e., a file transfer standard protocol and transfers the processed crash dump file to a storage medium 24 of the computer 2.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a crash dump management device and a crash dump management method that enable analysis of an abnormal state in a diskless computer.

[0002]

2. Description of the Related Art In a conventional crash dump management apparatus, when an error occurs in a diskless computer and the execution of a kernel becomes impossible, firmware which is a memory-independent process is stored in an image (crash) stored in the memory of the computer. Dump) to a disk device or DAT device connected to the computer via a SCSI bus,
Enables system error analysis.

In the crash dump management device disclosed in Japanese Patent Application Laid-Open No. 7-146807, when a failure occurs in a terminal, a host computer outputs a dump collection command, so that a terminal installed in a remote place can be monitored. After outputting the memory dump to the terminal file, the automatic power controller of the terminal temporarily turns off the power of the terminal and monitors the elapse of a certain time.

When a host computer detects a carrier signal after a lapse of a predetermined time, the automatic power supply controller turns on the power of the terminal and starts an initialization program to initialize the terminal. Then, when the host computer outputs the dump file download command, the automatic power supply controller transfers the dump file stored in the terminal to the host computer, and enables the system to analyze the abnormality.

[0005]

Since the conventional crash dump management device is configured as described above, when an error occurs in a diskless computer, a crash dump at the time of the error is transferred to a disk device or a DAT device, and the system is crashed. Can be analyzed, but it is not known when a system error occurs. Therefore, even if there is no other use, it is necessary to keep the disk device and DAT device on standby at all times, resulting in waste in system configuration. There were issues that occurred.

[0006] In the case of the crash dump management device disclosed in Japanese Patent Application Laid-Open No. 7-146807, the dump file stored in the terminal is initialized by initializing the terminal, so that the host computer can store the dump file. In some cases, empty data is transferred, and in this case, it becomes impossible to analyze the abnormality of the system.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and a crash dump management device and a crash dump management device capable of analyzing an abnormality that has occurred in a diskless computer without causing waste in the system configuration. The purpose is to obtain a crash dump management method.

[0008]

According to the present invention, there is provided a crash dump management device, comprising:
Transfer means for transferring a dump file processed by the file processing means to a storage medium of a computer having a P address is provided.

[0009] The crash dump management apparatus according to the present invention is provided with a transfer means for transferring a dump file using a file transfer standard protocol.

[0010] The crash dump management apparatus according to the present invention is provided with a management table for managing the IP addresses of other computers connected to the network and the file names of transfer destinations in the diskless computer.

[0011] A crash dump management device according to the present invention is such that a management table is registered in a nonvolatile memory in a diskless computer.

In the crash dump management device according to the present invention, when the size of the dump file processed by the file processing means exceeds the maximum size permitted by the file transfer standard protocol, the transfer means divides and transfers the dump file. Is provided.

[0013] The crash dump management device according to the present invention manages the transfer history of the divided file when the dump file is divided and transferred, and when the transfer of the divided file is completed, the divided file is transferred to the original dump file. A transfer means is provided for collecting the information.

In the crash dump management device according to the present invention, when the IP address of a plurality of computers is registered in the management table, if the transfer of the dump file to any computer fails, the transfer for changing the computer to be transferred is performed. Means are provided.

The crash dump management device according to the present invention is provided with transfer means for managing a transfer protocol corresponding to each computer and changing the transfer protocol each time the computer to be transferred is changed.

In the crash dump management method according to the present invention, the processed dump file is transferred to a storage medium of a computer having the acquired IP address.

In the crash dump management method according to the present invention, a dump file is transferred by using a file transfer standard protocol.

In the crash dump management method according to the present invention, a management table for managing the IP address of another computer connected to the network and the file name of the transfer destination is provided in the diskless computer.

In the crash dump management method according to the present invention, a management table is registered in a nonvolatile memory in a diskless computer.

In the crash dump management method according to the present invention, when the size of a dump file exceeds the maximum size allowed by the file transfer standard protocol, the dump file is divided and transferred.

According to the crash dump management method of the present invention, when a dump file is divided and transferred, the transfer history of the divided file is managed, and when the transfer of the divided file is completed, the divided file is transferred to the original dump file. It is intended to be put together.

In the crash dump management method according to the present invention, if the IP address of a plurality of computers is registered in the management table and the transfer of the dump file to any computer fails, the computer to be transferred is changed. It was made.

In the crash dump management method according to the present invention, a transfer protocol corresponding to each computer is managed, and the transfer protocol is changed every time the computer to be transferred is changed.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below. Embodiment 1 FIG. FIG. 1 is a configuration diagram showing a crash dump management device according to a first embodiment of the present invention. In the figure, reference numeral 1 denotes a diskless computer having no storage medium;
Reference numeral 2 denotes a server computer (other computer) that manages a crash dump file when the kernel of the diskless computer 1 is abnormal, and 3 denotes a diskless computer 1 and a server computer 2.
Is a LAN (network).

Reference numeral 11 denotes the memory of the diskless computer 1, 1
Reference numeral 2 denotes firmware of a memory-independent process of the diskless computer 1, and reference numeral 13 denotes a transfer IO module 14 for the server computer 2 connected to the LAN 3 when an error occurs in the diskless computer 1 and the kernel is stopped. An error processing module (error processing means, file processing means) for instructing acquisition of an IP address and a transfer destination file name, and processing an image stored in the memory 11 into a crash dump file to which the transfer destination file name is added; 14 uses a TFTP15 which is a file transfer standard protocol under the instruction of the abnormality processing module 13,
A transfer IO module (abnormality processing means, transfer means) for acquiring the IP address of the server computer 2 and the file name of the transfer destination and transferring the crash dump file processed by the error processing module 13 to the storage medium 24 of the server computer 2 , 15 are TFTP which is a file transfer standard protocol.

Reference numeral 21 denotes a bootptab that manages the IP address of the server computer 2 and the file name of the transfer destination.
BOOTP for transmitting the IP address and transfer destination file name stored in ab21 to the diskless computer 1,
23 is a TFTP which is a file transfer standard protocol, 2
Reference numeral 4 denotes a large-capacity storage medium for storing a crash dump file transmitted from the diskless computer 1. FIG.
5 is a flowchart showing a crash dump management method according to Embodiment 1 of the present invention.

Next, the operation will be described. The error processing module 13 transfers the image stored in the memory 11 to another computer when an error occurs in the diskless computer 1 and the kernel is stopped, so that the abnormal state of the diskless computer 1 can be analyzed. Transfer I
The O module 14 is started.

The transfer IO module 14 obtains the IP address of the computer connected to the LAN 3 and the file name of the transfer destination by using the TFTP 15 which is a file transfer standard protocol under the instruction of the abnormality processing module 13. A notification request such as an IP address is output to the LAN 3 (step ST1). Thereby, when the BOOTP 22 of the server computer 2 receives the notification request from the LAN 3 (in the example of FIG. 1, only the server computer 2 is connected to the LAN 3 except for the diskless computer 1, only the server computer 2 receives the notification request). Using TFTP23, which is a file transfer standard protocol, and
The IP address of the server computer 2 and the file name of the transfer destination stored in b21 are output to the LAN 3 (step ST2).

When the server computer 2 outputs the IP address and the file name of the transfer destination to the LAN 3, the transfer IO module 14 receives the IP address of the server computer 2 and the file name of the transfer destination from the LAN 3 and abnormally converts the file name. Output to the processing module 13. Error processing module 1
3 receives the transfer destination file name from the transfer IO module 14 and processes the image stored in the memory 11 into a crash dump file to which the transfer destination file name is added (step ST3).

When the abnormality processing module 13 processes the crash dump file, the transfer IO module 14
TFTP15,23 which is a file transfer standard protocol
A computer having the previously received IP address,
That is, the crash dump file is transferred to the storage medium 24 of the server computer 2 (step ST4). As a result, the abnormal state of the diskless computer 1 can be analyzed.

As is clear from the above, the first embodiment
According to the configuration, the IP address of the server computer 2 is acquired, and the crash dump file processed by the abnormality processing module 13 is transferred to the storage medium 24 of the server computer 2. Just to transfer the crash dump file, it is possible to save the crash dump file without always putting the disk device or the DAT device on standby. That is, there is an effect that it is possible to analyze an abnormality that has occurred in the diskless computer 1 without causing waste in the system configuration.

Embodiment 2 FIG. FIG. 3 is a configuration diagram showing a crash dump management device according to a second embodiment of the present invention. In the figure, the same reference numerals as those in FIG. 1 denote the same or corresponding parts, and a description thereof will be omitted. Reference numeral 25 denotes a crash dump transfer management table (management table) for managing the IP addresses and transfer destination file names of other computers connected to the LAN 3.

Next, the operation will be described. In the first embodiment, when the kernel is stopped, a notification request such as an IP address is output to the LAN 3 to obtain the IP address and the file name of the transfer destination from the server computer 2 which is another computer. However, the crash dump transfer management table 2 for managing the IP addresses of other computers connected to the LAN 3 and the file names of transfer destinations in advance
5 may be incorporated in the firmware 12 of the memory-independent process of the diskless computer 1. This has the effect of eliminating the need to obtain the IP address and the file name of the transfer destination from another computer.

Embodiment 3 FIG. 4 is a configuration diagram showing a crash dump management device according to Embodiment 3 of the present invention. In the figure, the same reference numerals as those in FIG. 26 is a non-volatile memory of the diskless computer 1, 27 is a transfer destination information management table (management table) for managing the IP address, transfer destination file name, transfer count, and final transfer time stamp of another computer connected to the LAN 3. It is.

Next, the operation will be described. In the second embodiment, the crash dump transfer management table 25 for managing the IP addresses of other computers and the file names of transfer destinations
Is incorporated in the firmware 12 of the memory-independent process. However, the transfer destination information management table 27 for managing the IP address of another computer, the file name of the transfer destination, the transfer count, and the final transfer time stamp is non-volatile. The information may be registered in the memory 26.

This makes it easy to manage the number of times and time of transfer of the crash dump file, easily change the computer to be transferred and the storage location of the file, and reduce the power of the diskless computer 1. Even so, there is an effect that the management information of the transfer destination information management table 27 is guaranteed.

Embodiment 4 FIG. In the first to third embodiments, the case where the crash dump file is transferred at one time has been described. However, when the size of the crash dump file exceeds the maximum size allowed by the TFTPs 15 and 23, the crash dump file is divided. What is necessary is just to transfer.

FIG. 5 is an explanatory diagram showing a transfer destination information management table according to the fourth embodiment. In FIG. 5, the total crash dump file size indicating the memory size of the crash dump file is shown. Is a transfer destination computer address INDEX indicating an index of a transfer destination computer address, is a transfer destination computer address indicating an IP address of a transfer destination computer set by a user, is a file INDEX indicating an index of a transfer file, and is set by a user. The transfer destination file name is the time at which the transfer is completed. FIG. 6 is a flowchart showing a crash dump management method according to the fourth embodiment of the present invention.

Next, the operation will be described. First, transfer I
When transferring the crash dump file, the O module 14 refers to the total crash dump file size in the transfer destination information management table 27, and the size of the image stored in the memory 11 exceeds 16 MB (TFTP1).
(Step ST11).

If the size of the image does not exceed 16 MB, the transfer request enable flag of the file INDEX is turned ON (step ST14), and the transfer request is enabled. In this case, since the number of transfer files is one, the number of transfer request possible flags is also one. On the other hand, when the size of the image exceeds 16 MB, when the image is divided and transferred in units of 16 MB, the number of transferred files is calculated (step ST12). Then, the transfer request enable flags of the file INDEX are prepared for the number of transfer files, the transfer request enable flags corresponding to the respective transfer files are turned ON (step ST13), and the transfer request is enabled.

As described above, the transfer request enable flag is set to O
If N is set, the IP address of the transfer destination computer registered in the transfer destination computer address of the transfer destination information management table 27 is obtained (step ST15), the transfer request possible flag is ON, and the file INDEX number is The minimum file name is obtained (step ST16).

Next, the transfer IO module 14 divides the frozen image of the memory 11 into 16 MB units, processes a crash dump file in 16 MB units (step ST 17), and uses the TFTPs 15 and 23 to generate 16 MB crash dump files. The file is transferred to the storage medium 24 of another computer (step ST).
18). The file name of the crash dump file in units of 16 MB is the file name obtained in step ST16.

Then, the transfer IO module 14
When the transfer of the crash dump file in MB units is completed, the transfer completion flag of the file INDEX is turned on (step ST19), and the file INDEX is turned on.
Turn off the transfer request enable flag (set the transfer request
N) to make a transfer request rejection state (step ST2).
0).

The transfer IO module 14 has a 16M
When the transfer of the crash dump file in units of B is completed, the transfer completion time is recorded in the time stamp area of the transfer destination information management table 27 (step ST2).
1). The transfer IO module 14 determines whether or not all the images stored in the memory 11 have been transferred (step ST22). If the transfer of all the images has been completed, a series of processing ends. If the transfer of the image has not been completed, the process returns to step ST16 to continue the process.

Thus, even if the size of the image stored in the memory 11 is large and the file cannot be transferred at one time, the image stored in the memory 11 can be transferred into a plurality of crash dump files. The abnormal state of the diskless computer 1 can be analyzed. However, since the transfer histories of a plurality of crash dump files are managed in the non-volatile memory 26, the plurality of crash dump files can be analyzed by referring to the transfer histories. They can be combined into one crash dump file.

As is apparent from the above, the fourth embodiment
According to the description, the image stored in the memory 11 is 16
If the size is larger than MB, the image is
B, and is configured to process and transfer a crash dump file in units of 16 MB.
In the case where the size of the image stored in the diskless computer is large and the files cannot be transferred at one time, an effect of analyzing the abnormal state of the diskless computer 1 can be obtained.

Embodiment 5 FIG. 7 is a configuration diagram showing a crash dump management device according to a fifth embodiment of the present invention. In the figure, the same reference numerals as those in FIG. 2a to 2d are server computers (other computers) similar to the server computer 2.

FIG. 8 is an explanatory view showing a transfer destination information management table according to the fifth embodiment. In FIG. 8, the total crash dump file size indicating the memory size of the crash dump file is shown. Number of registered computers and IND
Transfer destination computer registration information indicating information used for linking with EX, a computer address INDEX and a file / address INDEX (address) indicating a file INDEX, a computer IP address set by a user, and a file INDEX and computer address I
File / address INDEX for recording NDEX number, transfer request enable / disable flag and transfer completion / incomplete flag
(File) is the file name of the transfer destination set by the user, and is the time when the transfer is completed. FIG. 9 is a flowchart showing a crash dump management method according to the fifth embodiment of the present invention.

Next, the operation will be described. First, transfer I
When transferring the crash dump file, the O module 14 refers to the total crash dump file size in the transfer destination information management table 27, and the size of the image stored in the memory 11 exceeds 16 MB (TFTP1).
(Step ST11).

If the size of the image does not exceed 16 MB, the transfer request enable flag of the file / address INDEX is turned on (step ST14), and the transfer request is enabled. In this case, since the number of transfer files is one, the number of transfer request possible flags is also one. on the other hand,
If the size of the image is over 16MB,
When the image is divided and transferred in units of 16 MB, the number of transferred files is calculated (step ST).
12). Then, the transfer request permission flags of the file / address INDEX are prepared for the number of transfer files, and the transfer request permission flags corresponding to the respective transfer files are turned ON (step ST13), and the transfer request is enabled.

As described above, the transfer request enable flag is set to O
If N is set, the IP address of the transfer destination computer registered in the transfer destination computer address of the transfer destination information management table 27 is obtained (step ST15), the transfer request possible flag is ON, and the file INDEX number is The minimum file name is obtained (step ST16).

Next, the transfer IO module 14 divides the image of the memory 11 in the frozen state into 16 MB units, processes a crash dump file in 16 MB units (step ST17), and uses the TFTPs 15 and 23 to generate 16 MB crash dump files. The file is transferred to the storage medium 24 of another computer (step ST).
18). The file name of the crash dump file in units of 16 MB is the file name obtained in step ST16.

After performing the file transfer process, the transfer IO module 14 determines whether or not the transfer has been normally completed (step ST31). That is, it is checked whether or not the transfer has been abnormally terminated due to the transfer destination device being Full or no response, and the transfer has not been completed.

If the file transfer processing has not been completed normally, it is determined whether or not the transfer destination computer is registered in addition to the previously transferred computer by referring to the transfer destination computer registration information ( Step ST32).

If another computer is registered as the transfer destination, the computer address is incremented (step ST33), and processing for disabling the computer address for which transfer failed is executed (step ST3).
4), the process returns to step ST15. On the other hand, when the transfer destination computer is not registered, the transfer completion flag of the file / address INDEX is turned off to make the transfer incomplete (step ST35), and the time stamp at the time of error processing is stored in the time stamp area. Recording is performed (step ST36).

If the file transfer processing has been completed normally, the transfer completion flag of the file / address INDEX is turned ON (step ST19), and the transfer request enable flag of the file / address INDEX is turned off.
F (transfer request rejection flag is turned ON) to set a transfer request rejection state (step ST20).

Also, the transfer IO module 14
When the transfer of the crash dump file in units of B is completed, the transfer completion time is recorded in the time stamp area of the transfer destination information management table 27 (step ST2).
1). The transfer IO module 14 determines whether or not all the images stored in the memory 11 have been transferred (step ST22). If the transfer of all the images has been completed, a series of processing ends. If the transfer of the image has not been completed, the process returns to step ST16 to continue the process.

As is clear from the above, the fifth embodiment
According to the above, when the IP address of a plurality of computers is registered in the transfer destination information management table 27, if the transfer of the crash dump file to any computer fails, the computer to be transferred is changed. ,
Even if the transfer of the crash dump file to any computer fails, the effect of reliably saving the crash dump file is achieved.

Embodiment 6 FIG. FIG. 10 is a configuration diagram showing a crash dump management apparatus according to Embodiment 6 of the present invention. In the figure, the same reference numerals as those in FIG. 7 denote the same or corresponding parts, and a description thereof will be omitted. 15a to 15d are file transfer standard protocols of different types, 23a to 2
3d is a file transfer standard protocol used by the computers 2a to 2d, and 24a to 24d is a storage medium of the computers 2a to 2d.

FIG. 11 is an explanatory view showing a transfer destination information management table according to the sixth embodiment. In FIG. 11, the same reference numerals as those in FIG. 8 denote the same or corresponding parts, and a description thereof will be omitted. Is transfer destination computer use protocol information indicating protocol information used by the transfer destination computer. FIG. 12 is a flowchart showing a crash dump management method according to the sixth embodiment of the present invention.

Next, the operation will be described. In the fifth embodiment, the case where the crash dump file is transferred to any of the computers by using the same file transfer standard protocol has been described.
2d use different types of file transfer standard protocols when transferring a crash dump file, the file transfer standard protocol used by the transfer target computer from the transfer destination computer use protocol information in the transfer destination information management table 27. (Step ST4)
1) The crash dump file may be transferred using the standard file transfer protocol (step ST42).

As is apparent from the above, the sixth embodiment
According to, the file transfer standard protocol corresponding to each computer is managed, and the file transfer standard protocol is changed every time the computer to be transferred is changed.
When a plurality of computers using different types of file transfer standard protocols are connected to the LAN 3, all of the computers can be used as the computer to which the crash dump file is to be transferred, thereby saving system resources. It has an effect that can be utilized for.

[0063]

As described above, according to the present invention, the transfer means for transferring the dump file processed by the file processing means to the storage medium of the computer having the IP address obtained by the abnormality processing means is provided. Since the configuration is provided, it is possible to analyze an abnormality that has occurred in the diskless computer without causing waste in the system configuration.

According to the present invention, since the transfer means for transferring the dump file by using the file transfer standard protocol is provided, most of the computers connected to the network can be set as the transfer target computers. There is.

According to the present invention, the management table for managing the IP address of another computer connected to the network and the file name of the transfer destination is provided in the diskless computer. This has the effect of saving the trouble of acquiring the file name of the transfer destination.

According to the present invention, since the management table is registered in the non-volatile memory in the diskless computer, the management information in the management table is guaranteed even if the power of the diskless computer goes down. is there.

According to the present invention, when the size of a dump file processed by the file processing means exceeds the maximum size permitted by the file transfer standard protocol, a transfer means for dividing and transferring the dump file is provided. With this configuration, the dump file can be transferred even when the size of the image stored in the memory is large and the file cannot be transferred at one time.

According to the present invention, when a dump file is divided and transferred, the transfer history of the divided file is managed, and when the transfer of the divided file is completed, the transfer means for integrating the divided file into the original dump file Thus, even if the size of the image stored in the memory is large and the file cannot be transferred at one time, it is possible to analyze an abnormality occurring in the diskless computer.

According to the present invention, when the IP address of a plurality of computers is registered in the management table, if the transfer of the dump file to any computer fails, a transfer means for changing the computer to be transferred is provided. Therefore, even if the transfer of the dump file to an arbitrary computer fails, the dump file can be surely saved.

According to the present invention, the transfer protocol corresponding to each computer is managed, and the transfer means for changing the transfer protocol each time the transfer target computer is changed is provided. When a plurality of computers that use the transfer standard protocol are connected to a network, all of the computers can be used as a dump file transfer target computer, and the system resources can be effectively used. There is.

According to the present invention, the processed dump file is transferred to the storage medium of the computer having the acquired IP address, so that the diskless computer can be used without causing waste in the system configuration. There is an effect that the abnormality that has occurred can be analyzed.

According to the present invention, since the dump file is transferred by using the file transfer standard protocol, there is an effect that almost all computers connected to the network can be set as transfer target computers.

According to the present invention, the management table for managing the IP address of another computer connected to the network and the file name of the transfer destination is provided in the diskless computer. This has the effect of saving the trouble of acquiring the file name of the transfer destination.

According to the present invention, since the management table is registered in the non-volatile memory in the diskless computer, the management information in the management table is guaranteed even if the power of the diskless computer goes down. is there.

According to the present invention, when the size of the dump file exceeds the maximum size permitted by the file transfer standard protocol, the dump file is divided and transferred, so that the image stored in the memory is stored. There is an effect that the dump file can be transferred even when the size of the file is large and the file cannot be transferred at one time.

According to the present invention, when a dump file is divided and transferred, the transfer history of the divided file is managed, and when the transfer of the divided file is completed, the divided file is integrated into the original dump file. Because of the configuration, even if the size of the image stored in memory is large and the file can not be transferred at once,
There is an effect that an abnormality generated in the diskless computer can be analyzed.

According to the present invention, when the IP address of a plurality of computers is registered in the management table, if the transfer of the dump file to any computer fails, the computer to be transferred is changed. Even if the transfer of the dump file to any computer fails, the dump file can be saved without fail.

According to the present invention, the transfer protocol corresponding to each computer is managed, and the transfer protocol is changed each time the transfer target computer is changed. When a plurality of computers to be used are connected to a network, all the computers can be set as computers to which the dump file is to be transferred, and there is an effect that system resources can be effectively used.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a crash dump management device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing a crash dump management method according to the first embodiment of the present invention.

FIG. 3 is a configuration diagram showing a crash dump management device according to a second embodiment of the present invention.

FIG. 4 is a configuration diagram illustrating a crash dump management device according to a third embodiment of the present invention.

FIG. 5 is an explanatory diagram showing a transfer destination information management table according to Embodiment 4 of the present invention.

FIG. 6 is a flowchart showing a crash dump management method according to Embodiment 4 of the present invention.

FIG. 7 is a configuration diagram illustrating a crash dump management device according to a fifth embodiment of the present invention.

FIG. 8 is an explanatory diagram showing a transfer destination information management table according to Embodiment 5 of the present invention.

FIG. 9 is a flowchart showing a crash dump management method according to Embodiment 5 of the present invention.

FIG. 10 is a configuration diagram showing a crash dump management device according to a sixth embodiment of the present invention.

FIG. 11 is an explanatory diagram showing a transfer destination information management table according to Embodiment 6 of the present invention.

FIG. 12 is a flowchart illustrating a crash dump management method according to a sixth embodiment of the present invention.

[Explanation of symbols]

1 diskless computer, 2, 2a-2d server computer (other computer), 3LAN (network), 11
Memory, 13 error processing module (error processing means, file processing means), 14 transfer IO module (error processing means, transfer means), 15, 23 TFTP (file transfer standard protocol), 24 storage medium, 25 crash dump transfer management table (Management table), 27
Transfer destination information management table (management table).

Claims (16)

[Claims] An error processing means for acquiring an IP address and a transfer destination file name of another computer connected to a network when an error occurs in a diskless computer and a kernel is stopped, and the error processing means Obtains the file name of the transfer destination, the file processing means for processing the image stored in the memory of the diskless computer into a dump file to which the file name of the transfer destination is added, and the IP obtained by the abnormality processing means. A crash dump management device comprising: a transfer unit that transfers a dump file processed by the file processing unit to a storage medium of a computer having an address. 2. The crash dump management device according to claim 1, wherein the transfer means transfers the dump file using a file transfer standard protocol. 3. The crash dump management device according to claim 2, wherein a management table for managing an IP address of another computer connected to the network and a file name of a transfer destination is provided in the diskless computer. 4. A management table is registered in a nonvolatile memory in a diskless computer.
The crash dump management device described. 5. The transfer device according to claim 1, wherein when the size of the dump file processed by the file processing device exceeds the maximum size allowed by a file transfer standard protocol, the dump file is divided and transferred. Item 3. The crash dump management device according to Item 2. 6. When the transfer unit divides and transfers a dump file, the transfer unit manages a transfer history of the divided file,
6. The crash dump management device according to claim 5, wherein when the transfer of the divided files is completed, the divided files are combined into an original dump file. 7. The transfer means, when IP addresses of a plurality of computers are registered in the management table, if transfer of a dump file to any computer fails, the transfer target computer is changed. The crash dump management device according to claim 3. 8. The crash dump management device according to claim 7, wherein the transfer means manages a transfer protocol corresponding to each computer, and changes the transfer protocol each time the computer to be transferred is changed. 9. When an error occurs in the diskless computer and the kernel is stopped, the IP address and the transfer destination file name of another computer connected to the network are obtained and stored in the memory of the diskless computer. A crash dump management method for processing a transferred image into a dump file to which a transfer destination file name is added, and transferring the dump file to a storage medium of a computer having the IP address. 10. The crash dump management method according to claim 9, wherein the dump file is transferred using a file transfer standard protocol. 11. The crash dump management method according to claim 10, wherein a management table for managing an IP address of another computer connected to the network and a file name of a transfer destination is provided in the diskless computer. 12. The crash dump management method according to claim 11, wherein a management table is registered in a nonvolatile memory in the diskless computer. 13. The crash dump management method according to claim 10, wherein when the size of the dump file exceeds the maximum size allowed by the file transfer standard protocol, the dump file is divided and transferred. 14. When a dump file is divided and transferred, a transfer history of the divided file is managed, and when the transfer of the divided file is completed, the divided file is combined into an original dump file. Item 14. The crash dump management method according to item 13. 15. An IP address of a plurality of computers in a management table
12. The crash dump management method according to claim 11, wherein when the address is registered, if the transfer of the dump file to any computer fails, the computer to be transferred is changed. 16. The crash dump management method according to claim 15, wherein a transfer protocol corresponding to each computer is managed, and the transfer protocol is changed each time the computer to be transferred is changed.