JPH05265785A - Error recovery method for file transfer - Google Patents

Abstract

PURPOSE:To shorten the time required for recovery at the time of the occurrence of a fault by resending data from the data point, where reception was last confirmed, or following data in the case of the occurrence of the recoverable fault during the file transfer processing. CONSTITUTION:If the reception side detects a fault after reception of data 16 and data is secured up to data 11 at this time, the reception side requests the transmission side to resend data from data 12, and the transmission side resends data from data 12 after responding to this request. At this time, it is judged that transmission up to data 11 has been confirmed and preserved data ending with data 11 is logically deleted. If the transmission side detects a fault, the transmission side requests to resend data from data following the data point where reception is last confirmed, and the reception side returns the value of the data point following data, which is actually secured by itself, as the response. Thus, the error recovery processing of file transfer is performed by resending from the last confirmed data point or following data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an error recovery method in file transfer.

[0002]

2. Description of the Related Art In recent years, while the amount of data transferred in the information-oriented society has increased, the occurrence of problems in the transfer of files between different models has become a problem. Conventionally, for error recovery when such a failure occurs, as shown in FIG.
By inserting a checkpoint in the middle of the transmission data and receiving the response, the recovery from the checkpoint or the recovery from the beginning of the file is enabled.

[0003]

In such an error recovery method in the conventional file transfer, it is necessary to insert a checkpoint in the middle of the data, which complicates the processing of the program and inserts the checkpoint. By doing so, there was a problem that the transfer efficiency was significantly reduced.

According to the present invention, in error recovery processing in file transfer, the number of transmitted data is used as a data point instead of a checkpoint as a point for error recovery, and a data response is returned. By confirming the data transmission, and by resending from the data point that was confirmed last or the data after that when a failure occurs, the time required for recovery in the event of a failure can be shortened. With the goal.

[0005]

An error recovery method for file transfer according to the present invention comprises a data point recovery means for indicating the position at which data is retransmitted when a failure occurs by the number of transmitted data, and a conventional non-confirmation type primitive. The data response means for substituting a checkpoint for the data that has been a confirmation type primitive, the data storage means for storing the transmission data until the confirmation is received by the data response, and the data point confirmation Data deletion means to delete the data before the confirmation data point, and the data to be retransmitted from the last confirmed data point when a recoverable temporary failure occurs, or from the data point after that. If there is a recoverable failure during the file transfer process that has a resending means, the device that received the last confirmation By performing a retransmission from the other point, or later of the data, and performs a recovery in the event of a failure.

[0006]

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

FIG. 1 is a flow chart of a checkpointless error recovery processing method for file transfer according to the present invention.
2 shows an image of a storage area for storing data, and FIG. 3 shows a format of information regarding data for storage area management.

In FIG. 2, storage area 1 to storage area n, or storage area n and storage area 1 are logically continuous and are connected by an address to the next area.

In FIG. 3, the first data point and
The final data points both have "0" as the initial value,
The initial data address and the minimum data address both have an empty address as an initial value. The first and last data points are "1" due to the transmission of data (1)
The address to the storage area 1 is simultaneously stored in the first data address and the last data address. By sending the data (2), the final data point is (2)
Then, the address to the storage area 2 is stored in the final data address. With the transmission of the data (3) to (6), the final data point and the final data address are updated at any time. Further, the value of the head data point is updated to “6” by the reception of the data (5) response, and the head data address is also updated to the address to the storage area 6. In this way, each time the data is transmitted, the data is stored in the storage area, and when the response of the data is received, the head data point and the head data address are rewritten to logically delete the stored data.

In FIG. 1, when the receiving side detects a fault after receiving the data (16) and the data (11) is secured at that time, the receiving side sends the data (1) to the transmitting side.
The recovery request from (2) is made, the transmitting side returns a response to the request, and then the data (12) is retransmitted. At this time, the transmission side receives the recovery request from the data (12) and determines that the transmission confirmation up to the data (11) has been performed, and logically deletes the stored data up to the data (11).

Similarly, when the transmitting side detects a failure, the transmitting side requests retransmission from the data next to the last confirmed data point, whereas the receiving side does not Returns the value of the next data point of the secured data in the response. The transmitting side receiving the response retransmits from the data point designated by the receiving side.

In this way, when a failure occurs, the error is recovered in the file transfer without using the checkpoint by retransmitting from the last confirmed data point or the data after that. Processing becomes possible.

[0013]

As described above, according to the present invention, it is possible to save data, delete data, and retransmit data without using checkpoints in file transfer, and an error occurs when a recoverable failure occurs. For recovery, error recovery processing in file transfer is possible without using an inefficient method of inserting a checkpoint and only recovering from the checkpoint.

[Brief description of drawings]

FIG. 1 is a flow chart of one embodiment of the present invention.

FIG. 2 is a diagram showing an image of a storage area for storing data.

FIG. 3 is a diagram showing a format of information regarding data for storage area management.

FIG. 4 is a flowchart showing a conventional example.

[Explanation of symbols]

 1 to n storage area

Claims (1)

[Claims] 1. A checkpoint is substituted by a data point recovery means that indicates the position where data is retransmitted in the event of a failure by the number of transmitted data, and by replacing the data that was conventionally a non-confirmation type primitive with a confirmation type primitive. Data response means for saving, data storage means for storing transmission data until confirmation by the data response, and data deletion for deleting data before the confirmation data point when the data point is confirmed Means and a data resending means for resending data from the last confirmed data point or a subsequent data point when a recoverable temporary failure occurs, and a recoverable failure occurs during the file transfer process. if it occurs,
An error recovery method in file transfer, characterized in that recovery is performed when a failure occurs by retransmitting from the data point for which reception confirmation was finally performed or data after that.