JPS61292444A - Communication control system - Google Patents

Abstract

PURPOSE:To allow a host device to continue intermitted communication without taking the intermission of communication into consideration by allowing a communication controller of a packet terminal device to hold a required sending packet of a logical channel in a buffer and to reconnect the said logical channel to a packet communication network when a disconnection command is received. CONSTITUTION:In receiving a disconnection command from a network, a packet processing section 16 stops the packet transmission of a designated logic channel but does not inform it to a host device 3. The packet processing section 16 requests the transmission of the connection request packet designating an opposite terminal device after a proper time to a frame processing section 4 and tries the rest of the disconnected logic channel. When the network is recovered, the resetting of the logic channel is succeeded and if not, the trial is repeated with a proper interval until succeeded. When the network is recovered and the logic channel is reset with the opposite terminal device, the packet processing section 16 restarts the transmission of the said logic channel.

Description

[Detailed Description of the Invention] [Summary] This is a control method for managing logical channels of packet communication in a packet terminal connected to a packet communication network.

Depending on the circumstances of the network, the network may request the packet terminal to disconnect the logical channel in progress. In this case, the communication control device within the terminal retains the required transmission packets in a buffer without disconnecting the logical channel, and after network recovery,
By resuming communication and retransmitting the required packets,
Internally, it eliminates the need for disconnection/resume processing, improving processing efficiency.

[Industrial application field]

The present invention relates to a control method for managing logical channels of packet communication in a packet terminal connected to a packet communication network.3. One message data is divided into packets of a predetermined data length or less and exchanged/transmitted. Packet communication networks based on this method are well known.

A packet terminal, such as a computer, connected to such a packet communication network (hereinafter simply referred to as a network) uses one transmission path in a time-division manner to frame packets of multiple different communications. Transfer between units and the network.

The packet terminal processes a plurality of different communications in parallel in this way, and the virtual communication path that is considered to be set for each communication in this case is called a logical channel.

The control system within the network is usually composed of computer systems, and is often duplicated to increase the reliability of the system.

If any abnormality occurs that prevents these processes from being carried out normally, the system will switch from the current system to the backup system and continue processing, but at least during the switching period from the occurrence of a failure etc. to the restart of processing. Communication processing is interrupted.

In such cases, for example, CCITT Recommendation X, 25 provides a function for requesting the packet terminal from the network to disconnect the logical channel in order to stop transmission from the terminal.

[Prior art and problems to be solved by the invention] FIG. 2 is a block diagram showing an example of the configuration of a packet terminal.

A communication control device 1 is connected to a network via a communication line 2, and a host device 3 consisting of a computer system or the like controls the exchange of packets for communication over the network.

Information sent and received through the communication line 2 to and from the network is controlled, for example, according to a known high-level transmission control procedure (referred to as HDLC), and packets are transmitted as data in the information part of a frame that follows the HDLC information format shown in FIG.

In the frame format shown in the figure, as is well known, F is a flag, A is an address field, C is a control field, ■ is an information field, and Fe2 is a frame mucinic sequence. A packet transmitted in the information section (2) generally consists of a header section consisting of packet-level control information and a data section of transmitted/received data, but there are cases in which there is no data section.

When one application program executed on a host device communicates with another terminal via this network, a known method is used, for example, to communicate with the terminal from the application to the network via the communication control device 1. A connection request is made and the network establishes a logical channel for this communication.

In the communication control device 1, the packet processing unit 6 adds a header to the data of the message sent from the application connected to each logical channel and forms a packet on the hole couch 5, and the frame processing unit 4 A transmission frame for transmitting the information is constructed and sent to the wire via the communication line 2.

In this packet transmission, normally, as is well known, a predetermined number of packets (this is called a window, and the number of packets is called a window size, for example 1 to 4 packets) can be transmitted continuously without waiting for a response from the network, and once the window size has been transmitted, the next transmission is controlled based on the reception status response from the network. ,
Improves transmission efficiency.

For example, if a failure occurs in the network and communication processing is interrupted, a packet/node of a command to disconnect a logical channel (hereinafter referred to as a disconnection command) is sent from the network.

In the communication control device 1 , when the frame processing unit 4 receives the frame of the disconnection command packet, the frame processing unit 4 passes this packet to the packet processing unit 6 via the buffer 5 .

When the packet processing unit 6 identifies this disconnection command, it notifies the corresponding application of the host device 3 that the logical channel has been disconnected, and terminates the communication.

The application issues another connection request to the communication control device 1 to restart the interrupted communication if necessary, and the communication control device 1 attempts to set up a logical channel and establishes the network. When the logical channel is set up after recovery, the communication is started as a new communication.

For this reason, even if the network recovers in a relatively short time, for example by switching to a backup device, the applications on the host device are aware of the communication interruption, perform the necessary recovery processing, and set up logical channels. Since it is necessary to start over from scratch, there are problems such as complicating applications that use packet communication and unduly delaying the recovery of communication, so an improvement has been desired.

[Means for solving problems]

FIG. 1 is a block diagram showing the configuration of the present invention.

In the figure, 11 is a communication control device, 4 is a frame processing section, 5 is a buffer, 16 is a packet processing section, and 17 is a window buffer.

[For production]

In the packet transmission process, the packet processing unit 16
The transmission packet is passed to the frame processing unit 4 via the buffer 5 as in the conventional case, but at the same time, the packet is held in the window buffer 17.

Upon receiving the disconnection command from the network, the packet processing unit 16
temporarily stops transmission of the specified logical channel, but does not notify the higher-level device 3.

After an appropriate period of time, the packet processing unit 16 requests the frame processing unit 4 to send a connection request packet specifying the destination terminal in order to attempt resetting the disconnected logical channel.

When the network is restored and a logical channel is re-established with the other terminal, the packet processing unit 16 starts communication on the logical channel by retransmitting the packets held in the window buffer 17.

With the above control, upon recovery after a network failure or the like, the host device 3 can continue the interrupted communication without being concerned with the interruption of communication.

〔Example〕

In FIG. 1, it is assumed that a communication control device 11 executes control of packet communication in the same manner as the conventional communication control device 1, except for the transmission stop processing described below.

In the packet transmission process, the packet processing unit 16
The transmission packet is passed to the frame processing section 4 via the buffer 5 as in the conventional case, but at the same time, the packet is held in the window buffer 17.

The window buffer 17 stores packets of at least twice the window size for each logical channel in response to transmission management using the window method, and stores packets of the last window that was transmitted when communication is resumed, as described below. shall have the capacity to retransmit.

For this purpose, for example, the window buffer 17 is provided with two buffer areas for each window for each logical channel, and is used in a so-called alternating buffer system.

Upon receiving the disconnection command from the network, the packet processing unit 16
stops sending packets on the specified logical channel, but
The host device 3 is not notified.

After an appropriate period of time, the packet processing unit 16 requests the frame processing unit 4 to send a connection request packet specifying the other party's terminal, and attempts to reset the disconnected logical channel.

If the network has recovered, the logical channel reconfiguration will be successful, but if not, it will be reconfigured at appropriate intervals until it is successful.
Repeat the above trial.

When the network is restored and the logical channel is re-established with the other terminal, the packet processing unit 16 resumes transmission of the logical channel.

It is assumed here that transmission begins with the retransmission of the window packet for which the last acknowledgment of receipt was received from the network.

In other words, this window may have been successfully received by the network before, but a failure occurred during the process of relaying it to the other party's terminal, and it may have been lost within the network without reaching the other party's terminal. It is from.

However, on the contrary, there is a possibility that the window includes packets that have been successfully processed, such as packets received by the other party's terminal.

For this purpose, the packet processing unit 16 receives and receives, for example, a predetermined count value indicating the number of packets received by the other party by exchanging required control information packets with the other party's terminal, and based on the count value, the number of packets received by the other party is determined. identify the first packet that is not present, and transfer the packet to the wind buffer 17.
Retransmission is performed by extracting the data from the frame, passing through the buffer 5 and the frame processing unit 4, and transmitting it to the network.

With the above control, the top bag W
3, it becomes possible to continue communication without being concerned about the interruption of communication.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, when a logical channel is disconnected when an abnormality occurs in a packet communication network, communication can be quickly resumed without making the host device aware of the communication interruption. Therefore, there is a significant industrial effect of improving the processing efficiency of bucket communication.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is a block diagram of a conventional configuration example, and FIG. 3 is an explanatory diagram of a frame configuration. In the figure, 1.11 is a communication control device, 2 is a communication line, 3 is a host device, 4 is a frame processing unit, 5 is a buffer,
6.16 is the packet processing unit, 17 is the window sofa

Claims (1)

[Claims] In a packet communication network having a function of disconnecting a logical channel for packet communication set in a packet terminal by a disconnection command to the packet terminal, a communication control device (11) of the packet terminal: When the disconnection command is received, the packet terminal continues the connection state of the logical channel and buffers the required transmission packets of the logical channel (17).
A communication control system characterized in that: (16) the logical channel is held in the packet communication network, the logical channel is reconnected to the packet communication network, and the transmission packet in the buffer (17) is retransmitted to the packet communication network.