JPH0239753A - Fault processing method - Google Patents

Abstract

PURPOSE:To arbitrate a transmission right at the time of fault generation without interfering the transmission of an upstream communication node by rejecting a communication frame during communication and abandoning a transmitting right with a communication node having the transmitting right when the fault is generated on the transmission path. CONSTITUTION:The fault is generated on the transmission path between a communication nodes N1 and N2, when the communication node N2 has the transmission right at such a time point, a synchronization supervising part in the communication node N2 detects pull out, notifies it to a communication frame transmission control part, and from this, the communication frame during the transmission with communication equipment 20b is rejected. The pull out is successively detected in communication nodes N2 and N3 and in a supervisory node N0, and in the supervisory node N0, the frame initializing action is started. There, when the communication node N1 on the highest-stream is in the transmission right waiting condition, it obtains the transmission right, and starts the transmission. Since the communication node N2 stops the transmission, the transmission of the node N1 is not interfered.

Description

[Detailed Description of the Invention] [Summary of the Invention] Regarding a failure processing method for transmission right arbitration in a slotted ring type LAN, a communication node that has the transmission right also monitors input packets during communication to detect transmission abnormalities. is inserted into the transmission path of a slotted ring type LAN, with the aim of maintaining the transmission right arbitration function by determining that another communication node has started or will start transmission and thereby maintaining the transmission right arbitration function. Each communication node to which the communication device is connected has a synchronization monitoring unit that monitors the synchronization status of the transmission path, a storage unit that stores the ID number of the slot that the node uses as a token, and the usage status of the circulating token. When the synchronization monitoring section detects an out-of-synchronization, or the storage section and the token inspection section detect that a node has started using the token that is currently being used by its own node. , the transmission is stopped and the transmission of the node is prioritized.

[Industrial application field]

The present invention relates to a failure handling method for transmission right arbitration in a slotted ring type LAN.

Slotted Ring type L
Reserve any one slot in AN (Local Area Network) and use it as a token (
We have previously filed an application for a method of arbitrating transmission rights by using them as tokens (Japanese Patent Application No. 12391/1983, filed on January 22, 1986, Transmission Rights Arbitration Control Method in Throttling LAN with Multi-Frame Structure). The present invention relates to processing in the event of a fault in a LAN transmission line, and is intended to prevent duplicate tokens from being generated even when the transmission line is initialized due to the occurrence of a fault.

C. Prior Art] A slotted ring LAN is a LAN that transmits all data in the required number of fixed-length mini-packets. Note that in this specification, a slot refers to a container for transferring data, and a packet refers to data transferred by a slot. In a slotted ring type LAN, the length of the transmission frame in the ring (more specifically, the ring type transmission path) is always kept constant (memory etc. are interposed as necessary to keep the frame length, which is a temporal length, constant). ), there is always a fixed number of slots on the ring. An ID number is assigned to each slot.

FIG. 3(a) shows an outline of a slotted ring type LAN, where R is the ring of the LAN, 1, 2, 3 .・・・・・・
are the slots on the ring or their ID numbers. The first one of the slot touches the last one of the slot, creating a loop. A plurality of communication nodes N, N2, . communicate with. The format of the slot data (packet) is fbl in the same figure.
As shown in , the packet header PH, data DATA,
Consists of bucket and throat trailer PT, packet header PH
includes a control (CTL) bit indicating the nature of the data (busy/idle, etc.), a destination address DA, a source address SA, etc. 32 to 6 in the DATA section
Can store 4 bytes of data.

For example, when transmitting from communication node N1 to communication node N2 on the LAN ring, the destination address DA is N2,
The source address SA is N1, etc., and the data part DATA
Enter the data to be sent and send.

Communication node N2 receives this and takes in the destination address DA since it points to itself. If the DA does not point to itself, simply send the packet downstream.

Now, there are three or more communication nodes N,,N2゜N3 on the ring.
．． ..., the destination of each node is arbitrary (undefined), and considering that node N sends long data that does not fit in one slot to node N2, in such a case, the data The receiving node N2 receives these by dividing them into one slot each and transmitting them multiple times. However, when data is also sent from another node N3 to N2, node N2 receives the data from N1 and the data from N3. will be mixed, and processing will be required to sort them out. To avoid this, it is necessary to arbitrate transmission rights so that only one of the plurality of communication devices (three or more) accommodated in the slotted ring type LAN can transmit at the same time. A communication device that has acquired the transmission right transmits one communication frame (one weight of data) in one acquisition of the transmission right, and then
Transmission rights are delegated to the next communication device to equalize transmission opportunities.

This arbitration control can be realized in the upper layer of the communication protocol between communication devices, but in order to obtain high communication throughput, it is necessary to realize it between each communication node as a function of the slotted ring type LAN. good. The above-mentioned applications are based on this viewpoint, and the transmission right arbitration is realized by the following procedure.

A communication node that has received a transmission request from an affiliated communication device and whose transmission buffer is ready for transmission first starts acquiring tokens. In FIG. 3, the communication nodes N1 to N3 in ring R that may communicate with each other are notified in advance of one common slot number that may be used as a token, so that the sending node ! If so, cough N! checks the reception slot with the above number, and if it is empty, inserts specified data indicating that it is in use into the slot, assumes that it has the right to transmit, and sequentially sends the transmission data to the subsequent empty slots. insert. Also, when a token that has returned to an empty slot after one rotation (in the multi-frame method, an empty slot with the same number as the token slot) is captured, the preparation for sending the subsequent data may not be ready yet. If so, send dummy data and capture the token again so that subsequent data can be sent. When there is no more data to be transmitted (transmission ends), the token slot that has been rotated once is made vacant to return to the initial state.

In this slotted ring type LAN, a monitoring station is placed either separately from the communication nodes or in any communication node, and performs control to maintain the frame length constant. This control is performed by writing received data into memory and reading it at a timing that makes the frame length constant.

[Problem to be solved by the invention]

In this transmission right arbitration system, if a failure occurs in the transmission path during operation, all the slots issued by the monitoring node become empty and the transmission path is initialized.

Note that this slotted ring type LAN is a dual system, so even if a failure occurs, transmission can be continued using the healthy part. That is, as shown in FIG. 4, the monitoring station No. manages the network and controls the monitoring device SV #0. Normally, SV#O manages and monitors the network like (al), and if a failure occurs in SV#O in this state, the control authority is transferred to SV#1 as shown in ('b). It moves from #0 to SV#1 and returns to the operating state at SV#1 as shown in tel.

In addition, as shown in Figure 5, the transmission line is also doubled, and f
As shown in a), the #O transmission line was always operated, but as shown in ('b), the transmission line was changed to node N2. N3
If a failure occurs between them, the monitoring station NO switches from the #0 system loop to the #1 system loop, resulting in state (C). Also, as shown in Figure 6 fal, transmission lines #O and #1 are connected to node N.
2. If a failure occurs between N3, monitoring station N.

is N2. A loopback is commanded to N3, as shown in FIG. 4(b). Furthermore, each communication node is equipped with a repeater RP as shown in FIG. let

As mentioned above, when a failure occurs in a transmission line, etc., operation is switched to using a healthy part, but - at least temporarily, the input data to the monitoring station is lost.

Therefore, the monitoring station ends up initializing the transmission system. Therefore, the communication node that had the transmission right at the time of the failure continues to transmit, while the communication node that was waiting to acquire the transmission right and is upstream from the above communication node uses the free token issued by the monitoring station. When a slot is flashed, the node recognizes that it has acquired the right to transmit and starts transmitting, resulting in both nodes transmitting at the same time, causing a problem in which normal control cannot be continued.

When transmission right arbitration becomes abnormal, a recovery procedure is performed in the upper layer of the communication protocol between the communication devices, but this increases the time required to handle the failure and reduces the operational efficiency of the system.

The present invention aims to improve this point, and the communication node that has the transmission right also monitors input packets during communication, and if a transmission abnormality is detected, other communication nodes start transmission. The purpose of this method is to maintain the transmission right arbitration function by stopping the transmission when the transmission is determined to start.

[Means to solve the problem]

As shown in FIG. 1, in the present invention, the communication node Ni is provided with a synchronization monitoring section 11, a token checking section 12, and a storage section 13.

The synchronization monitoring unit 11 monitors the synchronization state (synchronization establishment/synchronization loss) of the transmission path by detecting the synchronization of data received while circulating around the ring transmission path R. That is, the sending node writes the destination address DA etc. in the packet header PH, puts the sending data in the data section DATA and sends it out, and the receiving node receives this, but it does not write information indicating that it has been received to the packet header. Just by writing, the data in the data section D^↑A is left as is and sent to the downstream node. The synchronization monitoring unit 11 of the sending node checks the synchronization flag pattern in the slot to recognize the synchronization pull-in state. If synchronization is detected when the self-communication node has the right to transmit, the synchronization monitoring unit instructs the communication frame transmission control unit 14 to stop transmission and discard the communication frame that is being transmitted. do.

The storage unit 13 stores the total loft number used by the self-communication node, and the token check unit 12 compares the number of the slot flowing through the transmission path R with the slot number stored in the storage unit 13 and determines whether If the slot is vacant, the communication frame transmission control unit 14 is notified that the transmission right can be acquired. Also, when the self-communicating node has obtained the transmission right (using its own node), a number other than its own node number is written in the token and slot, indicating that the node is using the cough totance loft. and instructs the communication frame transmission control unit 14 to stop transmission and discard the communication frame being transmitted, in the same way as above.

In the figure, 15 is a packet add/drop unit, 16 is a reception buffer, 17 is a transmission buffer, and 18 is a packet assembly/disassembly unit. transmission line R even when
The synchronization monitoring unit 11 and the token inspection unit 12 monitor the transmitted packets, and if an out-of-synchronization is detected or a token abnormality is detected, it is determined that the node has started transmission or will start transmission, and the communication frame is transmitted. Control unit 14 is made to stop transmission. This prevents multiple nodes from transmitting at the same time, and maintains the transmission right arbitration function even if a failure occurs in the transmission path.

When a transmission path failure occurs, all monitoring stations begin issuing empty slots, and communication nodes downstream of the monitoring station that are close to the monitoring station receive the empty slots, assume that they have the right to transmit, and start transmitting, but when a failure occurs A downstream communication node that previously obtained the transmission right realizes transmission right arbitration by delegating the transmission right to the upstream communication node.

The communication node that had the right to send and was transmitting when the failure occurred will lose the communication frame, but this will not affect other communication devices and normal communication will resume after the transmission path is restored. can be restarted.

Transmission/reception of the communication device 20 is performed in a known manner. That is, the communication node Ni takes in the transmission packet on the transmission path R,
If the data is addressed to the own node, the data in the data section DATA is stored in the reception buffer section 16, converted into a data format that can be passed through the communication device 2o by the packet assembly/disassembly section 18, and sent to the communication device 2o.

The transmission data of the communication device 20 is converted in the packet assembling and disassembling section 18 in the opposite manner to the above, and then stored in the transmission buffer section 17. When the transmission right is obtained, the transmission data is transferred to the packet branching and inserting section 15.
A packet header PH and a packet trailer PT are added to the packet, and the packet is sent to the transmission path R. This transmission control is performed by the communication frame transmission control section 14.

〔Example〕

An example of the operation of the fault handling method for transmission right arbitration of the present invention will be described with reference to FIG. As throughout the design, parts that are the same as in other figures are numbered the same.

First, let us assume that a failure occurs in the transmission path between communication nodes N1 and N2 (as in al), and communication node N2 had the right to transmit at this time. is detected and notified to the communication frame transmission control unit, whereby the communication frame being transmitted by the communication device 20b is discarded.

The communication node N2 is out of synchronization. N3, the frame is sequentially detected at the monitoring node No., and the frame initialization operation is started at the monitoring node No. Therefore, if the most upstream communication node N1 is in a state of waiting for the transmission right, it acquires the transmission right and starts transmission. Since the communication node N2 has stopped transmitting, there is no problem with the transmission of the node N.

Figure 2 (blO) In the example, it is assumed that a failure occurred in the transmission path between communication nodes N3 and N2, and that communication node N2 had the right to transmit at this time.A failure occurred in the transmission path, When synchronization occurs, this is detected in the order of communication node N3 and monitoring node N, and the monitoring node Na starts frame initialization operation.Then, the most upstream communication node N
1 is waiting to acquire the transmission right, it acquires the transmission right and starts transmission. In the communication node N2, the token inspection unit detects a token abnormality (not own node ID) and notifies the communication frame transmission control unit of this, thereby causing the communication device N
The communication frame No. 3 was being transmitted is discarded. As a result, the communication node N1 can transmit data without any trouble.

〔Effect of the invention〕

As explained above, according to the present invention, when a failure occurs on a transmission path, the communication node that had the right to transmit discards the communication frame in progress and relinquishes the right to send, so that the transmission path becomes By being initialized, transmission right arbitration can be performed in the event of a failure without interfering with the transmission of an upstream communication node that has newly acquired the transmission right. Further, it is not necessary for each communication device to perform a recovery operation when transmission right arbitration is disturbed, and the failure processing time of the entire system can be shortened.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of the principle of the present invention, Fig. 2 is a detailed explanatory diagram of the invention, Fig. 3 is an explanatory diagram of a slotted ring type LAN, and Figs. It is an explanatory diagram of an example of operation. In FIG. 1, R is a transmission path, 20 is a communication device, N1 is a communication node, 11 is a synchronization monitoring section, 13 is a storage section, and 12 is a token inspection section.

Claims (1)

[Claims] 1. A communication node (Ni
), a synchronization monitoring unit that monitors the synchronization state of the transmission path,
A storage unit that stores the ID number of the slot used by its own node as a token, and a token inspection unit that inspects the usage status of the circulating token are provided, and the synchronization monitoring unit detects out of synchronization, or the storage unit and When the token inspection unit detects that another node has started using the token that is currently being used by the own node,
A failure handling method in transmission right arbitration of a slotted ring type LAN, characterized by stopping transmission and giving priority to transmission from other nodes.