CN100370751C - Methods for broadcasting data over semi-permanent connections - Google Patents

Abstract

The invention relates to a method for broadcasting data through a semi-permanent connection. By applying for a record block for each semi-permanent connection and a machine frame where its endpoint is located, the information of the semi-permanent connection and the information of a machine frame where the endpoint of the semi-permanent connection is located are recorded. Networking information: assign a unique index number to each semi-permanent connection record block, record it in the endpoint information of the end point of the chassis where the semi-permanent connection is located, and send it to each chassis when it is connected to the network for the first time. When the endpoint of the connection fails, it is guaranteed to index to the corresponding record block according to the endpoint information, to ensure the correct release of the semi-permanent connection, to prevent the convergence of the endpoints, and to ensure the successful reconstruction of the network; at the same time, the endpoint code does not need to be rewritten when the network is rebuilt, and no message is required The confirmation mechanism reduces the communication volume between the chassis and improves the code reuse rate of the endpoint, which not only achieves the purpose of system expansion, but also flexibly realizes various single-chassis services.

Description

Methods for broadcasting data over semi-permanent connections

technical field

The present invention relates to a communication networking method, more specifically, relates to a method for broadcasting data from one end point to multiple end points by using a semi-permanent connection.

Background technique

Semi-permanent connection, that is, the two endpoints are always connected to the network. If abnormal conditions such as endpoint failure occur, it is required to rebuild the network in time when the abnormal situation recovers. Multicasting of semi-permanent connections, that is, broadcasting data from one endpoint to multiple endpoints through multiple semi-permanent connections. For example, two semi-permanent connections: A->B (forward connection from point A to point B) and A->D (forward connection from point A to point D), for point A, there are two semi-permanent connections, so that Data coming into point A is broadcast to points B and D respectively.

For services such as signaling monitoring, functions such as semi-permanent multicast are required to realize. For example, establish a signaling semi-permanent connection A<->B (two-way semi-permanent connection between A and B) on the switch. Point A is the relay time slot for connecting other devices. If you want to monitor other devices from point C to point A Signaling, another semi-permanent connection A->C needs to be established, so as to realize the function of monitoring signaling at point C through the signaling monitoring device.

Due to capacity expansion, sometimes multiple chassis need to be cascaded to expand capacity. However, due to reasons such as instability of inter-chassis communication, data in each chassis may be inconsistent, causing problems. For example, as shown in Figure 1, the forward network A->B between endpoint A of chassis 1 and endpoint B of chassis 2 is cascaded through 3 optical fibers, and endpoint C is selected as the intermediate time slot , C' indicates that the middle time slot is in the time slot of chassis 1, and C" means that the middle time slot is in the time slot of chassis 2. C' and C" are connected through optical fibers, through A->C', C"->B , realize the forward networking of A->B.

If chassis 1 detects a failure of endpoint A, it can disconnect the network between A and C', and then send a message to chassis 2 to disconnect the network between point B and C", but due to the The communication is not stable. If frame 2 does not receive the message of disconnection, it will not disconnect the connection between B and C". In this way, while waiting for the failure of endpoint A to be eliminated and rebuilt, the network connection message (C”->B) for connecting B and C” will be sent to chassis 2 again, and the endpoints will converge due to the network connection of point B, and return failure , the semi-permanent connection cannot be reestablished.

In the case of multiple chassis, due to the needs of services such as signaling monitoring, it is often necessary to implement multicast of semi-permanent connections. When the semi-permanent connection is reestablished, due to abnormal conditions such as unstable communication between the chassis, if the time slot is not released in time, the endpoints will converge and the reestablishment will fail.

For the semi-permanent connection, the most important thing is to ensure the successful re-establishment of the semi-permanent connection in a short period of time when the endpoint failure is eliminated in any abnormal situation. Ordinary single-chassis semi-permanent connection, if there is an abnormal situation such as an endpoint failure, the entire network can be removed, and the network can be rebuilt when the failure is eliminated.

For semi-permanent connections under multi-chassis conditions, if an abnormal situation occurs, the entire network can also be dismantled, and the network can be rebuilt when the endpoint fault is eliminated. For example, the above semi-permanent connection A->B. If there is no multicast, there may only be a semi-permanent connection between point A and point B. When point A fails and the network is rebuilt, if point B is connected to the network, the existing point B can be connected to the chassis 2. All the networks are dismantled and then rebuilt, so as to ensure that the reconstruction will not fail due to the aggregation of endpoints.

But for the case of multicast, there may be many semi-permanent connections (B->D, B->E) at point B. In this case, it is impossible to simply release all the connections of point B to ensure that no endpoint aggregation occurs. Because if all the connections of point B are released, some connections of other semi-permanent connections may be released by mistake, causing other semi-permanent connections to be disconnected.

Contents of the invention

The technical problem to be solved by the present invention is to provide a method for broadcasting data through a semi-permanent connection, which overcomes the defect in the prior art that the semi-permanent connection fails to be rebuilt due to the convergence of the endpoints after the endpoint fails.

For this reason, the present invention adopts the following technical solutions: a method for broadcasting data through a semi-permanent connection, utilizing a semi-permanent connection to perform data transmission in or between the machine frames, characterized in that the method includes the following steps:

S1. Apply for a record block corresponding to a semi-permanent connection and a chassis where the endpoint is located, and record the information of the semi-permanent connection and the networking information of a chassis where the endpoint is located;

S2. Assign a unique index number to each semi-permanently connected record block;

S3. Record the record block index number into the endpoint information of the end point of the chassis where the record block corresponds to the semi-permanent connection, so as to ensure that the corresponding record block is indexed according to the endpoint information;

S4. The record block index number is sent to each frame when the semi-permanent connection is established for the first time;

S5. When an endpoint of a semi-permanent connection fails, index to the corresponding record block according to the index number in the endpoint information, and send a network teardown message to the target endpoint respectively according to the semi-permanent connection information of the record block, and remove the corresponding semi-permanent connection;

S6. When an endpoint failure of a semi-permanent connection is eliminated, index to the corresponding record block according to the index number in the endpoint information, and send a networking message to the target endpoint respectively according to the semi-permanent connection information of the record block. The corresponding record block index number is carried in the network message, and the corresponding semi-permanent connection is rebuilt according to the record block information, and the network connection is restored.

The above-mentioned method for broadcasting data through a semi-permanent connection is characterized in that: the record block in the step S1 is stored in the memory on the corresponding chassis single board;

The above method of broadcasting data through a semi-permanent connection is characterized in that: the target endpoint in the step S5 and the faulty endpoint are in the same frame, directly send a network removal message to the frame, and disconnect the corresponding semi-permanent connection.

The above-mentioned method for broadcasting data through a semi-permanent connection is characterized in that: the target endpoint in the step S6 is in the same frame as the faulty endpoint, and when the fault of the faulty endpoint is eliminated, a network connection message is directly sent to the frame to rebuild the target endpoint A semi-permanent connection to this endpoint.

The above-mentioned method of broadcasting data through a semi-permanent connection is characterized in that: the step S6 also includes: indexing to the record block according to the record block index number, and performing verification according to the information in the record block, when the recorded in the record block When the semi-permanent connection is not released, release the semi-permanent connection first, and then rebuild the semi-permanent connection to ensure that the code corresponding to the semi-permanent connection can be reused.

The above-mentioned method for broadcasting data through a semi-permanent connection is characterized in that: the target endpoint and the faulty endpoint in the step S5 are located in two machine frames respectively, and when the faulty endpoint and the middle frame where the faulty endpoint is located and the frame where the target endpoint is located are removed The semi-permanent connection between the intermediate time slot and the target endpoint is sent to the machine frame where the target endpoint is located, and the semi-permanent connection between the intermediate time slot and the target endpoint is removed.

The above method of broadcasting data through a semi-permanent connection is characterized in that: the target endpoint and the faulty endpoint in the step S6 are respectively in two machine frames, and when the fault of the faulty endpoint is eliminated, the faulty endpoint and the intermediate time slot are connected Semi-permanent connection, and send a network connection message to the frame where the target endpoint is located. The message carries the corresponding record block index number, and checks according to the information in the record block. When the semi-permanent connection recorded in the record block has been released , to rebuild the semi-permanent connection; when the semi-permanent connection recorded in the record block has not been released, release the semi-permanent connection first, and then re-establish the semi-permanent connection to ensure that the code corresponding to the semi-permanent connection can be reused.

The above method of broadcasting data through a semi-permanent connection is characterized in that: when the user deletes the information of the semi-permanent connection through configuration, a machine frame where the endpoint of the semi-permanent connection is located sends a network teardown message to the machine frame corresponding to the semi-permanent connection, and the message carries the index number of the corresponding record block, and the corresponding frame performs verification according to the information in the record block;

If the semi-permanent connection still exists, remove the semi-permanent connection, delete the record block corresponding to the semi-permanent connection, and release the network;

If the semi-permanent connection does not exist, delete the record block corresponding to the semi-permanent connection, release the network, and ensure that the endpoint code will be reused in the network connection in the future; when the frame that has deleted the record block receives the network message, according to The record block index number carried in the message re-applies for the record block and rebuilds the network.

The beneficial effects of the present invention are:

1 Through indexing and verification of record blocks, it overcomes the defect of semi-permanent connection reconstruction failure caused by endpoint aggregation in the existing technology, and ensures that semi-permanent connection reconstruction succeeds after an endpoint fails;

2. This solution does not need to rewrite the networking of each part, but only needs to borrow the processing of ordinary networking, which greatly increases the degree of code reuse;

3 Release the network when the semi-permanent connection is released. Even if the release fails, there is no need for a message confirmation mechanism to guarantee. Instead, when connecting to the network again, check the network information according to the record block index in the message. If it will cause endpoint convergence, you can Release the network according to the network information in the record block, and will not affect other semi-permanent connections;

4 Since the message confirmation mechanism is not required to ensure the correct release of the network between the chassis, the communication between the chassis is reduced.

Description of drawings

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Figure 1 is a schematic diagram of the semi-permanent connection of two frames;

Figure 2 is a schematic diagram of the semi-permanent connection of three frames;

Figure 3 is a schematic diagram of another 3-frame semi-permanent connection.

Detailed ways

In Figure 1, frame 1 and frame 2 are cascaded through three optical fibers 102, and endpoint C is the middle time slot between frame 1 and frame 2, where C' indicates that the middle time slot is in the time slot of frame 1 , C" means that the middle time slot is in the time slot of the frame 2, C' and C" are connected through the optical fiber 102, the end point A of the frame 1 and the end point B of the frame 2 are semi-permanently connected A->C' and C"- >B, realize the forward networking of A->B.

If there is no semi-permanent multicast and there is only one semi-permanent at most at endpoint B, then all connections of endpoint B can be released directly to ensure the correctness of semi-permanent reconstruction. However, since there may be multiple connections at point B (such as B->D, B->E, not shown in the figure), all connections at point B cannot be simply released. If you send a message to release the network connection of C”->B, through the message confirmation mechanism, if the release is not successful, you can continue to release, which can also achieve the purpose of release, but if multiple messages fail, the chassis will definitely be added The message flow of inter-communication, but also need to increase the confirmation message of releasing the network, which increases the complexity of the code.

If point A of chassis 1 fails, the network connection of chassis 2 (C”->B) is not released. When the failure of point A recovers, it is feasible to only rebuild the network connection of chassis 1. However, it is impossible to reuse common The code for network connection, because of ordinary network connection, the input condition is the two endpoints (A and B) of the terminal, and it cannot be connected to the network according to the middle time slot (C' and C"). If it is only for A of chassis 1 To connect with C', it is necessary to re-implement the semi-permanent connection mechanism, which reduces the reusable code.

Therefore, in order to ensure the successful reconstruction of the semi-permanent network when the endpoint fails, before connecting to the network for the first time, apply for a record block corresponding to a semi-permanent connection and a chassis where its endpoint is located, and record the information of the semi-permanent connection and the semi-permanent connection. Networking information of the chassis where the connected endpoint resides.

The record block is a storage area opened up in the memory of each chassis single board, which is used to save the semi-permanent connection and the networking information of a chassis where its endpoint is located, such as the information of the source endpoint and destination endpoint of the semi-permanent connection and the network connection information. direction, and the result of the establishment of the semi-permanent connection (whether the semi-permanent network has been established, and if the establishment fails, record the reason for the failure) and other information.

Assign a unique index number to each semi-permanent connection record block, which is used to identify the semi-permanent connection information. It is a one-to-one correspondence, ensuring that the corresponding record block is indexed according to the endpoint information. When the two endpoints of the semi-permanent connection are in the same chassis, the semi-permanent connection corresponds to one record block; when the two endpoints of the semi-permanent connection are in two chassis, the semi-permanent connection corresponds to two record blocks, and the two record blocks They are respectively stored in the memory of the boards corresponding to the chassis.

The record block index number is sent to each chassis when the network connection is established for the first time. When the semi-permanent connection is restored, the record block is indexed according to the record block index number, so as to find the relevant semi-permanent connection information. For example, apply for a record block for the semi-permanent connection A->B and the chassis 1 where the endpoint A is located, record the semi-permanent connection A->B and its networking information in the chassis 1, and keep it in the memory of the single board of the chassis 1; Apply for a record block for the semi-permanent connection A->B and the chassis 2 where the endpoint B is located, record the semi-permanent connection A->B and its networking information in the chassis 2, and keep it in the memory of the chassis 2 board. Assign an index number 11 to the two record blocks of the semi-permanent connection A->B, and record the index number 11 of the record block into the end points of the chassis where the record block corresponds to the semi-permanent connection (that is, A, C', C", B end points) In the endpoint information, in chassis 1, according to the record block index number 11, you can find the semi-permanent connection A->C', and according to the endpoint A or C', you can find the record block index number 11, so as to index to the corresponding record block; Box 2, according to the record block index number 11, you can find the semi-permanent connection C”->B, and according to the endpoint B or C”, you can find the record block index number 11. When the semi-permanent connection A->B is established for the first time, send to the chassis 2 When sending the C”->B network connection message, at the same time send the record block index number 11 to the chassis 2, so that the chassis 2 will store the information of C”->B in the record block corresponding to the record block index number 11 , so that it can be queried when rebuilding in the future.

If there is a failure at point A, when the network connection of frame 1A->C' is removed, a network removal message will be sent to frame 2 where the target endpoint B is located, and the network connection of C”->B will be removed. When the fault at point A recovers , connect the network of A->C' of chassis 1, and send a network connection message (ie C"->B) to chassis 2 where the target endpoint B is located, and carry the record block index number 11 in the message. After receiving the above-mentioned network connection message (that is, C"->B), frame 2 indexes to the corresponding record block according to the record block index number 11 carried in the message, and performs verification according to the information of the record block. If the If the network (C”->B) recorded in the record block has been released, then directly re-establish the semi-permanent connection; otherwise, release the network (C”->B) first, and then re-establish the network.

If the user deletes the information of the semi-permanent connection A->B during system configuration, the frame 1 or frame 2 corresponding to the semi-permanent connection A->B needs to send the network removal to the other frame corresponding to A->B message, the message carries the index number of the corresponding record block, removes the connection between A->C' and C”->B, and deletes the record block corresponding to the semi-permanent connection in chassis 1 and chassis 2. If the message is sent If it fails, the connection between frame 1 and frame 2 will not be released. If C’ and point B are connected to the network in the future, it will also fail due to the aggregation of endpoints. Therefore, when frame 1 or frame 2 receives For the networking message of point C' or point B, if it will lead to the convergence of the endpoints, you can find the record block with the index number 11 according to C' and B, and check according to the information in the record block. If the semi-permanent connection If it no longer exists, the network will be released, and the corresponding record block will be deleted, so as to ensure the reuse of codes at points C' and B of the network in the future.

When the chassis that has deleted the record block receives the network connection message again, it re-applies for the record block according to the record block index number carried in the message, and re-establishes the network connection.

In Figure 2, frame 1 and frame 2 are cascaded through three optical fibers 102, and between frame 1 and frame 3 are cascaded through three optical fibers 103, and the end point F is the middle of frame 1 and frame 3 time slot, where F' indicates that the middle time slot is in the time slot of frame 1, and F" represents the time slot in which the middle time slot is in frame 3. F' and F" are connected through optical fiber 103, and endpoint A of frame 1 and the frame The endpoint E of box 3 realizes the forward networking of A->E through the semi-permanent connection of A->F' and F"->E.

In Figure 2, there are other connections at point A, A->D (points A and D are both in chassis 1), A->E (points A and E are in chassis 1 and chassis 3 respectively), Then apply for these two semi-permanent record blocks separately, and allocate the record block with index number 12 to save A->D and the networking information of the corresponding chassis, and the record block with index number 13 to save A->E and its Networking information of the corresponding chassis. According to the information of point A, a linked list is established, and all semi-permanent connection information related to point A can be indexed according to point A (semi-permanent connection information corresponding to record blocks with index numbers 11, 12, and 13).

If point A fails, according to point A, record blocks with index numbers 11, 12, and 13 are indexed in chassis 1, and these record blocks are processed separately: for record block with index number 11 (record A->B and The network connection information of the corresponding subrack), since C' is the time slot between subracks, it is necessary to send a message to subrack 2 to process the semi-permanent connection in subrack 2, and the method is as described above.

For the record block with the index number 12 (record A->D and the network connection information of the corresponding chassis), since point D is the endpoint of the local chassis, it is only necessary to send a message to the local chassis (chassis 1) to disconnect Open the A->D connection.

For the record block with index number 13 (record the connection information of A->E and its corresponding chassis), the processing method is the same as A->B. Point F is the selected inter-frame time slot, then in frame 1, according to the record block with index number 13, you can find the semi-permanent connection A->F'; in frame 3, according to the record block with index number 13, You can find the semi-permanent connection F”->E. If there is a failure at point A, when the network connection of frame 1A->F’ is removed, a network removal message is sent to frame 3. The message carries the record block index number 13, and the removal F”->E’s networking. When the fault at point A is eliminated, connect the network connection of frame 1A->F', and send a network connection message (F”->E) to frame 3 at the same time, and carry the record block index number 13 in the message. If the frame 3 Received the message of dismantling the network and dismantled the F"->E network in time, then when the network is rebuilt, there will be no endpoint convergence. However, if chassis 3 does not receive the message of dismantling the network connection, it will return a failure when connecting to the network. Therefore, when establishing the network connection in chassis 3, it is necessary to find the corresponding record block according to The network connection (F”->E) recorded in has not been released, you should release the network connection first, and then connect to the network.

In Figure 3, frame 1 and frame 2 are cascaded through three optical fibers 102, and frame 2 and frame 4 are cascaded through three optical fibers 204, and endpoint I is the middle of frame 2 and frame 4 time slot, where I' indicates that the middle time slot is in the time slot of frame 2, and I" represents the time slot in which the middle time slot is in frame 4. I' and I" are connected by optical fiber 204, and the end point B of frame 2 and the frame The endpoint H of box 4 realizes the forward networking of B->H through semi-permanent connection B->I' and I"->H.

If there is a failure at point B, handle it in the same way as above. The network connection status is A->B, B->G, B->H, point G and point B are both in chassis 2, points H and point B are in chassis 4 and chassis 2 respectively. The index numbers of the semi-permanent connection record blocks are 11, 14, and 15 respectively. The record block with the index number 14 stores B->G and the networking information of the corresponding chassis, and the record block with the index number 15 stores B->H and The connection information of the corresponding chassis. When point B fails, according to the information of point B, record blocks with index numbers 11, 14, and 15 are indexed in chassis 2, and these record blocks are processed respectively.

For the record block with the index number 11 (recording the network connection information of A->B and its corresponding frame), when the network connection of frame 2C”->B is removed, a network removal message is sent to frame 1, the message Carry the record block index number 11 in the record block, and remove the network connection of A->C'. When the fault at point B recovers, connect the network connection of C"->B of frame 2, and send a network connection message to frame 1 at the same time ( A->C'), carry the record block index number 11 in the message. If chassis 1 receives the network teardown message and removes the A->C’ network in time, then when the network is rebuilt, there will be no aggregation of endpoints. However, if chassis 1 does not receive the message of dismantling the network connection, it will return failure when connecting to the network. Therefore, when chassis 1 reconnects to the network, it needs to search for the corresponding record block according to the record block index number 11. If the record block The network connection (A->C') recorded in has not been released, you should release the network connection first, and then connect to the network.

For the record block with the index number 14 (record B->G and the network connection information of the corresponding frame), since point G is the endpoint of the frame, it is only necessary to send a message to the frame (frame 2) to disconnect the frame. Chassis networking.

For the record block with the index number 15 (recording B->H and the networking information of its corresponding chassis), the processing method is the same as A->B. Point I is the selected time slot between frames, then in frame 2, according to the record block whose index number is 15, the semi-permanent connection B->I' can be found; in frame 4, according to the record block whose index number is 15, You can find the semi-permanent connection I”->H. If there is a failure at point B, when the network connection of frame 2B->I’ is removed, a message is sent to frame 4, and the message carries the record block index number 15, and I” is removed -> H's networking. When the fault at point B is eliminated, connect the network connection of B->I' of frame 2, and send a network connection message (I”->H) to frame 4 at the same time, and carry the record block index number 15 in the message. If Frame 4 has received the message of dismantling the network and dismantled the I”->H network in time, so when the network is rebuilt, there will be no aggregation of endpoints. However, if frame 4 does not receive the message of dismantling the network connection, it will return a failure when connecting to the network. The network connection (I”->H) recorded in has not been released, you should release the network connection first, and then connect to the network.

The method proposed by the present invention is not only applicable to two chassis, but also can be smoothly expanded to a multi-chassis situation of more than three chassis, and is also applicable to various multi-chassis management methods, such as the centralized management mode of the central chassis, etc. . For example, three chassis are cascaded, and chassis 0 is used as the central switching chassis, and chassis 1 and 2 are service chassis, then you can send messages to chassis 1 and 2 through chassis 0 to control the semi-permanent The case of connection reestablishment.

Adopting the technical scheme of the present invention can bring the following beneficial effects:

1. By indexing and verifying record blocks, it overcomes the defect of semi-permanent connection reconstruction failure caused by endpoint convergence in the existing technology, and ensures that semi-permanent connection reconstruction succeeds after an endpoint failure;

2. This solution does not need to rewrite the networking of each part, it only needs to borrow the processing of ordinary networking, which greatly increases the intensity of code reuse;

3. When the semi-permanent connection is released, the network connection is released. Even if the release fails, the message confirmation mechanism is not required. Instead, when the network is connected again, the network information is checked according to the record block index in the message. If it will cause endpoint convergence, The connection can be released according to the connection information in the record block, and will not affect other semi-permanent connections;

4. Since the message confirmation mechanism is not required to ensure the correct release of the inter-frame networking, the inter-frame communication is reduced.

This solution can realize signaling monitoring and other telecommunication services under the conditions of single-chassis and multi-chassis. It not only achieves the purpose of capacity expansion, but also can flexibly realize various single-chassis services and ensure the correct reconstruction and release of semi-permanent multicast.

Claims (8)

1. A method for broadcasting data by semi-permanent connection, utilizes semi-permanent connection to carry out data transmission in machine frame or between machine frames, it is characterized in that, described method comprises the steps: S1. Apply for a record block corresponding to a semi-permanent connection and a chassis where the endpoint is located, and record the information of the semi-permanent connection and the networking information of a chassis where the endpoint is located; S2. Assign a unique index number to each semi-permanently connected record block; S3. Record the record block index number into the endpoint information of the end point of the chassis where the record block corresponds to the semi-permanent connection, so as to ensure that the corresponding record block is indexed according to the endpoint information; S4. The record block index number is sent to each frame when the semi-permanent connection is established for the first time; S5. When an endpoint of a semi-permanent connection fails, index the corresponding record block according to the index number in the endpoint information, and then send a network teardown message to the target endpoint respectively according to the semi-permanent connection information of the record block, and remove corresponding semi-permanent connections; S6. When the terminal failure of a semi-permanent connection is eliminated, index to the corresponding record block according to the index number in the terminal information, and send a networking message to the target terminal respectively according to the semi-permanent connection information of the record block, in the network connection The corresponding record block index number is carried in the message, and the corresponding semi-permanent connection is rebuilt according to the record block information, and the network connection is restored. 2. The method for broadcasting data through a semi-permanent connection according to claim 1, characterized in that: the recording block in the step S1 is stored in the memory on the corresponding chassis board. 3. The method for broadcasting data through a semi-permanent connection according to claim 1 or 2, characterized in that: the target end point in the step S5 is in the same machine frame as the fault end point, and directly sends a network removal message to the machine frame, and Disconnect the corresponding semi-permanent connection. 4. The method for broadcasting data through a semi-permanent connection according to claim 3, characterized in that: the target endpoint in the step S6 is in the same machine frame as the faulty endpoint, and when the fault of the faulty endpoint is eliminated, directly send the data to the machine frame Send a networking message to re-establish the semi-permanent connection between the target endpoint and the endpoint. 5. The method for broadcasting data through a semi-permanent connection according to claim 4, wherein the step S6 further comprises: indexing to a record block according to the record block index number, and performing calibration according to the information in the record block Check, when the semi-permanent connection recorded in the record block has not been released, release the semi-permanent connection first, and then re-establish the semi-permanent connection to ensure that the code corresponding to the semi-permanent connection can be reused. 6. The method for broadcasting data through a semi-permanent connection according to claim 1 or 2, characterized in that: the target endpoint and the faulty endpoint in the step S5 are located in two machine frames respectively, and the faulty endpoint and the machine where the faulty endpoint is located are removed The subrack is semi-permanently connected to the middle time slot of the target endpoint, and sends a network disconnection message to the target endpoint to remove the semi-permanent connection between the middle time slot and the target endpoint. 7. The method for broadcasting data through a semi-permanent connection according to claim 6, characterized in that: the target endpoint and the faulty endpoint in the step S6 are respectively in two machine frames, and when the fault of the faulty endpoint is eliminated, the faulty endpoint is connected The endpoint is semi-permanently connected to the intermediate time slot, and sends a network connection message to the frame where the target endpoint is located. The message carries the corresponding record block index number, and the information in the record block is verified. When the record block records When the semi-permanent connection has been released, rebuild the semi-permanent connection; when the semi-permanent connection recorded in the record block has not been released, release the semi-permanent connection first, and then rebuild the semi-permanent connection to ensure that the code corresponding to the semi-permanent connection can be reused. 8. The method for broadcasting data through a semi-permanent connection according to claim 7, characterized in that: when the user deletes the information of the semi-permanent connection through configuration, a machine frame where the endpoint of the semi-permanent connection is located sends a message to the machine corresponding to the semi-permanent connection The box sends a network teardown message, and the message carries the index number of the corresponding record block, and the corresponding machine box performs verification according to the information in the record block; If the semi-permanent connection still exists, remove the semi-permanent connection, delete the record block corresponding to the semi-permanent connection, and release the network; If the semi-permanent connection does not exist, delete the record block corresponding to the semi-permanent connection, release the network, and ensure that the endpoint code will be reused in the network connection in the future; when the frame that has deleted the record block receives the network message, according to The record block index number carried in the message re-applies for the record block and rebuilds the network.

Images