CN1319320C - Method for realizing link state feed-through in network - Google Patents

Abstract

The invention discloses a method for realizing link status break-through in a network. In order to notify the opposite end when a link fails, the method includes the following steps: a service device detects whether a link in the network fails; When a fault occurs on the link, the service device uses the control frame to transmit the link failure information to the peer service device; the peer service device disconnects the user link in response to the received control frame containing the link fault information and performs corresponding processing; and the user at the fault end communicates with the user at the opposite end through the backup network. establish communication between the users over the link. The present invention can notify the opposite end when the link fails and can ensure that the extra cost is as small as possible when the network works normally.

Description

A Method for Realizing Link State Breakthrough in Network

technical field

The invention relates to network communication, in particular to a method for realizing pass-through of access point link state and service link state on a point-to-point access network.

technical background

In engineering applications, data between users enters the network through access devices to achieve data interaction. FIG. 1 is a schematic diagram of data interaction between users through a point-to-point access network. As shown in FIG. 1 , under normal circumstances, user 1 and user 2 establish communication through a service device and a link, and the link mainly includes an access point link 1 , a service network, and an access point link 2 . When a link fails, that is, when the access point link or the service network fails, the communication between user 1 and user 2 is interrupted. For important users, a backup network needs to be activated when such a failure occurs to ensure normal transmission of important data. The backup network can take various forms, such as renting one more link, or setting up wireless communication or satellite communication between user 1 and user 2, etc. When the access point link fails, the user can detect the link failure, start the backup network, and try to use the backup network to communicate with the peer user. However, for the opposite end user, it does not know that a link failure has occurred, and still uses the original network for communication; as a result, the communication between the two users is disconnected. When the service network fails, the users at both ends do not know that the network cannot be used, so the communication at both ends is disconnected. If the state of the link can be transmitted to the peer user, that is, the pass-through function of the link state is realized, then the peer user can start the backup network and re-establish the link with user 1 or user 2 on the backup network.

In order to let the remote user know that the link has failed, various methods can be adopted. A typical method is: user 1 and user 2 tell the other party the link information of the local end through regular communication. When the link information sent by the opposite end is not received, it is considered that the link has failed, the user stops reporting the link information, and automatically uses the backup network for communication; another typical method is: user 1 and user 2 detect the status of the backup network in real time In terms of usage, as long as any party is detected to use the backup network, the other party will also use the backup network instead.

Although the above method can solve the problem to a certain extent, there are some disadvantages: first, under normal circumstances, the regular reporting link information will occupy the network bandwidth, which requires additional network overhead; second, when the network is congested , there will be a delay in reporting link information, which may lead to wrong operations due to wrong judgments; third, if the backup network is detected, it needs to occupy part of the bandwidth of the backup network, which will also bring additional overhead to the network; 4. When the link returns to normal, how both parties leave the backup network and return to the working network is also a problem.

Contents of the invention

In order to solve the above-mentioned drawbacks in the prior art, the purpose of the present invention is to propose a link in the network that can notify the opposite end in the event of a link failure and ensure that the network works normally with as little extra overhead as possible. State passthrough method.

In order to achieve the above-mentioned purpose of the present invention, a method for realizing link status pass-through in the network is provided. The method includes the following steps: the service device detects whether the link in the network fails; when the link in the network fails, the service device Use the control frame to transmit the link failure information to the peer service device; the peer service device disconnects the user link in response to the received control frame containing the link fault information and performs corresponding processing; and the faulty end The user communicates with the peer user through the backup network.

The method according to the present invention also includes the following steps:

When the link returns to normal, the service device at the recovery end sends a control frame containing link failure recovery information to the service device at the opposite end, and re-establishes the communication between the users on the link.

The control frame includes at least network status information, which is used to indicate whether a link failure occurs. The control frame can be a standard Ethernet frame, and the network state information is filled in the operation code field of the Ethernet frame, or an encapsulation protocol control frame, and the network state information is filled in the encapsulation protocol control frame In the control field, or the signal label byte, the network status information is filled in the signal label byte.

Step A also includes the following steps of the debounce process:

Only when the link continues to be unavailable for a predetermined time, it is judged that the link is faulty; and only when the link is continuously available for a predetermined time, it is judged that the link returns to normal.

If the link failure is an access point failure, the service device at the faulty end will send the control frame containing the failure information of the access point to the opposite service device, and after receiving the control frame containing the failure information of the access point After the control frame, the opposite end service device will disconnect the connection with the opposite end user, and then the user at the faulty end communicates with the opposite end user through the backup network;

If the link failure is a failure of the service network, the service device at the faulty end will send the control frame containing the failure information of the service network to the opposite service device, and disconnect the link with the user at the faulty end. After the control frame of failure information appears on the network, the peer service device will disconnect from the peer user, and then the user at the fault end communicates with the peer user through the backup network.

In the case that the link failure is a service network failure, the processing steps corresponding to the service network failure are executed only after it is determined that the protection function of the service network itself fails.

The beneficial effects of the present invention are: 1) when the link is working normally, no control frame will be sent for control, so that the bandwidth of the working network and the backup network will not be occupied, thereby saving extra overhead; 2) by using the control frame mechanism to control the disconnection and recovery of the link, which will not produce side effects on the original network, thereby facilitating the upgrading and maintenance of the network; 3) in the present invention, "debounce" is adopted for the judgment of link failure and recovery processing, thereby improving the reliability and ease of use of the network.

Description of drawings

Figure 1 is a schematic diagram of data interaction between users through a point-to-point access network;

Fig. 2 is a block flow diagram of the method for passing through the link state according to the present invention;

Fig. 3 is a link state traversal process diagram in standard mode according to an embodiment of the present invention;

FIG. 4 is a process diagram of link state traversal in extended mode according to an embodiment of the present invention;

FIG. 5 is a timing diagram of debounce processing according to an embodiment of the present invention;

FIG. 6 is an illustration of reporting alarms and events according to an embodiment of the present invention;

FIG. 7 is a state transition diagram for realizing link pass-through according to an embodiment of the present invention;

Fig. 8 is a schematic diagram of composition of a control frame according to an embodiment of the present invention;

FIG. 9 is a network structure applicable to the link pass-through method of the present invention;

Fig. 10 is another network structure applicable to the link pass-through method of the present invention.

Detailed ways

Fig. 2 is a block diagram of the overall technical flow of the link state pass-through method according to the present invention. As shown in Figure 2, the service device first detects whether a link in the network is faulty S1; when a link fault occurs S2, the service device at the faulty end uses the control frame to transmit the link fault information to the peer service device through the service network S3, after receiving the control frame from the service device, the service device at the opposite end will perform corresponding processing S4, and then the user at the faulty end communicates with the user at the opposite end through the backup network S5.

The control frame can be defined in different ways, three implementation ways are introduced below.

(1) Use the standard Ethernet frame to define the control frame, and realize the link state pass-through function. In the IP OVER SDH (IP based on SDH) network, the user is an Ethernet switch, and the service device is an SDH Ethernet access device. First define the control frame as shown in Figure 8 according to the Ethernet frame format, the control frame includes the following content:

Destination address (6 bytes): globally unique media access control (MAC) address;

Source address (6 bytes): globally unique MAC address;

Type field (2 bytes): 0xE140-indicates that the frame is a link state passthrough function control frame;

Opcode field (2 bytes):

0x0000-network fault elimination (Non_Broken message),

0x0200-Network failure (Broken message);

Padding field (44 bytes): fill in padding bytes;

Check field (4 bytes): fill in the cyclic redundancy Crc32 check value of Ethernet.

Secondly, several time parameters are defined: T0=500 milliseconds, T1=1-5 seconds, T2=5-10 seconds, T3=20 milliseconds-1 second, T4=30 seconds-300 seconds can be referred to.

After defining the control frame format and time parameters, the service device detects the working status of the service network and the link status of the access point link, and controls according to the above description to realize the link status passthrough function.

Considering that in network maintenance, it is sometimes necessary to distinguish fault points to facilitate network maintenance; at this time, the opcode field should be expanded. The expanded opcode is defined as:

0x0000-The access point fault is eliminated (Non_Broken message),

0x0100-the service network fault is eliminated (Non_Broken message),

0x0200-access point failure (Broken message),

0x0300-service network failure (Broken message);

When the failure end detects an access point failure, it sends an access point failure message; when it detects a service network failure, it sends a service network failure message. Devices can choose whether to support extended mode opcodes.

(2), using the encapsulation protocol control frame to realize the link state pass-through function

When the encapsulation protocol is used to implement the link state passthrough function, the control field of the encapsulation protocol can be used to define the control frame. If Link Access Procedure-SDHLAPS/P2P Protocol/High-level Data Link Control HDLC encapsulation protocol is used, the control field can be defined as follows:

0x70-network failure elimination (Non_Broken message), 0x72-network failure (Broken message);

Normally, the control field is filled with the default control value. When the link fails, the faulty end device fills in the control field with the control value with the above special meaning, which is used to notify the peer device of the link status at this time. When the peer device receives a control value representing a link failure, it processes it according to the control value. When the link is restored, the device resumes sending normal control values.

Similarly, the device can choose to use the extended control value, and the extended control field is defined as follows:

0x70-access point failure elimination (Non_Broken message), 0x71-service network failure elimination (Non_Broken message), 0x72-access point failure (Broken message), 0x73-service network failure (Broken message);

(3) In SDH, the signal label byte is used to realize the link state pass-through function

The following is the specific definition of the signal label:

0x70-network failure elimination (Non_Broken message), 0x72-network failure (Broken message);

The expanded definition:

0x70-access point failure (Non_Broken message), 0x71-service network failure (Non_Broken message), 0x72-access point failure (Broken message), 0x73-5 service network failure (Broken message) ;

Signal labels in SDH are transmitted using C2/V5 bytes in channel overhead.

For VC3/VC4-level services, under normal circumstances, the C2 byte transmits a normal signal label value. When a link fails, the fault detection end device sends a link fault C2 value to inform the peer end device of the link status at this time. After the peer device receives the C2 byte representing the link failure, it should not report the signal mismatch alarm, but process the link state passthrough function; and perform state transition according to the state machine of the link state passthrough function. When the link at the faulty end recovers and the device returns to the normal working state, the device resends the normal C2 value.

For VC1/2-level services, V5 bytes are used to transmit link state passthrough function control frames, and the working mechanism is consistent with VC3/VC4-level services. It should be noted that: under normal circumstances, the 5-7 bits of the V5 byte transmit normal signal labels. Since the link state passthrough function is added, the V5 byte needs to be extended. According to the standard, when extended, fill in 101 in the 5~7 bits of the V5 byte, and fill in the bit1 of the K4 byte for the extended signal label. The bit1 of the 32-frame K4 constitutes a 32-bit bit string, and use 12~19 bits to transmit the extended signal label.

1. Link passthrough in access point failure mode

Fig. 3 is a link state traversal process diagram in the standard mode according to an embodiment of the present invention. As shown in Fig. 3, the processing process is as follows:

At time t0, the link between user 1 and service device A fails. At this time, service device A sends continuous Broken primitives to service device B at the opposite end, and tells service device B the fault of the local end. The primitives use the control frame format.

At time t1, service device B receives the Broken primitive of the link failure; after processing, it knows that there is a link failure at service device A, so it disconnects the link with user 2; the link status of user 1 and service device A The connection state of the link between service device B and user 2 is reached. At this point, user 1 and user 2 can use the backup network to communicate.

At time t2, the link between user 1 and service device A is restored; at this time, service device A sends a Non_Broken primitive to service device B, telling device B that the local link is restored.

At time t3, service device B receives the Non_Broken primitive; after processing, it knows that the link failure at device A has disappeared; then re-establishes a link with user 2 on the original network.

2. Link state passthrough in service network failure mode

Fig. 4 is a process diagram of link state traversal in the extended mode according to an embodiment of the present invention. As shown in Fig. 4, the processing process is as follows:

At time t0, service device A detects that the service network is unavailable, sends a Broken primitive to the peer end, and cuts off the link with the local user;

At time t1, service device B receives the Broken primitive and cuts off the connection with user 2; after that, user 1 and user 2 use the backup network to communicate until the service network is restored.

It should be noted that the service network usually has its own network protection function and has certain network protection capabilities. In order to avoid confusion, the link state passthrough function is required to be started late, that is, the link state passthrough function is started only after it is determined that the protection function of the serving network is invalid.

3. Debounce processing of link state pass-through function

When the link status pass-through function is implemented, there may be erroneous operation and processing due to the ups and downs (jitter) of the link. In order to eliminate the jitter problem of the link, dejitter processing can be used. FIG. 5 is a timing diagram of debounce processing according to an embodiment of the present invention. As shown in FIG. 5 , the main process is:

Only when the link is unavailable for T0 time, the link is judged to be faulty; if the link failure time is less than T0, then the original link status remains unchanged;

Only when the link is available for T0 time, can it be judged that the link is working normally.

4. Alarms and events under the link status passthrough function

During link state traversal, various conditions need to be monitored and reported. Figure 6 shows the alarms and events under the link state passthrough function, as shown in Figure 6, its main content is:

I. When the service device detects the failure of the access point link, it reports the local access point failure alarm, and only stops reporting when it detects that the access point link failure disappears;

II. When the service device receives the Broken control frame for the first time, it reports the peer link failure event, and when it receives the Non_Broken control frame for the first time, it reports the peer link fault elimination event;

III. When the service device detects a fault in the service network, it can use the original alarm of the service network to report, or use a special alarm to report.

5. State machine conversion of the link state pass-through function

There are multiple states under the link state pass-through function, and transitions between these states can occur as the link situation changes. Various states and their transitions under the link state passthrough function will be described in detail below with reference to FIG. 7 . Fig. 7 is a state transition diagram for realizing link pass-through according to an embodiment of the present invention. As shown in Fig. 7, the specific realization process of the link state pass-through function consists of a state machine including 4 basic states. The following is the state machine The 4 basic state definitions of :

Normal working state: both the access point and the service network are in normal state;

Waiting for recovery state: When the service device receives the Broken control frame, it enters the state of waiting for recovery;

Access point failure state: the path between the user and the service device is faulty, and normal communication between the user and the service device cannot be carried out;

Service network failure status: The service network between service devices fails, and normal communication cannot be performed.

The state transition process can be summarized as follows:

◆Normal working status:

1). If it is detected that the service network is unavailable for T0 time, then send a Broken control frame to the peer device to tell the peer device that the service network is detected to be unavailable, and then enter the service network failure state;

2). If it is detected that the access point network is unavailable for T1 time, then send a Broken control frame to the peer device to tell the peer device that the local end detects that the access point network is unavailable, and then enter the access point failure state;

3). If the Broken control frame is received, indicating that the peer network is faulty, then the local service device cuts off the link port with the user, and then enters the waiting recovery state;

◆Waiting for recovery state: If the device enters this state, it means that it has received a Broken control frame. At this time, the access point network of the local device may be good, and a tentative link to the access point network should not be performed.

1). If the device detects that the service network continues to be unavailable at T0, then send a Broken control frame to the peer device to tell the peer device that the service network is unavailable at the local end, and then enter the service network failure state; this step is optional ;

2). The device disconnects the link with the user every T1 interval to prevent the accident that the service device does not disconnect the link with the user; if it can be guaranteed that the access point link between the device and the user is cut off before entering the waiting recovery state, Then this step is also optional;

3). If the device does not receive the Broken message for T2, it may be that the peer end has entered the normal working state or is waiting for recovery; the local device should link to the user's access point link and actively return to normal work state to prevent deadlocks;

4). If the device receives the Non_Broken control frame, it means that the peer link has recovered and can work normally now; then the device should link with the user's access point link, and then enter the normal working state;

5). If the device receives the Broken message, it means that the link of the peer device is still in a fault state, and restarts T2 counting;

◆Service network failure state: The device enters this state, indicating that the service network has detected a failure.

1). Continue to check whether the service network is normal. If the service network is still faulty, then periodically send a Broken message at T3 to tell the peer device that the service network is faulty (whether the peer device can receive it or not);

2). Continue to check whether the service network is normal. If the service network continues to have no faults for T0, then it is considered that the service network has returned to normal; at this time, a Non_Broken message is sent to the peer, and it returns to the normal working state directly, and a new detection is started . .

◆Access point failure state: The device enters this state, which means that the local access point is detected to be faulty. In this state, regular attempts are made to restore the link to the access point.

1). Detect the state of the service network at the local end. If it is detected that the service network is unavailable, then send a Broken message and enter the service network failure state; this step is optional;

2). T1 and T4 cycles work alternately; during T1 (T1 should at least ensure the required time between the normal link service device and the user), the device tries to link the access point link, and if the link is successful, it sends a Non_Broken control frame to The opposite end, and then enter the normal working state; otherwise, cut off the access point link again and enter the T4 stage; in the T4 stage, the service device sends a Broken message to the opposite end at regular intervals T3, telling the opposite end that a network failure has occurred at the end; T4 After the stage ends, the equipment re-enters the T1 stage;

3). If a Broken control frame is received, it means that the service network or the peer access point is faulty, and the T4 stage counting is restarted;

4). If a Non_Broken message is received, it means that the peer network has been restored; at this time, the local device should immediately enter the T1 stage; try to restore the network link.

Although the above descriptions of the present invention are made with reference to specific embodiments, those skilled in the art should understand that these descriptions are not meant to limit the present invention. For example, in the foregoing embodiments, the link state passthrough technology is applied to the network structure shown in FIG. 1 . However, this technology is also applicable to other forms of network structures, and Fig. 9 and Fig. 10 show two forms of network structures in which the link pass-through technology of the present invention can be used. As shown in Figure 9, the first network structure is: users 1 and 2 can activate the services of two ports at the service device, and the two ports can be active and standby, or they can be active and standby each other. At this point, both ports can use the link state passthrough function. However, when the service network fails, the communication between user 1 and user 2 will still be disconnected. As shown in Figure 10, the second network structure is: the working link and the backup link do not go through the same network, and are completely independent. User 1 activates services on service equipment A0 and service equipment A1 respectively; similarly, user 2 activates services on service equipment B0 and service equipment B1. Under normal circumstances, user 1 and user 2 use service network 0 to communicate, and when a fault occurs, user 1 and user 2 use service network 1 to communicate. This structure can ensure that the communication between users can still be carried out when the service network 0 fails. Therefore, the limitation of the present invention is defined by the claims, and any modification and transformation of the present invention will not depart from the spirit and scope of the present invention.

Claims (13)

1, a kind of method that realizes Link State Pass Through in the network is characterized in that may further comprise the steps:

The A service equipment detects whether link breaks down in the network;

B is when link breaks down in the network, and service equipment utilizes control frame that link failure information is passed to the opposite end service equipment;

The described control frame that contains link failure information that service equipment response in C opposite end receives and disconnect user link are also handled accordingly;

The user of D fault end and the user of opposite end communicate by backup network.

2, the method for Link State Pass Through in the realization network according to claim 1 is characterized in that further comprising the steps of:

E works as described link-recovery just often, and the reset terminal service equipment sends the control frame that contains link failure recovery information to the opposite end service equipment, again in the communication of setting up on the described link between the described user.

3, the method for Link State Pass Through in the realization network according to claim 1 is characterized in that comprising network state information in the described control frame at least, is used to show link failure whether occurs.

4, the method for Link State Pass Through in the realization network according to claim 3 is characterized in that described control frame is a standard ethernet frame, and described network state information is inserted in the opcode field of described ethernet frame.

5, the method for Link State Pass Through in the realization network according to claim 3 is characterized in that described control frame is the tunneling control frame, and described network state information is inserted in the control field of described tunneling control frame.

6, the method for Link State Pass Through in the realization network according to claim 3 is characterized in that described control frame is a signal label byte, and described network state information is inserted in the described signal label byte.

7, according to the method for Link State Pass Through in the described realization network of arbitrary claim in the claim 1,2 or 3, it is characterized in that steps A is further comprising the steps of:

Only when link is lasting unavailable in the given time, judge that just link breaks down;

But only, judge that just link-recovery is normal when the lasting in the given time time spent of link.

8, the method for Link State Pass Through in the realization network according to claim 7 is characterized in that:

If described link failure is the access point fault, then the service equipment of fault end will send to the opposite end service equipment and contain the break down described control frame of information of access point, contain access point and break down behind the described control frame of information receiving, described opposite end service equipment will disconnect with to being connected of end subscriber, then the user of fault end with end subscriber is communicated by backup network;

If described link failure is the service network fault, then the service equipment of fault end will send to the opposite end service equipment and contain the break down described control frame of information of service network, and disconnection links with the fault end subscriber, contain service network and break down behind the described control frame of information receiving, described opposite end service equipment will disconnect with to being connected of end subscriber, then the user of fault end with end subscriber is communicated by backup network.

9, the method for Link State Pass Through in the realization network according to claim 8; it is characterized in that; at link failure is under the situation of service network fault, only after the defencive function of having determined described service network self lost efficacy, carries out and the corresponding treatment step of service network fault again.

10, the method for realization Link State Pass Through according to claim 8 is characterized in that, also comprises the following steps of regularly attempting recovering the access point link when link failure is the access point fault:

In the first predetermined period stage, service equipment attempts recovering the access point link, if link successfully control frame that then transmission contains link failure recovery information to the opposite end service equipment, enter normal operating conditions then, otherwise disconnection access point link entered for the second predetermined period stage;

In the second predetermined period stage, service equipment regularly sends to the opposite end service equipment and contains the break down control frame of information of link, enters for the first predetermined period stage then.

11, the method for Link State Pass Through in the realization network according to claim 1 is characterized in that step B is further comprising the steps of:

When service equipment detects the access point link failure, report the fault warning of local terminal access point, stop to report when the access point link failure disappears when detecting;

When service equipment detects the service network fault, use original alarm of service network or special alarm report service network fault warning.

12, the method for Link State Pass Through in the realization network according to claim 1 is characterized in that step C is further comprising the steps of:

When service equipment receives that containing link breaks down the control frame of information, report opposite end link failure incident,, then enter the Wait-to-Restore state if local terminal access point link is normal, if local terminal access point link failure, then regularly send and contain the break down control frame of information of link to the opposite end service equipment;

When service equipment is received the control frame that contains link failure recovery information, report the opposite end link failure to eliminate incident,, then enter normal operating conditions, if local terminal access point link failure then attempts recovering the access point link if local terminal access point link is normal.

13, the method for Link State Pass Through in the realization network according to claim 12, it is characterized in that: be in the service equipment of Wait-to-Restore state if continuing not receive in the given time contains the break down control frame of information of link, then recover also initiatively to get back to normal operating conditions with user's access point link.

Images