WO2012167620A1

WO2012167620A1 - Method and system for reducing calling failure of color ring back tone service

Info

Publication number: WO2012167620A1
Application number: PCT/CN2012/070102
Authority: WO
Inventors: 贺镇海
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-06-08
Filing date: 2012-01-06
Publication date: 2012-12-13
Also published as: CN102821365B; CN102821365A

Abstract

Disclosed in the present invention are a method and a system for reducing calling failure of color ring back tone service, wherein, the system comprises: a primary device, a backup device and a share storage, and the method comprises: the primary device writes the calling data of said calling into the share storage after the calling service flow enters a designated state, the backup device reads said calling data from the share storage and recovers the calling after the handoff from the primary device to the backup device. The method and the system can effectively enable the color ring back tone playing and the call of the broadband color ring back tone service in the network comprising double devices for fault tolerance to be recovered after the handoff, reduce the calling failure rate during the period of the breakdown and recovery of a broadband color ring back tone host, and at the same time avoid influencing the normal service stability and the system performance.

Description

Method and system for reducing call failure of CRBT service

Technical field

The present invention relates to the field of mobile communication technologies, and in particular, to a method and system for reducing a call failure of a CRBT service after an abnormality of a broadband ring tones primary device service is switched to a standby device. Background technique

In the current ring network application of the CRBT service, the Service Control Point (SCP) of the intelligent network is built in a fault-tolerant manner. The main SCP is down, the network is disconnected, the database is abnormal, and the abnormal process of the intelligent network platform is detected by the dual-machine software, and the command is issued by the standby SCP. When the above-mentioned primary and backup SCP switching occurs, there are usually several calls in progress, and all the calls will fail due to the failure of the primary device. The length of the switching process depends on the setting of the heartbeat detection of the dual-machine, the hardware configuration level of the standby device, the network bandwidth, and the startup process of the standby intelligent network service. If the traffic peaks and the switching time is too long, it may cause a large number of call loss and cause more user complaints.

The broadband ring back tone service platform and the switching side and the playback device pass a session layer signaling control protocol (SIP), a bearer independent call control protocol (BICC), and a media gateway. The IP Gateway-based protocol, such as the Media Gateway Control Protocol (MGCP), interacts. The call flow is divided into a crossover scheme and a non-crossover scheme. The service platform in the crossover solution indicates that the CRBT plays and continues to be called. In the non-crossover solution, the service platform only indicates the CRBT playback, and the connected call is completed by the switching device. Summary of the invention

Currently, in the existing broadband CRBT service process, if the primary and backup devices are switched, the call release session is abandoned, and the standby device resumes accepting the initiated new call to complete the new CRBT service, and the backup device cannot restore the original call. Therefore, there is an urgent need for a method for recovering CRBT playback and call status after the main ring device switching of the CRBT service platform, and reducing the call failure rate during the collapse and recovery of the broadband CRBT host. The technical problem to be solved by the present invention is to provide a method and system for reducing the call failure of the CRBT service. After the primary and backup devices of the CRBT service platform are switched, the CRBT playback and call status can be restored, and the call failure during the collapse and recovery of the broadband CRBT host is reduced. rate.

In order to solve the above technical problem, the present invention provides a method for reducing a call failure of a CRBT service, including:

After the calling service process enters the specified state, the calling device writes the call data of the call to the shared memory, and after the active device and the standby device are switched, the standby device reads the call data from the shared memory. , resume the call.

Preferably, the call service process enters the specified state, where: the CRBT service platform uses the non-crossover solution to call the service process to enter the ring back tone state to the calling party, or the CRBT service platform uses the crossover solution to call the service process to enter the calling and calling party. Call status.

Preferably, the call data includes: data that plays the CRBT to the calling party when the CRBT service platform uses the non-crossover solution, or data of the calling and the called call when the CRBT service platform uses the bridging scheme.

Preferably, the method further includes:

After the standby device resumes the call, if the on-hook signaling is received, the call is released; otherwise, the state of the call service flow before the primary device and the backup device are switched is maintained or discarded according to the result of the in-call detection.

Preferably, the maintaining or abandoning the status of the call service flow before the switching of the active device and the standby device according to the result of the in-call detection includes:

If the CRBT service platform uses a non-crossover solution, the standby device sends an in-call detection message to the calling party, and if the response message of the calling party is received, the ring back tone state is kept to the calling party, otherwise the call is released. ;

If the CRBT service platform uses the bridging scheme, the standby device sends an in-call detection message to the calling party and the called party respectively, and if the response message of the calling party and the called party is received, the main message is maintained. The called party is in a call state, otherwise the call is released.

Preferably, after the calling device writes the call data of the call to the shared memory, the method further includes: marking the call data as valid, and if the primary device and the standby device are not switched after the call ends, The primary device marks the call data of the call in the shared memory as invalid Or delete;

After the standby device reads the call data from the shared memory, the call data written by the active device in the shared memory is marked as invalid or deleted.

Preferably, after the calling service process enters the specified state, the calling device writes the call data of the call to the shared memory, including:

In a given time or a given number of calls, the ratio of the number of calls written by the active device to the shared memory to the total number of calls in the given time or to the number of calls is not greater than Call data write ratio.

Preferably, the reading, by the standby device, the call data from the shared memory comprises: the number of calls corresponding to the valid call data read by the backup device from the shared memory is stored by the primary device The ratio of the number of calls corresponding to the valid call data in the shared memory is not greater than the call data read ratio.

Preferably, the call data write ratio/read ratio is determined as follows:

The system administrator sets the call data write ratio/read ratio;

The system periodically collects system performance indicators, and determines the corresponding call data writing ratio/read ratio according to the system performance indicators;

Comparing the corresponding call data write ratio/read ratio determined by the system with the call data write ratio/read ratio set by the system administrator, and selecting the smaller value as the final call data write ratio / Read ratio.

In addition, in order to solve the above technical problem, the present invention further provides a system for reducing a call failure of a CRBT service, including an active device, a backup device, and a shared memory, where

The active device is configured to: after the call service flow enters the specified state, write the call data of the call to the shared memory;

The standby device is configured to: after the primary device and the standby device are switched, read the call data from the shared memory to resume the call.

Preferably, the call service process enters a specified state, where: the CRBT service platform uses non-cross The incoming call service process enters the ring back tone state to the calling party, or the color ring tone service platform uses the crossover solution to call the service process to enter the calling and called call state.

Preferably, the standby device is further configured to: release the call if the on-hook signaling is received after the call is resumed; otherwise, maintain or abandon the call service flow before the primary device and the backup device are switched according to the in-call detection result. status.

Preferably, the standby device is further configured to:

If the CRBT service platform uses the non-crossover scheme, the in-call detection message is sent to the calling party, and if the response message of the calling party is received, the CRBT state is kept to be played to the calling party, otherwise the conversation is released; The CRBT service platform uses the bridging scheme to send an in-call detection message to the calling party and the called party respectively. If the response message of the calling party and the called party is received, the calling party and the called party are kept in a call state, otherwise the call is released. dialogue.

Preferably, the primary device is further configured to: mark the call data as valid after the call data of the call is written into the shared memory, and if the primary device and the backup device do not switch after the call ends, The call data of the call in the shared memory is marked as invalid or deleted; the standby device is further configured to: after reading the call data from the shared memory, write the primary device in the shared memory Incoming call data is marked as invalid or deleted.

Preferably, the primary device is configured to: the number of calls written by the primary device to the shared memory and the total number of calls in the given time in a given time or a given number of calls Or the ratio to the number of calls is not greater than the call data write ratio.

Preferably, the standby device is configured to: the number of calls corresponding to the valid call data read from the shared memory and the number of calls corresponding to the valid call data stored by the active device in the shared memory The ratio is not greater than the call data read ratio.

Preferably, the system further comprises a performance monitoring and data writing/reading control device, configured to: determine the call data writing ratio/read ratio according to the following manner:

Collecting system performance indicators at regular intervals, and determining corresponding call data writing ratios/read ratios according to the system performance indicators; Comparing the corresponding call data write ratio/read ratio determined by the system with the call data write ratio/read ratio set by the system administrator, and selecting the smaller value as the final call data write ratio/read Proportion.

Compared with the prior art, after the embodiment of the present invention is used, the CRBT playback and the call after the switching of the broadband CRBT service in the fault-tolerant network can be effectively restored, and the call failure rate during the collapse and recovery of the broadband CRBT host is reduced. At the same time, avoid affecting the stability of normal business and system performance. BRIEF abstract

FIG. 1 is a structural diagram of a system for reducing a call failure of a CRBT service in an embodiment; FIG.

2 is a flowchart of a method for reducing a call failure of a CRBT service in an embodiment;

FIG. 3 is a structural diagram of a system for reducing a call failure of a CRBT service in an application example; FIG.

FIG. 4 is a flow chart showing the operation of the system for reducing the failure of the CRBT service call shown in FIG. 3. Preferred embodiment of the invention

In order to make the objects, the technical solutions and the advantages of the present invention more clearly, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, in the case of no conflict, the features in the embodiments and the embodiments of the present application may be arbitrarily combined with each other.

Embodiment 1

As shown in FIG. 1 , an embodiment of the present invention provides a system for reducing a call failure of a CRBT service, where the system includes an active device, a backup device, a dual-machine monitoring device, and a shared memory, where: the primary device is used for After the call service flow enters the specified state, the call data of the call is written into the shared memory;

The call service process enters the specified state, and the CRBT service platform uses the non-crossover solution to call the service process to enter the ring back tone state to the calling party, or the CRBT service platform uses the crossover solution to call the service process to enter the calling and called call. status.

The call data includes: The CRBT service platform plays the ring back tone to the calling party when the non-crossover solution is used. The data, or the data of the calling party's call when the ring back tone service platform uses the bridging scheme.

Wherein, when the calling device writes the call data of the call to the shared memory, the number of calls written by the active device to the shared memory and the given time in a given time or a given number of calls The ratio of the total number of calls in the fixed time or the number of calls to the number of calls is not greater than the call data write ratio, that is, the active device writes the call data of the call according to a call data write ratio. Memory. The call data write ratio refers to the ratio of the number of calls written to the shared memory to the total number of calls in the given time or the number of calls, which may be set by the system administrator, It can be dynamically modified after the system is running, as other dynamically modifiable parameters are periodically read by the system and validated in the next cycle, or dynamically determined in conjunction with system performance indicators.

The standby device is configured to read the call data from the shared memory after the primary device and the standby device are switched, and resume the call.

The standby device reads the valid call data from the shared memory according to the call data read ratio. The call data read ratio refers to a percentage of the number of calls read out in the shared memory, and the call data read ratio is determined similarly to the call data write ratio, and may also be set by the system administrator. It can also be dynamically modified after the system is running. As other dynamically modifiable parameters are periodically read by the system and validated in the next cycle, or dynamically determined in combination with system performance indicators.

After the backup device resumes the call, the standby device is further configured to release the call if the on-hook signaling is received; otherwise, the call service flow is maintained or abandoned according to the in-call detection result. status.

The standby device maintains or discards the state of the call service flow before the switching between the active device and the standby device according to the result of the in-call detection, including:

If the CRBT service platform uses a bridging scheme, the standby device divides the calling party and the called party into Do not send an in-call detection message. If the response message of the calling party and the called party is received, the calling and calling state is maintained, otherwise the call is released.

The dual-machine monitoring device is configured to monitor the running status of the active device, and enable the standby device when the primary device is abnormal.

Correspondingly, as shown in FIG. 2, the embodiment of the present invention provides a method for reducing a call failure of a CRBT service, including the following steps:

S101: After the call service process enters the specified state, the active device writes the call data of the call to the shared memory. The call service process enters the specified state, where the call service process uses the non-crossover solution to call the service process. The CRBT status is played to the calling party or the CRBT service platform uses the bridging scheme to call the service flow in the main and called call state, and the active device writes the current call data to the shared memory; the status can also be set as needed. The current call data written by the active device to the shared memory includes: the CRBT service platform plays the CRBT data to the calling party when the non-crossover solution is used, or the data of the calling and the called call when the CRBT service platform uses the crossover solution. .

Wherein, when the calling device writes the call data of the call to the shared memory, the number of calls written by the active device to the shared memory and the given time in a given time or a given number of calls The ratio of the total number of calls or the number of calls to the call is not greater than the call data write ratio, that is, the active device writes the call data to the shared memory according to the call data write ratio. For example, the monitoring period can be set. Within each monitoring period, the ratio of the number of calls written to the shared memory to the total number of calls in the period does not exceed 30%, or, using a counter, set each N calls. The number of calls written to the shared memory does not exceed 30%, and the N value is set as needed.

The call data write ratio can be set by the system administrator, or can be dynamically modified after the system is running. The other dynamically changeable parameters are periodically read by the system and validated in the next cycle, or dynamically combined with system performance indicators. Determined.

S102: After the primary device is abnormal, the primary device and the standby device perform switching; S103: The standby device reads the call data from the shared memory to resume the call. The standby device reads the valid call data from the shared memory according to the call data read ratio. The ratio of the number of calls corresponding to the valid call data read by the standby device from the shared memory to the number of calls corresponding to the valid call data stored by the active device in the shared memory is not greater than the call data read Proportion.

The determination of the call data read ratio is similar to the call data write ratio, and may also be set by the system administrator, or may be dynamically modified after the system is running, and the other dynamically changeable parameters are periodically read by the system, and Effective within the next cycle, or dynamically determined in conjunction with system performance metrics.

After the standby device reads the call data from the shared memory, the call data written by the active device in the shared memory is marked as invalid or deleted. In this way, shared memory can save storage space, data that is identified as invalid can be overwritten by new data, and can guarantee good performance and stability of subsequent traffic. In the above steps, the call data writing ratio/read ratio may be determined in the following manner but not in other dynamic manners, including:

The system administrator sets the call data write ratio/read ratio;

Comparing the corresponding call data write ratio/read ratio determined by the system with the call data write ratio/read ratio set by the system administrator, and selecting the smaller value as the final call data write ratio / Read ratio. In addition, in step S103, after the standby device resumes the call, if the on-hook signaling is received, the call is released; otherwise, the state of the call service flow before the switching between the active device and the standby device is maintained or abandoned according to the in-call detection result. , Specifically:

If the CRBT service platform uses a non-crossover solution, the standby device sends an in-call detection message to the calling party, and if the response message of the calling party is received, the ring back tone state is kept to the calling party, otherwise the call is released. ; If the CRBT service platform uses the bridging scheme, the standby device sends an in-call detection message to the calling party and the called party respectively, and if the response message of the calling party and the called party is received, the main message is maintained. The called party is in a call state, otherwise the call is released.

In an application example, as shown in FIG. 3, the system for reducing a CRBT service call failure may further include: a performance monitoring and data writing/reading control device, configured to determine a call data writing ratio/read ratio, It can be determined in the following ways but not in other dynamic ways:

The performance monitoring and data writing/reading control device periodically collects system performance indicators, and determines a corresponding call data writing ratio/read ratio according to the system performance indicators;

The performance monitoring and data writing/reading control device compares the corresponding call data writing ratio/read ratio determined by the system with the calling data writing ratio/reading ratio set by the system administrator, and selects The smaller value is used as the final call data write ratio/read ratio.

Among them, the system administrator manually sets the call data write/read ratio, and sets the upper limit of the number of calls in the monitoring period (usually 10 minutes, which can be configured according to the actual situation). This value can be dynamically modified after the system is running. It can be read periodically by the system with other dynamically modifiable parameters and will take effect in the next cycle. The system performance indicators include overall CPU utilization and memory usage, critical process CPU and memory usage, and 10 throughput (optional). In a specific implementation manner, a system performance indicator in different situations may be associated with a call data write/read ratio, and the performance monitoring and data write/read control device may collect the performance data after the corresponding relationship. Matching to the performance-optimized record that can be satisfied, obtaining the corresponding call data write/read ratio, after entering the next monitoring cycle, the performance monitoring and data write/read control device continues to collect performance data and generate new Call data write/read ratio. In the specific implementation, the corresponding relationship between the system performance indicator and the call data currently allowed by the system can be written as shown in Table 1:

CPU usage Memory idle rate Current total traffic Call data write ratio

1 2% or less 50% or more 150 calls per second Up to 80%

Called the following 2 6% or less 40% or more 200 calls per second up to 60%

the following

Table 1: Correspondence between system performance indicators and the current call data write ratio allowed by the system. Assume that the upper limit of the data write ratio set by the system administrator is 70%. If the current system performance indicator corresponds to the first correspondence, that is, call data. The write ratio is up to 80%, and the smaller of the two is selected after the comparison, that is, 70% is the final call data write ratio written to the shared memory; if the current system performance indicator corresponds to the second correspondence, that is, the call data The write ratio is up to 60%, and the smaller of the two is selected after the comparison, that is, 60% as the final call data write ratio written to the shared memory. In this application example, the active device may also set a service status counter for every hundred calls, and the initial value of the counter is 0. In each monitoring period, a CRBT call triggers the counter to increase by 1, and reaches 100 and then resets. 0, re-accumulate. For example, if the final call data write ratio is 30%, the active device can set the call data corresponding to the counter count 0~29 to be written to the shared memory. In this way, it is possible to write the number of calls corresponding to the counter to the shared memory in accordance with the final call data write ratio. Correspondingly, the correspondence between the system performance index and the call data readout ratio currently allowed by the system can also be set as shown in Table 2:

Table 2: Correspondence between the system performance indicator and the current allowed call data read ratio of the system. Assume that the upper limit of the data read ratio set by the system administrator is 70%. If the current system performance indicator corresponds to the first correspondence, that is, the call data. The readout ratio is 80%, and after the comparison, the two are selected. The smaller value, that is, 70% as the final call data readout ratio read from the shared memory; if the current system performance index corresponds to the second correspondence, that is, the call data readout ratio is 60%, after the comparison, the two are selected. The smaller value, 60%, is the final call data readout ratio read from the shared memory. As shown in FIG. 4, in the application example, the workflow of the system for reducing a CRBT service call failure is as follows:

S201: The broadband ringtone intelligent network service platform starts normally, the active device runs normally, and the dual-machine monitoring device monitors the working status of the active device; whether the dual-machine monitoring device has an intelligent network platform, database or system abnormality for the primary device Monitoring; S202: The system administrator sets the call data write/read ratio, starts the performance monitoring and data write/read control device, periodically collects the system performance indicators (within the monitoring period), and generates corresponding Call data write/read ratio;

S203: The interactive side call, the CRBT service platform enters the ring tone playing state by using the non-crossover solution, or the color ringing service platform uses the bridging scheme and the calling party enters the call state; S204: the main device writes to the performance monitoring and data The in/out control device sends a message requesting to obtain a call data writing ratio of the current shared memory;

S205: The performance monitoring and data writing/reading control device compares the call data writing ratio corresponding to the system performance indicator (as shown in Table 1) with the call data writing ratio set by the system administrator, and selects both The smaller value is sent to the active device as the current shared memory call data write ratio; S206: The primary device writes the current call data to the shared memory according to the obtained call data write ratio according to the statistics of the service status counter; for example, The obtained call data write ratio is 30%, and the active device writes the call data corresponding to the counter count from 0 to 29 to the shared memory, and can also write the call data corresponding to the call with the count of 30 to 59 into the shared memory. Or the call data corresponding to the call under the discontinuous count is written into the shared memory as long as the call data of 30 calls per 100 calls is satisfied. The data structure for writing call data to the shared memory mainly includes: a dialog protocol identifier (such as a call-id, a from-tag, and a to-tag field in the sip protocol), a user number (including a calling and called number, Forwarding number, _ur i, etc., routing data of the protocol stack, current media negotiation status, and saved media information, and identifying the validity of the data area in the fixed area of the shared memory file (initially "valid,"), the data The segment is fixed in length in the shared memory.

S207: The dual-machine monitoring device monitors whether the primary device is abnormal. If no abnormality is found, the primary device continues to work normally, and the call data in the shared memory is marked as invalid or deleted after the call ends; otherwise, the process goes to step S208;

S208: The dual-machine monitoring device starts the standby device;

S209: The standby device sends a message to the performance monitoring and data writing/reading control device, requesting to obtain a call data read ratio of the current shared memory; S210: performance monitoring and data writing/reading control device comparing the system performance indicator Corresponding call data read ratio (as shown in Table 2) and the call data read ratio set by the system administrator, select the smaller of the two as the current shared memory final call data read ratio, and send to the standby device;

S211: The standby device reads out the call data marked as valid in the shared memory according to the obtained call data read ratio, and then marks the call data written by the active device in the shared memory as invalid or deleted, according to the read. Call data to recover this part of the call;

S212: If the standby device receives the on-hook signaling, release the session, and end the call; otherwise, go to step S213;

S213: Send an in-call detection message to the calling party and the calling party according to the bridging and non-crossover scheme configured by the service;

In order to avoid abnormality such as network congestion and packet loss caused by excessive data transmission in an instant, the detection message is sent in several batches. The default value can be used for each batch of data. It can also be configured by the administrator based on the current traffic of the current network and the status of the network device.

S214: If the response is detected in the session, the session is continued, otherwise the session is released; in the above steps S213-S214, for the spanning scheme, the detection message is separately sent to the calling party, and the waiting period is not received after the sending. The called detection response or the response carries an error code to inform that the call does not exist (such as the sip protocol 481 Call/Transaction Dose Not Exist response), then release the call in memory Call data; otherwise, if the calling party side sends a correct response (such as the sip protocol 200 OK response), the calling party is still in the call state, the memory call data continues to be maintained, and the hang-up from the calling party side is resumed, and the media negotiation is performed again. Wait for the request.

For the non-crossover scheme, the detection message is sent to the calling side, and the caller's detection response is not received within the waiting period after the transmission, or the response carries the error code to inform that the call does not exist, and the call data in the memory is released; otherwise, if the calling side sends If it responds correctly, it is still in the ring tone playback state, the memory call data continues to be maintained, and the on-hook request from the calling side continues to be processed later for the calling party or the calling party to release the call.

S215: The standby device replaces the primary device service, and the dual-machine monitoring device monitors the working state of the standby device; subsequent requests are processed according to the primary device.

Determine the current write shared memory call data ratio similarly to the active device, and play the CRBT

In the (non-crossover scheme) or call (crossover scheme) state, the data area is written to the shared storage for re-switching.

So far, this application example has completed the workflow of the system that failed to call the CRBT service.

It can be seen from the above-mentioned embodiments that the method and system for reducing the CRBT service call failure provided by the present invention are used for the CRBT playback state or the crossover scheme of the non-crossover solution before the dual-machine switching of the broadband CRBT service platform is compared with the prior art. Under the scenario of the call state, the CRBT playback and the call after the switching of the broadband CRBT service in the fault-tolerant network can be effectively restored, and the call failure rate during the collapse and recovery of the broadband CRBT host is reduced, and the normal service is not affected. Stability.

In addition, the method and system for reducing the call failure of the CRBT service provided by the present invention overcome the problem of writing a large amount of call data to the shared storage, which may cause a decrease in CPU performance and 10 is too busy, by controlling the write/read data of the shared memory. Defects greatly improve CPU performance and avoid system waste.

One of ordinary skill in the art will appreciate that all or a portion of the above steps may be performed by a program to instruct the associated hardware, such as a read only memory, a magnetic disk, or an optical disk. Alternatively, all or part of the steps of the above embodiments may also be implemented using one or more integrated circuits. Correspondingly, each module/unit in the above embodiment can be used The form of hardware implementation can also be implemented in the form of software function modules. The invention is not limited to any specific form of combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. In view of the present invention, various other modifications, equivalents, improvements, etc., should be made without departing from the spirit and scope of the invention. It is included in the scope of protection of the present invention.

Industrial applicability

After the embodiment of the present invention is used, the CRBT playback and the call after the switching of the broadband CRBT service in the fault-tolerant network can be effectively restored, and the call failure rate during the collapse and recovery of the broadband CRBT host is reduced, and the normal service is avoided. Stability and system performance.

Claims

Claim

1. A method for reducing a call failure of a CRBT service, including:

2. The method of claim 1 wherein

The call service process enters the specified state, including: the CRBT service platform uses the non-crossover solution to call the service process to enter the CRBT state to the calling party, or the CRBT service platform uses the bridging scheme to call the service process to enter the calling and calling state.

3. The method of claim 1, wherein

The call data includes: data that the CRBT service platform plays to the calling party when the non-crossover solution is used, or the data of the calling party and the called party when the CRBT service platform uses the bridging scheme.

4. The method of claim 1, wherein the method further comprises:

5. The method of claim 4, wherein

And maintaining, according to the result of the in-call detection, the status of the call service flow before the switching between the active device and the standby device, including:

6. The method of claim 1, wherein

After the calling device writes the call data of the call to the shared memory, the method further includes: marking the call data as valid, and if the primary device and the standby device are not switched after the call ends, the primary device The device marks the call data of the call in the shared memory as invalid or deleted;

The method of claim 1, wherein the method further comprises:

After the calling service enters the specified state, the calling device writes the call data of the call to the shared memory, including:

8. The method of claim 1, wherein

The reading, by the standby device, the call data from the shared memory includes: the number of calls corresponding to the valid call data read by the standby device from the shared memory is stored with the active device to the share The ratio of the number of calls corresponding to the valid call data in the memory is not greater than the call data read ratio.

9. The method according to claim 7 or 8, wherein

The call data write ratio/read ratio is determined as follows:

The system administrator sets the call data write ratio/read ratio;

10. A system for reducing a call failure of a CRBT service, comprising a primary device, a backup device, and a shared memory, wherein

The standby device is configured to: after the primary device and the standby device are switched, read the call data from the shared memory to resume a call.

11. The system of claim 10, wherein

12. The system of claim 10, wherein

13. The system of claim 10, wherein

The standby device is further configured to: release the call if the on-hook signaling is received after the call is resumed; otherwise, maintain or abandon the state of the call service flow before the primary device and the standby device switch according to the in-call detection result.

14. The system according to claim 13, wherein the standby device is further configured to: if the CRBT service platform uses a non-crossover scheme, send an in-call detection message to the calling party, if the calling party is received The response message keeps playing the CRBT status to the calling party, otherwise the dialog is released; if the CRBT service platform uses the bridging scheme, the caller and the called party respectively send an in-call detection message, if the call is received The response message of the calling party and the called party keeps the calling and calling state of the call, otherwise the dialogue is released.

15. The system of claim 10, wherein

The primary device is further configured to: mark the call data as valid after the call data of the call is written into the shared memory, and if the primary device and the backup device are not sent after the call ends Switching, the call data of the call in the shared memory is marked as invalid or deleted; the standby device is further configured to: after reading the call data from the shared memory, the shared memory The call data written by the primary device is marked as invalid or deleted.

16. The system of claim 10, wherein

The primary device is configured to: in a given time or a given number of calls, the number of calls written by the primary device to the shared memory and the total number of calls in the given time or The ratio of the number of calls to the threshold is not greater than the call data write ratio.

17. The system of claim 10, wherein

The standby device is configured to: the ratio of the number of calls corresponding to the valid call data read from the shared memory to the number of calls corresponding to the valid call data stored by the active device in the shared memory is not Greater than the call data read ratio.

18. The system of claim 16 or 17, wherein

The system also includes a performance monitoring and data write/read control device configured to: determine the call data write ratio/read ratio as follows:

Collecting system performance indicators periodically, and determining the corresponding call data writing ratio/read ratio according to the system performance index; writing the corresponding call data determined by the system to the ratio/read ratio and the system administrator setting The predetermined call data write ratio/read ratio is compared, and the smaller value is selected as the final call data write ratio/read ratio.