CN1976370A

CN1976370A - Method for keeping call data uniformity after master-slave gateway replacing

Info

Publication number: CN1976370A
Application number: CN 200510101531
Authority: CN
Inventors: 唐娜; 王进
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2005-11-28
Filing date: 2005-11-28
Publication date: 2007-06-06

Abstract

A method for holding consistency of calling data after master and back-up gateways are switched in turns includes sending a request affairs to inform MGC by MG, sending response message by MGC, carrying out one time of audit on each context existed in H.248 protocol course by MGC and sending audit message (AM) to gateway, judging whether this context is existed on MG or not after AM is received by MG, sending normal response to MG if it is or otherwise sending response with error description symbol to MGC, releasing context and relevant terminal in response message with said symbol when AM is received by MGC.

Description

The method of keeping call data uniformity after master-slave gateway replacing

Technical field

The present invention relates to field of mobile communication, relate in particular to the call data disaster tolerance of calling out gateway in control and the carrying isolation technics in the mobile communcations system.

Background technology

Along with mobile communcations system in the increasingly extensive development of economic and social life every field, mobile communcations system develops rapidly, third generation communication system is perfect day by day, the demand of user to communication system satisfied in technical development day by day.

In the control and calling piece-rate system of mobile communcations system, be gateway controller (MGC, Media Gateway Control) and gateway (MG, MediaGateway) two parts by traditional calling is controlled and carried the device separates of concentrating.The unified management on gateway of bearing resource in the calling is called out to be controlled on the gateway controller and is realized, gateway controller can how easier introducing business.H.248 agreement is as main agreement mutual between gateway controller and the gateway.Gateway controller is finished foundation, modification and the release of bearer path and media channel by agreement control gateway H.248, to the configuration of the attribute of media channel and bearer path; To the operation of media channel and bearer path, comprise playback, audit etc.Gateway can be to the gateway controller reported event.

Using in the call flow of agreement H.248, H.248 protocol process is arranged all on MGC and MG.In call establishment, all can create in the H.248 protocol process of MGC and MG and call out relevant context and terminal.During unusual the or upgraded version of MG main frame, if there is standby host, standby host can be born the call business of main frame, and standby host is called out and can be continued after becoming new main frame, and user's communication normally carries out.

Because MG can not be to the all-calls data backup, but the calling that will enter talking state as far as possible backups to and carries out the disaster tolerance protection on the standby host.Only exist on the standby host of MG to enter talking state or soon enter the call data of talking state, the call data that MG will create two terminals at least backup on the standby host.Break down at the MG main frame, after the service of standby host input became new main frame, the call data of MGC and MG may be inconsistent, causes the communication system disorder, and then cause call loss.

Summary of the invention

Technical problem to be solved by this invention provides a kind of method, solve in the MC network element of prior art H.248 in the protocol process disaster-tolerant backup mechanism, during MG network element generation masterslave switchover, it is inconsistent that MGC and MG go up the call data of preserving, thereby cause the extension of the last data field of MGC dead, and then cause the problem of communication system disorder.

Technical scheme of the present invention is that when masterslave switchover took place MG, the H.248 protocol process on the MG carried out active and standby synchronous at least for the calling of having created two terminals.

Particularly, when gateway generation masterslave switchover, carry out following work:

The first step: MG sends a request transaction notice MGC:MG masterslave switchover takes place;

Second step: after MGC receives transaction request message, send response message to MG;

The 3rd step: MGC begins each context that exists in the protocol process is H.248 once audited, and sends audit message to gateway;

The 4th step: MG receives audit message, judges whether there is this context on the MG, if there is this context, then sends and replys normally to MGC; If there is not this context, then transmission has replying to MGC of error description symbol;

The 5th step: MGC receives replying of audit message, analyzes message content, and the response message for having the error description symbol discharges context and relevant terminal in this message.

The present invention has in time deleted MGC and has gone up abnormal call out corresponding terminal and contextual data region after masterslave switchover takes place MG, makes the call data of MGC and MG to be consistent.

Description of drawings

Fig. 1 is the context created in the H.248 protocol process of MGC and MG of a calling and the corresponding relation schematic diagram of terminal;

The synchronous schematic diagram of call data in the H248 protocol process of MGC and MG behind the masterslave switchover takes place in Fig. 2 for MG;

Fig. 3 is a call data Synchronous Processing schematic flow sheet in the H248 protocol process behind the MG masterslave switchover.

Embodiment

Below in conjunction with the drawings and specific embodiments the method for the invention is described further.

As shown in Figure 1, an important parameter of calling out in the corresponding H.248 process on MGC and MG is context and terminal.For same calling, context, terminal in the H.248 process of MGC and MG are corresponding.Among the figure, Ctxt calls out corresponding context ID, and T1, T2 are the ID of two terminals.

H.248 protocol process on the MG is synchronized on the standby host the call information of creating two terminals, only synchronous a part of call information on the standby host of MG.After masterslave switchover took place MG, it is not quite identical that MGC and MG go up the calling that exists, how that the call data on MGC and the MG are synchronous after being described in detail in MG below masterslave switchover taking place, and how to discharge MGC and go up and hang dead context and terminal data district.

The call data that the H.248 protocol process of MG will be created at least two terminals are synchronized on the standby host.After masterslave switchover takes place MG, have only the call data of having created at least two terminals on the new main frame.When masterslave switchover takes place in MG, if a calling has only been created a terminal in carrying out, this terminal and context of calling out correspondence exists on the H.248 protocol process of MGC, but behind the MG masterslave switchover, do not exist on the new main frame of MG, and this calling can't be proceeded.These abnormal context and terminals thereof should in time discharge in the H.248 protocol process of MGC.

A kind of method is described below, in time that the call data on MG and the MGC are synchronous when masterslave switchover takes place MG, discharge MGC and go up abnormal context and the terminal data district that exists.

As shown in Figure 2, supposing has two callings before masterslave switchover takes place MG, and one of them calls out corresponding context ID is C1, and two Termination ID are respectively T1, T2.It is C2 that another one is called out corresponding context ID, has only a terminal in this context, and Termination ID is T3.Context ID is that the calling of C1 has backuped on the MG standby host, and context ID is calling not backup on the standby host of MG of C2.When masterslave switchover takes place MG, standby host becomes new main frame input service, only having context ID on the new main frame of MG is the call data of C1, but to have context ID on MGC be that C1 and context ID are the call data of C2, and context ID to be the calling of C2 can't proceed.Should in time delete MGC and go up abnormal call out corresponding terminal and contextual data region, make the keeping calling data uniformity of MGC and MG.In conjunction with shown in Figure 3, concrete implementation step is as follows:

301:MG sends transaction request message to MGC, and masterslave switchover takes place notice MGC:MG, and standby host begins to start service.Can use the ServiceChange order in this transactions requests, method wherein is Failover, and Termination ID is a root termination.

After 302:MGC receives message, by analyzing message, learn that the MG standby host starts, MGC sends affairs and replys to MG.

303:MGC begins each context that exists in the protocol process is H.248 once audited.MGC begins context C1 is audited, and sends request transaction to MG, orders to be Auditvalue, and wherein context ID is C1, and Termination ID uses asterisk wildcard " * ", all terminals among the expression context C1.

304:MG receives the audit message of MGC to context C1, judges that MG goes up this context of existence C1, sends and replys affairs normally to MGC.

305:MGC begins context C2 is audited.Send request transaction to MG, order to be Auditvalue, wherein context ID is C2, and Termination ID uses asterisk wildcard " * ", all terminals among the expression context C2.

306:MG receives the audit message of MGC to context C2, judges on the MG not have this context C2, send have an error description symbol reply affairs to MGC, wherein the error description symbol can be " No_ContextIDs_Availables ".

307:MGC receives acknowledgement messaging, analyzes message content, for the response message that has error code, and free context C2 and terminal T3 wherein.

Claims

1, a kind of method of keeping call data uniformity after master-slave gateway replacing, when gateway generation masterslave switchover, the H.248 protocol process on the gateway carries out active and standby synchronous at least for the calling of having created two terminals, it is characterized in that, when gateway generation masterslave switchover, carry out following work:

The first step: gateway sends a request transaction notification gateway controller: gateway generation masterslave switchover;

Second step: after gateway controller is received transaction request message, send response message to gateway;

The 3rd step: gateway controller begins each context that exists in the protocol process is H.248 once audited, and sends audit message to gateway;

The 4th step: gateway is received audit message, judges whether there is this context on the gateway, if there is this context, then sends and replys normally to gateway controller; If there is not this context, then transmission has replying to gateway controller of error description symbol;

The 5th step: gateway controller is received replying of audit message, analyzes message content, and the response message for having the error description symbol discharges context and relevant terminal in this message.

2, the method for the described keeping call data uniformity after master-slave gateway replacing of claim 1 is characterized in that, utility command ServiceChange in the request transaction of the described first step, and method is Failover.

3, the method for the described keeping call data uniformity after master-slave gateway replacing of claim 1, it is characterized in that the auditing method in described the 3rd step is to use the order Auditvalue in the agreement H.248, context ID wherein is the context ID of appointment, and Termination ID uses asterisk wildcard.