WO2007009289A1

WO2007009289A1 - A method for configuring router device in both online and offline

Info

Publication number: WO2007009289A1
Application number: PCT/CN2005/001050
Authority: WO
Inventors: Youquan He; Changjiang Yang
Original assignee: Zte Corporation
Priority date: 2005-07-15
Filing date: 2005-07-15
Publication date: 2007-01-25
Also published as: CN100549981C; CN101133399A

Abstract

A method for configuring router device in both online and offline manner adopts setting up two independent configuration tables, one for the online and the other for the offline, in the realtime in-memory database to support both of the online and offline configuration concurrently to use the realtime in-memory database and to store the online and offline configuration data independently. The method contains: defining the protocol-stack data retrieves data from the in-memory database in the order of online-data-first or offline-data-first, and marking one of the two tables valid by setting a status bit on the table.

Description

Method for online and offline hybrid configuration of routing devices

The present invention relates to the field of telecommunications technology, and in particular, to a method for realizing online and offline hybrid configuration of a device having a routing function by using a real-time memory database. Background technique

In the configuration of the device with routing function in the prior art, for convenience of description, the device with the routing function is simply referred to as a router, and generally adopts an online configuration mode, that is, directly using the human-machine command. The router is configured and waits for the response from the router to determine whether the setting is successful. The configuration is as shown in the dotted line in Figure 1.

This method has been patented by Cisco. The US patent number is US6704752. The processing flow of this patent is: The user inputs the command, and the processing subsystem converts the command into a database table after receiving the online configuration command. The record form is ready to be inserted into the database. If the database does not allow the operation, it is returned directly to the user, otherwise it is verified. If the error message is reported if the verification fails, the verification is inserted into the database; if other subsystems need the database table configuration change message, the database can also provide.

The advantage of this technology is that the operator can immediately know if the set command is successful. However, the shortcomings of this approach are also very obvious, including:

1) It must be set when the router is running.

2) The boards to be configured must also be configured to run.

3) Different router settings are generally different. In particular, engineers and post-maintenance personnel at the start of the process must be familiar with these configuration methods, which increases the difficulty of these engineers and maintenance personnel, and thus is difficult to meet the needs of the project start-up. .

In addition, the prior art also has an offline configuration mode, as shown in the solid line frame in FIG. 1, which utilizes an in-memory database for caching, so that the protocol stack is acquired in the database when needed, and the way of changing the database is to notify the protocol stack. . The specific process is that the interface (client) performs configuration data, and after the client performs the necessary legality check, it is stored in the commercial database, and a large amount of data can be stored in the commercial database. When the configuration of the running router needs to be changed, the data in the commercial database is taken out by the adaptation/transfer process and transmitted to the in-memory database of the router, and the in-memory database discovers that the data changes, such as data addition, deletion, modification, etc. The processing process of the protocol stack is notified, and the protocol stack processing process re-acquires the corresponding data from the in-memory database for checking and verifying, and then returns the status field of the table for labeling. The advantage of this method is that it can be separated from the running board during configuration, reducing maintenance costs. Summary of the invention

The object of the present invention is to provide a method for performing on-line and off-line hybrid configuration of a routing device, and adopting a new configuration mode, which combines the advantages of the two existing technologies, so that the configuration can be performed on a board that is out of operation. Conditions can also be configured on-site using an online method to further reduce the difficulty of engineering and maintenance personnel and further increase the corresponding speed of processing problems.

The technical solution of the present invention is as follows: '

A method for online and offline hybrid configuration of a routing device, which adopts a separate online configuration table and an offline configuration table in a real-time in-memory database to simultaneously support online and offline configuration, and completely and independently save offline configuration data and online configuration. Data; the method further includes:

It is agreed that the protocol stack obtains data from the real-time in-memory database at power-on, and the rules of online data prioritization or off-line data prioritization are performed in order, and are validated by setting the status bits of the online or offline configuration table.

The method further includes the step of: the real-time in-memory database process loading data on the physical medium or the commercial database into the memory, including storing an online configuration table or an offline configuration. In the table, all status bits in the offline configuration table associated with the routing configuration are initialized to an invalid state; _f

Bl), when the protocol stack process is powered on, obtain online configuration information from the real-time memory database, and after simple verification, put into the data area of the protocol stack;

Cl), obtaining the information of the offline configuration table, the database interface process obtains the data from the offline configuration table and returns the data to the protocol stack, and the protocol stack is verified by the legality check and written into the data area of the protocol stack to determine whether it conflicts with the online data, and passes The database interface returns the status of the table record and completes the power-on.

The method further includes: the online configuration process is -a2), the online configuration process uses the client to directly interact with the protocol stack process, and the protocol stack process receives the message sent by the client, and determines whether Compliance with the rules, whether it conflicts with the configuration in its own data area;

B2) If the rules are not met or conflict with the configuration in the data area, the online configuration process directly returns the configuration failure; otherwise, the data is written into the protocol data area, and the interface of the calling database is written to the online memory database. In the configuration table, it is stored on the physical medium.

The method, wherein the method further includes an offline configuration change process as

A3), after receiving the data sent from the adaptation/transmission process, the real-time in-memory database inserts the data into an offline configuration table of the real-time memory database on the board and responds to the adaptation/transmission process;

B3), when it is found that the data in the offline configuration table changes, a configuration change message is sent to the protocol stack process. The method, wherein the data of the real-time in-memory database is changed by a table record level change notification, which includes the following steps:

C31), when the table in the in-memory database detects that the record has changed, immediately sends the changed record to the protocol stack process;

D31), the protocol stack process processes the message immediately after receiving the message, determines whether it conflicts with the online data, and satisfies the configuration rule. If the set configuration information can take effect, write back the real-time memory database, and put the record into effect. Sign; if it does not work, set the non-validation flag and the reason.

The method, wherein the data change of the real-time in-memory database adopts a table-level change notification, which is to send a notification message to the protocol stack process after the entire table synchronization is completed, and the protocol stack process receives the notification message. Go to the real-time in-memory database and go to the offline configuration table to read the records in turn, and check whether the configuration can take effect, and then return the record whether the record is valid.

The method for performing the online and offline hybrid configuration on the routing device according to the present invention can be configured offline while the online data is configured by the protocol stack, and the two configuration modes do not interfere with each other, so that the rear personnel can be allowed to be opened at the beginning of the deployment. It can be configured at home according to the opening requirements of the administrator to avoid misconfiguration of the on-site staff. It can also be configured on-line by on-site personnel as needed, so that faults can be resolved more quickly and reduced in later maintenance. Traveling to the equipment site reduces maintenance costs. DRAWINGS

The specific principles and applications of the method of the present invention will be described in more detail below with reference to the accompanying drawings, wherein - FIG. 1 is a flow chart showing the simultaneous support of the online and offline configuration modes;

2 is a schematic diagram of a power processing process of a protocol stack of the method of the present invention;

3 is a schematic diagram of an online configuration process of the method of the present invention;

4 is a schematic diagram of an offline configuration process of the method of the present invention;

The principles of online data prioritization are employed in the various flow charts of the method of the invention shown. detailed description

The benefits of the present invention will be more apparent from the following detailed description of the invention. The method for performing online and offline hybrid configuration of a routing device according to the present invention simultaneously supports an online and offline configuration scheme using a real-time in-memory database, and completely independent of the offline configuration data and the online configuration data by using a separate table, and It is agreed that the protocol stack acquires data from the database at power-on, using online data prioritization or offline data prioritization rules in order. Since the data of the online configuration and the data of the offline configuration are stored in different tables, the configuration data between the two does not interfere with each other in the database, and the method of the present invention cannot modify the offline configured data in an online manner, nor can it be offline. The way to modify online data, offline configuration data can only be changed offline, online configuration data can only be modified online. The order of effectiveness between the two configuration data is achieved through the convention of the rules, so that online and offline can exist at the same time.

When the online configuration data has been validated, that is, stored in the online table, then the offline configuration will not take effect, but the offline table will have this record, but the status bit is invalid; when this online data is available After the deletion, the protocol stack can obtain offline data from the offline data table to make the data take effect, and reset the state; similarly, a certain data configured offline is valid in the data area of the protocol stack, that is, the corresponding record in the offline data table. The status bit is set to the active state by the protocol stack, then the data will not be configured successfully online, and there will be no such record in the online table. Unless the data is deleted offline first, the data can be configured online.

As shown in FIG. 1 to FIG. 4, a flow chart of an application example of offline and online configuration adopted by ZTE Corporation is used to illustrate the specific application process of the present invention. The protocol stack data area in the figure refers to a board with routing function. The data area of the protocol stack process.

The internal part of the solid line frame shown in Figure 1 is a schematic diagram of the offline configuration. First, it is configured according to the interface, that is, the client, such as IP, MAC (Media Access Control) required to implement the RTP (Real Time Transport Protocol) and SCTP (Stream Control Transmission Protocol) functions. The data is checked by the client, and the data is verified by the client. If the data is legal, it is stored in the commercial database. The configuration does not care whether the program of the board runs. After the data configuration is completed, the data in the commercial database is taken out by the adaptation/transfer process to be synchronized to the in-memory database on the board. The management and maintenance process of the in-memory database writes the configuration to the data area of the protocol stack through the message exchange with the protocol stack process, so that the configuration takes effect. The inside part of the dotted line shown in Figure 1 is a schematic diagram of the online configuration. Since the data of the online configuration needs to be stored on the physical medium, the configured data is written to the online table of the in-memory database after the data area of the protocol stack is valid. The storage mechanism of the in-memory database is used to store on the physical medium. It should be noted that the online data can be directly stored on the physical medium without passing through the database, and the processing is slightly changed, and is known to those skilled in the art, and thus does not affect the design idea of the invention.

The protocol stack process on the board and the inter-process process message interaction mainly have the following parts -

1. Power-on process.

Figure 2 is a schematic diagram of the power-on processing flow of the protocol stack. Generally, the in-memory database management process on the board is powered on earlier than the protocol stack process. The database process loads the data on the physical medium or the commercial database into the memory in some manner, including storing the online configuration table and the offline configuration table. Medium, all offline related to routing configuration The status bits in the table are initialized to an invalid state. When the protocol stack process is powered on, the online configuration information is obtained from the database. Since the online data is completely in accordance with the rules, the data is considered to be correct, and after a simple verification, it can be directly placed in the data area of the protocol stack. Then, the information of the offline configuration table is obtained, and the database interface process obtains the data from the offline related table and returns it to the protocol stack. The protocol stack is processed, and the legality is verified and written into the data area of the protocol stack, and it is necessary to determine whether it conflicts with the online data. And the status is recorded back to the table through the database interface, and the power is turned on. The database interface is an interface of an in-memory database.

When the protocol stack process receives a peripheral board, it refers to the board in a system other than the router. After registering (powering on or deleting) the notification message, go to the database to query the record status of the offline configuration table corresponding to the board. After the judgment process, the status of the table is recorded, and the routing function of the board is implemented. This part is primarily related to offline configuration and is well known in the art and is not shown in the figure.

2, online configuration process

The online configuration uses the interface (client) to directly interact with the protocol stack process. After receiving the message sent by the interface (client), the protocol stack process determines whether it meets the rules and whether it conflicts with the configuration in its own data area. Or directly return configuration failure in the conflict with the configuration in the data area; otherwise, the data is written into the protocol stack data area, and then the interface of the calling database is written into the online table of the database, and then stored on the physical medium. Due to the principle of online priority, online configuration is basically independent of offline.

3. Offline configuration change process.

After receiving the data sent from the adaptation/transmission process, the in-memory database on the board inserts the data into the offline configuration related two-dimensional relation table of the memory database on the board and responds to the adaptation/transmission process. When the data in the two-dimensional relation table is found to change, a configuration change message is sent to the protocol stack process, as shown in FIG.

Data changes to the database can take the form of two notifications: a table record level change notification and a table level change notification. The so-called table record level notification form is that the table in the in-memory database detects that the record has changed, and immediately sends the changed record to the protocol stack process. The protocol stack process processes the message immediately after receiving the message, and determines whether it is in conflict with the online data. Configure rules. If the protocol stack finds that the configured configuration information can be valid, it writes back the database and puts the record on the valid flag. If it does not take effect, it sets the invalidation flag and the reason. This configuration change mode is called the record-level configuration change mode. The table-level change notification is to send a notification message to the protocol stack after the synchronization of the entire table is completed. After receiving the notification message, the protocol stack goes to the database to read the records in turn, and checks whether the configuration can take effect, and then resets them in turn. A flag that records whether it is in effect. In practice, both methods are used. For example, the configuration of the IP uses a record-level configuration to change the message flow, and for the static route, the table-level record is used to change the configuration process.

While the preferred embodiment of the present invention has been shown and described, it will be understood that the various modifications, variations and variations of the components may be practiced without departing from the scope of the invention. Therefore the invention does not The invention is intended to be limited to the specific embodiments of the invention disclosed herein.

Claims

Rights request

A method for performing online and offline hybrid configuration on a routing device, which uses a separate online configuration table and an offline configuration table in a real-time in-memory database to simultaneously support online and offline configuration and utilize the real-time in-memory database, and The offline configuration data and the online configuration data are completely and independently saved; the method further includes: agreeing that, when the power is on, the protocol stack obtains data from the real-time in-memory database, and adopts an online data priority or an offline data priority rule in sequence, and passes Set the status bit corresponding to the online or offline configuration table to make it valid.

2. The method according to claim 1, wherein the method further comprises the step of powering up:

Al), the real-time in-memory database process loads data on a physical medium or a commercial database into the memory, including storing an online configuration table or an offline configuration table, and all status bits in the offline configuration table related to the routing configuration are initialized to Invalid state

The method according to claim 1, wherein the method further comprises: the online configuration process is: a2), the online configuration process uses the client to directly interact with the protocol stack process, and the protocol stack process receives the client. After the sent message, it is judged whether it conforms to the rule, whether it conflicts with the configuration in its own data area;

B2) If the rules are not met or conflict with the configuration in the data area, the online configuration process directly returns the configuration failure; otherwise, the data is written into the protocol stack data area, and the interface of the calling database is written to the online real-time memory database. In the configuration table, it is stored on the physical medium.

4. The method according to claim 1, wherein the method further comprises an offline configuration change process as

B3), when it is found that the data in the offline configuration table changes, a configuration change message is sent to the protocol stack process.

5. The method according to claim 4, wherein the data of the real-time in-memory database is changed by a table record-level change notification, which includes the following steps - c31), and the table in the in-memory database detects that the record has changed. Immediately send the changed record to the stated Protocol stack process;

The method according to claim 4, wherein the data change of the real-time in-memory database adopts a table-level change notification, which is to send a notification message to the protocol stack process after the entire table synchronization is completed, After receiving the notification message, the protocol stack process goes to the offline real-time memory database to sequentially read the record in the offline configuration table, and checks whether the configuration can take effect, and sequentially returns whether the record is valid or not.

δ