CN112187519A

CN112187519A - Multi-activity concurrency method and device for realizing policy control based on BGP (Border gateway protocol)

Info

Publication number: CN112187519A
Application number: CN202010937558.8A
Authority: CN
Inventors: 张琪
Original assignee: Zhongying Youchuang Information Technology Co Ltd
Current assignee: Zhongying Youchuang Information Technology Co Ltd
Priority date: 2020-09-09
Filing date: 2020-09-09
Publication date: 2021-01-05

Abstract

The invention discloses a BGP protocol-based multi-active concurrency method and a device for realizing policy control, wherein the method comprises the following steps: the BGP controller starts active registration; if the number of the BGP controllers is larger than or equal to 3 and no leader controller exists, electing the leader controller; after receiving a request for building a BGP peer, a leader controller distributes and builds a plurality of active BGP peers; after the Leader controller receives a request for issuing the strategy, the strategy is issued more actively; after discovering that a certain Node controller reports no keepalive message for 2 times continuously, the Leader controller performs down processing on the Node controller; and the node controller and the Leader controller are interactive without response, and the Leader controller is down for processing. The method and the device ensure the reliability of the controller by realizing a multi-active concurrency mechanism of the BGP peer, and provide guarantee for network safety.

Description

Multi-activity concurrency method and device for realizing policy control based on BGP (Border gateway protocol)

Technical Field

The invention relates to the field of BGP (Border gateway protocol), in particular to a multi-activity concurrency method and a multi-activity concurrency device for realizing policy control based on a BGP protocol.

Background

The BGP protocol has excellent extension attributes, extends BGP SR/SRv6-Policy, BGP FLOWSPEC, BGP-LS and EVPN routing, and is a natural SDN control plane protocol. Under the basic premise of switching control separation, the controller can learn the routing information and issue the corresponding routing strategy in real time through a peer of the BGP, and control the traffic forwarding path of the network according to the intention.

The following problems exist for the BGP peers of forwarding devices by BGP controllers:

1. the policy issued by the peer of the single BGP cannot guarantee the reliability of the policy.

2. The multiple strategies are distributed to different BGP peers according to distribution rules, and when a certain strategy is specified, the strategy is still single active, so that the reliability of the strategy cannot be guaranteed.

Disclosure of Invention

In order to overcome the problems of the BGP peers of the forwarding equipment of the conventional BGP controller, the invention provides a multi-activity concurrency method and a device for realizing policy control based on a BGP protocol.

In order to achieve the purpose, the invention adopts the following technical scheme:

in an embodiment of the present invention, a method for implementing multi-active concurrency for policy control based on a BGP protocol is provided, where the method includes:

deploying N (N is more than or equal to 3) BGP controllers, and after the single BGP controller is started, realizing registration to a service bus and acquiring information of other BGP controllers;

if the number of the BGP controllers is larger than or equal to 3 and no leader controller exists, electing the leader controller;

after receiving a request for establishing a BGP peer, a leader controller distributes and establishes a plurality of active BGP peers;

after receiving the request of issuing the strategy, the Leader controller carries out a multi-activity issuing strategy;

after discovering that a certain Node controller has no keepalive message to report for 2 times continuously, the Leader controller carries out Node controller down processing;

and (4) the node controller and the Leader controller keep no response in interaction, and the Leader controller is down processed.

Further, after the single BGP controller is started, it registers to the service bus and obtains information of other BGP controllers, including:

configuring the name and the address of a service bus of the BGP controller and a management domain where the BGP controller is located, starting the BGP controller through a script, and realizing active registration of the BGP controller by calling a registration interface of the service bus through the BGP controller;

after the registration is successful, the BGP controller acquires other BGP controllers in the domain and related roles thereof through a service bus, calls other BGP controller notification interfaces in the domain, notifies a new BGP controller of going online, and synchronously updates a BGP controller active list by the other BGP controllers in the domain;

and when detecting that the leader controller exists, sending a keepalive interaction message to the leader controller 5s once through the restful interface actively detected by the leader controller.

Further, the electing leader controller comprises:

detecting that the number of BGP controllers is more than or equal to 3, actively calling restful interfaces of leader requests of other BGP controllers, and initiating the leader requests;

the other BGP controllers which do not initiate the leader request preferentially reply to the received BGP controller of the first leader request as ok;

after the BGP controller initiating the leader request receives more than half of the ok messages, the notification interfaces of all the other BGP controllers are called concurrently to become the leader;

after receiving the notification leader message, the BGP controller becomes a slave node controller, calls a restful interface of active detection of the leader controller, and sends a keepalive interaction message to the leader controller;

and after the Leader election is finished, the Leader controller replies a keepalive message request of the node controller.

Further, the distributing establishes a multi-active BGP peer, comprising:

detecting that the number of the BGP controllers is 2, and actively and synchronously establishing all BGP peers and all BGP strategies by the new online controller;

after receiving a request for establishing a BGP peer, a Leader controller allocates a controller with the least two peers and a forwarder device to establish the BGP peer according to a least-priority allocation principle;

and calling the notification interfaces of the other node controllers by the Leader controller to notify the newly established BGP peers, and updating the local BGP peer list after the node controllers receive the notification interfaces.

Further, the multi-activity issuing strategy comprises the following steps:

after receiving a request for issuing a strategy, the Leader controller searches a local BGP peer table;

if the peer is not established, executing distribution to establish a multi-active BGP peer;

and searching two node controllers establishing BGP peers with corresponding repeaters according to the local BGP peer table, calling issuing strategy interfaces of the two node controllers, issuing strategies to repeater equipment, and informing all the node controllers of updating the strategy list by the leader controller.

Further, the Node controller down processing includes:

after finding that a certain node controller does not report keepalive messages for 2 times continuously, the Leader controller initiatively initiates dial testing, the dial testing is normal, and a common alarm is sent to warn that the keepalive interaction of the node controller is abnormal;

if the Leader controller is abnormal in dial test, the other node controllers are informed to carry out dial test, if the dial test of more than half of the node controllers does not respond, a serious alarm is sent out, the node controller is interrupted, the Leader controller informs all the node controllers to delete the peer list of the node controller, and the BGP peer and the multi-active sending strategy are triggered and distributed again;

and if the dial testing of the node controller is normal, sending a common alarm to alarm the network abnormality between the node controller and the leader controller.

Further, the Leader controller down processing includes:

the node controller and the leader controller are interactive without response and send out a notice;

and if more than half of the node controllers do not respond to the interaction with the leader controller, re-triggering the election leader controller, and distributing and establishing the multi-active BGP peers and the multi-active issuing strategy.

In an embodiment of the present invention, a multi-active concurrent device for implementing policy control based on a BGP protocol is further provided, where the device includes:

the BGP controller registration module is used for realizing registration to the service bus after the single BGP controller is started and acquiring information of other BGP controllers;

the leader controller election module is used for discovering that the number of the BGP controllers is more than or equal to 3 and no leader controller exists, and electing the leader controller;

the distribution and establishment module of the multi-active BGP peer is used for distributing and establishing the multi-active BGP peer after the leader controller receives a request for establishing the BGP peer;

the multi-activity issuing strategy module is used for carrying out multi-activity issuing strategies after the Leader controller receives the strategy issuing request;

the Node controller down processing module is used for performing Node controller down processing after the Leader controller finds that a certain Node controller has no keepalive message to report for 2 times continuously;

and the Leader controller down processing module is used for performing the Leader controller down processing on the node controller and the Leader controller keepalive without response in an interactive mode.

Further, the BGP controller registration module is specifically configured to:

Further, the leader controller election module is specifically configured to:

Further, the multi-active BGP peer distribution establishment module is specifically configured to:

Further, the multi-activity delivery policy module is specifically configured to:

Further, the Node controller down processing module is specifically configured to:

if the Leader controller is abnormal in dial test, the other node controllers are informed to carry out dial test, if the dial test of more than half of the node controllers does not respond, a serious alarm is sent out, the node controller is interrupted, the Leader controller informs all the node controllers to delete the peer list of the node controller, and the multi-active BGP peer distribution establishing module and the multi-active sending strategy module are triggered again;

Further, the Leader controller down processing module is specifically configured to:

and if more than half of the node controllers do not respond to the interaction with the leader controller, the leader controller election module, the multi-active BGP peer distribution establishing module and the multi-active sending strategy module are triggered again.

Has the advantages that:

the invention guarantees the reliability of the controller by realizing the multi-active concurrency mechanism of the BGP peer, thereby providing guarantee for network safety.

Drawings

FIG. 1 is a simplified flow diagram of a dual-activity mechanism according to an embodiment of the present invention;

FIG. 2 is a flow diagram of a multi-election distribution mechanism according to an embodiment of the invention;

FIG. 3 is a low latency traffic routing diagram for the cloud and metro network interworking traffic through the backbone portion via bgp srv6-policy distribution in accordance with an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a multi-active concurrency device according to an embodiment of the invention.

Detailed Description

The principles and spirit of the present invention will be described below with reference to several exemplary embodiments, which should be understood to be presented only to enable those skilled in the art to better understand and implement the present invention, and not to limit the scope of the present invention in any way. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

As will be appreciated by one skilled in the art, embodiments of the present invention may be embodied as a system, apparatus, device, method, or computer program product. Accordingly, the present disclosure may be embodied in the form of: entirely hardware, entirely software (including firmware, resident software, micro-code, etc.), or a combination of hardware and software.

Aiming at two problems of the prior BGP controller aiming at the BGP peers of the forwarding equipment, the reliability scheme of the BGP controller provided by the invention is divided into the following two types:

1. the simple dual-activity mechanism is shown in fig. 1, and specifically comprises the following steps:

the method comprises the steps that main BGP controllers and standby BGP controllers are deployed, all routing strategies published through BGP are issued in two copies through the main BGP controllers in real time, a single strategy realizes double activities, and under the condition that a certain single BGP controller fails, the routing strategies on a repeater are switched and cannot fail, so that the reliability of the BGP controllers is realized.

2. The multi-activity election distribution mechanism, as shown in fig. 2, is specifically as follows:

deploying N (more than or equal to 3) BGP controllers, enabling a single BGP controller to register to a service bus and acquire information of other BGP controllers after starting, if finding that the BGP controller is more than or equal to 3 and no leader controller exists, actively initiating leader requests to all the controllers without responding to the leader requests of other controllers, preferentially replying the controller which receives the first leader request by the rest controllers without initiating the leader requests to ok, actively successfully sending the leader and notifying all the controllers after receiving the leader notification message by the rest controllers, actively issuing a leader message to the leader controllers by all the node controllers, and detecting the active states of all the node controllers by the leader controllers.

After the election is finished, the leader controller receives a request for establishing a BGP peer, 2 controllers are allocated to each repeater device according to the least-priority allocation principle to establish the BGP peer, all the node controllers are notified, all the controllers have the latest allocation information, all the node controllers are notified when the controller is allocated to establish the BGP peer every time, and the controller actively establishes and keeps connection with the repeater device after receiving the notification.

The method comprises the steps that after a Leader controller finds that a certain node controller does not have a keepalive message for 2 times continuously, active dial testing is carried out, if no response is found, the other node controllers are informed to carry out dial testing, if more than half of the node controllers do not have responses in dial testing, the Leader controller redistributes and informs BGP peers of repeater equipment distributed on the controller, and after the BGP peers on the node controllers are built, a router strategy is redistributed to the repeaters.

The principles and spirit of the present invention are explained in detail below with reference to several representative embodiments of the invention.

The invention provides a BGP protocol-based multi-activity concurrency method for realizing policy control, which comprises the following steps:

BGP controller initiated active registration:

configuring the BGP controller name, the service bus address and the management domain of the controller, starting the controller through a script, and realizing the active registration of the controller by calling a registration interface of the service bus through the controller;

after the registration is successful, the controller acquires other controllers in the domain and related roles of the controllers through a service bus, calls other controller notification interfaces in the domain, notifies a new BGP controller to be on-line, and synchronously updates the controller active list by other controllers in the domain;

2. Electing a leader controller:

detecting that the number of BGP controllers is more than or equal to 3, actively calling restful interfaces of leader requests of other controllers, and initiating the leader requests;

the other controllers which do not initiate the leader request preferentially reply to the controller of the received first leader request as ok;

after the controller initiating the leader request receives more than half of the ok messages, the notification interfaces of all the other controllers are called concurrently to become the leader;

the controller becomes a slave node controller after receiving the notification leader message, calls a restful interface of active detection of the leader controller, and sends a keepalive interaction message to the leader controller;

3. The distribution establishes multi-active BGP peers:

detecting that the number of BGP controllers is equal to 2, and actively and synchronously establishing all the BGP peers and all BGP strategies by the new online controller;

and calling the announcement interfaces of the other node controllers by the Leader controller to announce the newly established bgp peer, and updating the local bgp peer list after the node controller receives the notification interfaces.

4. And (3) multi-activity issuing strategy:

after receiving a request for issuing a strategy, the Leader controller searches a local bgp peer body surface;

And 5, Node controller down processing:

if the Leader controller is abnormal in dial test, the other node controllers are informed to carry out dial test, if more than half of the node controllers do not respond to the dial test, a serious alarm is sent out, the node controller is interrupted, the Leader controller informs all the node controllers to delete the peer list of the node controller, and the step 3-4 is triggered again;

And 6, performing down processing on a Leader controller:

and if more than half of the node controllers do not respond to the interaction with the leader controller, re-triggering the step 2-4.

It should be noted that although the operations of the method of the present invention have been described in the above embodiments and the accompanying drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the operations shown must be performed, to achieve the desired results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions.

For a clearer explanation of the above multi-active concurrency method for implementing policy control based on the BGP protocol, a specific embodiment is described below, but it should be noted that the embodiment is only for better illustrating the present invention and is not to be construed as an undue limitation on the present invention.

Taking fig. 3 as an example, it is ensured that the inter-access traffic from the cloud company and the metropolitan area network passes through the backbone network portion as the lowest latency plane, and a low latency traffic path is issued through bgp srv6-policy, which is specifically as follows:

3 controllers are deployed, and leader controllers are elected through the election mechanism introduced above;

the S2 equipment is used as RR and is intended to initiate a bgpls establishing request, after receiving the request, the Leader controller distributes the Leader controller and the node1 controller to establish a bgpls peer with the S2 equipment, and the ISIS routing change information of the network and the link quality delay (4ms) jitter packet loss information of the link are sensed in real time;

s1, S2, C1 and C2 are used as repeater equipment of a head end, a peer of bgp srv6-policy needs to be established with a controller, the peer request of establishing bgp srv6-policy is intended to be initiated, after the leader controller receives the request, the leader controller and S1, S2 and C2 equipment are distributed to establish a peer of bgp srv6-policy, the node1 controller and S2, C1 and C2 equipment are used to establish a peer of bgp srv6-policy, the node2 controller and S1 and C1 equipment are used to establish a peer of bgp srv6-policy, and the multiple activities of the bgp peer are realized;

the intention is calculated according to link quality, a low-delay link from S1 to C1 is S1-C5-C1, a low-delay link from S2 to C2 is S2-C6-C2, after a leader controller receives an intention of issuing a low-delay link strategy, the result of establishing a peer between the controller and forwarding equipment is inquired, the S1 is used as a strategy of head-end forwarding equipment, the leader controller and a node2 controller are notified to concurrently issue, the C1 is used as a strategy of the head-end forwarding equipment, the node1 controller and the node2 controller are notified to concurrently issue, the S2 is used as a strategy of the head-end forwarding equipment, the leader controller and the node1 controller are notified to concurrently issue, the C2 is used as a strategy of the head-end forwarding equipment, the leader controller and a node1 controller are notified to concurrently issue, and multiple activities are achieved for concurrent association.

Based on the same inventive concept, the invention also provides a multi-active concurrency device for realizing policy control based on the BGP protocol. The implementation of the device can be referred to the implementation of the method, and repeated details are not repeated. The term "module," as used below, may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.

As shown in fig. 4, a multi-active concurrency device implementing policy control based on BGP protocol includes:

the BGP controller registration module 101 is configured to register to the service bus after a single BGP controller is started and acquire information of other BGP controllers, and specifically includes:

The leader controller election module 102 is configured to discover that the number of BGP controllers is greater than or equal to 3 and that no leader controller exists, and elect a leader controller, which is specifically as follows:

The multi-active BGP peer distribution establishing module 103 is configured to, after the leader controller receives a request for establishing a BGP peer, perform distribution to establish the multi-active BGP peer, and specifically includes:

The multi-activity issuing policy module 104 is configured to perform a multi-activity issuing policy after the Leader controller receives the request for issuing the policy, and specifically includes:

The Node controller down processing module 105 is configured to perform Node controller down processing after the Leader controller finds that a certain Node controller has no keepalive message to report for 2 consecutive times, and specifically includes the following steps:

The Leader controller down processing module 106 is used for performing Leader controller down processing on the node controller and the Leader controller keepalive without response, and specifically comprises the following steps:

It should be noted that although several modules of a multi-active concurrency device implementing policy control based on the BGP protocol are mentioned in the above detailed description, such partitioning is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the modules described above may be embodied in one module according to embodiments of the invention. Conversely, the features and functions of one module described above may be further divided into embodiments by a plurality of modules.

The multi-activity concurrency method and device for realizing policy control based on the BGP protocol ensure the reliability of the controller by realizing the multi-activity concurrency mechanism of the BGP peer, thereby providing guarantee for network security.

While the spirit and principles of the invention have been described with reference to several particular embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, nor is the division of aspects, which is for convenience only as the features in such aspects may not be combined to benefit. The invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

The limitation of the protection scope of the present invention is understood by those skilled in the art, and various modifications or changes which can be made by those skilled in the art without inventive efforts based on the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A multi-active concurrency method for realizing policy control based on BGP protocol is characterized in that the method comprises the following steps:

2. The BGP protocol-based policy control implementing multi-active concurrency method of claim 1, wherein a single BGP controller initiates a post-registration with a service bus and obtains other BGP controller information, comprising:

3. The BGP protocol-based policy control based multi-active concurrency method of claim 1, wherein electing a leader controller comprises:

4. The BGP protocol-based policy control implementing multi-active concurrency method of claim 1, wherein the distributing and establishing multi-active BGP peers comprises:

5. The BGP protocol-based method of implementing policy control for multi-active concurrency according to claim 1, wherein the multi-active policy issuing includes:

6. The BGP protocol based policy controlled multi-active concurrency method of claim 1, wherein the Node controller down process comprises:

7. The BGP protocol-based policy control multi-active concurrency method of claim 1, wherein the Leader controller down process comprises:

8. A multi-active concurrency device for implementing policy control based on BGP protocol, the device comprising:

9. The BGP protocol-based policy control based multi-active concurrency device of claim 8, wherein the BGP controller registration module is specifically configured to:

10. The BGP-protocol-based policy control based multi-active concurrency device of claim 8, wherein the leader controller election module is specifically configured to:

11. The BGP protocol-based policy control implementing multi-active concurrency device of claim 8, wherein the multi-active BGP peer distribution establishing module is specifically configured to:

12. The BGP protocol-based multi-active concurrency device for implementing policy control according to claim 8, wherein the multi-active delivery policy module is specifically configured to:

13. The BGP protocol-based policy control based multi-active concurrency device of claim 8, wherein the Node controller down processing module is specifically configured to:

14. The BGP-protocol-based policy control based multi-active concurrency device of claim 8, wherein the Leader controller down processing module is specifically configured to: