WO2012100571A1

WO2012100571A1 - Centralized management method and system for multiple tunnels with the same path

Info

Publication number: WO2012100571A1
Application number: PCT/CN2011/081291
Authority: WO
Inventors: 刘必振; 邱军; 金飞蔡
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-01-24
Filing date: 2011-10-25
Publication date: 2012-08-02
Also published as: CN102611610A; CN102611610B

Abstract

Disclosed in the invention is a centralized management method for multiple tunnels with the same path. The method comprises: creating a tunnel cluster for multiple tunnels with the same path at the two endpoint devices of the tunnel, bundling the tunnel cluster to a maintenance entity group(MEG), the MEG informing the tunnel cluster of current detected status when a physical link failure occurs, and the tunnel cluster performing a batch process according to the detected status. Also disclosed in the invention is a centralized management system for multiple tunnels with the same path. With the method and the system, unnecessary maintenance message transmission can be reduced in the network, the link failure status can be uniformly informed to management layer in a group way, and it can be avoided to print the same alerts for every MEG, thus reducing the burden of the devices.

Description

Method and system for centralized management of multiple same path tunnels

The present invention relates to a multi-protocol label switching/Internet Protocol (MPLS/IP) technology in network communication, and more particularly to a method and system for centralized management of multiple tunnels of the same path. Background technique

The most basic goal of the evolution of telecom transport network is to reduce the transmission cost of information in communication. However, with the expansion of packet access bandwidth and the deployment of Next Generation Network (NGN), time-division channelized transmission is no longer completely Meet this basic need. The connection-oriented transport mechanism in a large number of packet services makes channel utilization difficult to increase, so that the information transmission cost cannot be reduced. In order to solve this problem, Transmission-Multi Protocol Label Switching (T-MPLS) has emerged under the dual driving of the market and services.

T-MPLS technology is a Packet Transport Network (PTN) transmission technology standardized by the International Telecommunication Union (ITU-T). It is characterized by a connection-oriented packet transmission technology, and its data forwarding plane is multi-protocol. A subset of Multi Protocol Label Switching (MPLS), which is forwarded based on T-MPLS labels. It simplifies some of the complex functions of the MPLS data forwarding plane, removes the IP-based connectionless forwarding feature, and adds connection-oriented operation, management, maintenance (Operation, Administration, Maintenance, ΟΑΜ) functions and protection recovery. The function is to use the Automatically Switched Optical Network/Generalized Multiprotocol Label Switching (ASON/GMPLS) as its control plane. The "connection-oriented" feature means that T-MPLS technology can provide telecom carrier-grade services that require traditional delivery. Net equivalent OAM function. Therefore, OAM function becomes an indispensable key link of T-MPLS technology.

OAM is configured based on the Maintenance Entity Group (MEG), that is, a T-MPLS port is deployed on a tunnel. Such a deployment can monitor the status of the tunnel, so that the user can be notified in time when the tunnel is abnormal, for example: printing an alarm on the user configuration interface and the network management interface; taking corresponding measures to detect, diagnose, and Positioning and fault recovery, etc., for example: Tunnel for protection switching or tunnel advertising upper layer application.

The T-MPLS mechanism has the following characteristics: 1. Each ΟΑΜ detection management entity needs to specify T-MPLS connection support; 2. Support client layer/service layer ΟΑΜ relationship, and need corresponding configuration mode at each level; The service provider shall be able to automatically detect and notify the faults related to the maintenance entity; 4. The corresponding measures must be taken when the fault occurs.

In actual applications, because of different services or different user requirements, multiple tunnels of the same path may exist for the same two endpoints. The corresponding detection management entity group needs to be deployed on the corresponding tunnel for end-to-end detection. Features. The MEG of the same path is generated on the corresponding T-MPLS scope. This deployment increases the redundancy of the OAM Protocol Data Unit (PDU) in the network and appears on the network of the path. When an abnormality occurs, each MEG needs to print an alarm on the user configuration interface and the network management interface, which increases the burden on the device. Summary of the invention

In view of the above, the main object of the present invention is to provide a method and system for centralized management of multiple tunnels of the same path, which can solve the same function of the MEG on the same path, and in the case of network failure, redundant message transmission avoids each The MEG prints the same warning.

In order to achieve the above object, the technical solution of the present invention is achieved as follows: The present invention provides a method for centralized management of multiple tunnels of the same path, and the method includes:

Create a tunnel cluster for multiple tunnels of the same path on the two endpoint devices of the tunnel; bind the tunnel cluster to the detection management entity group;

When the physical link is abnormal, the management entity group notifies the tunnel cluster of the current detection status, and the tunnel group completes batch processing according to the detection status.

After the tunnel cluster is created, the method further includes: when a new tunnel is added to the tunnel cluster, determining whether the existing tunnel information in the tunnel cluster matches, and if yes, adding, otherwise the adding fails.

The tunnel cluster includes: a tunnel cluster number, a tunnel cluster type, a tunnel cluster management entity binding interface, a tunnel number statistics, a tunnel cluster state, and a tunnel information database.

The tunnel cluster type includes: a tunnel cluster detected by the layer, and a tunnel cluster detected by the lower layer.

The batch processing includes: the tunnel cluster notifies the tunnel included in the tunnel information database one by one, switches the tunnel with protection, and prints an alarm on the device.

The present invention also provides a system for centralized management of multiple tunnels of the same path, where the system includes: a tunnel endpoint, a detection management entity group;

The tunnel endpoint is configured to create a tunnel cluster for multiple tunnels of the same path, and the tunnel cluster on the tunnel endpoint completes batch processing according to the received detection state;

The detection management entity group is configured to bind the tunnel cluster. When the physical link is abnormal, the management entity group sends the current detection status to the tunnel endpoint.

The tunnel endpoint is further configured to: when the new tunnel needs to join the tunnel cluster, determine whether the existing tunnel information in the newly added tunnel and the tunnel cluster match, and if yes, join, otherwise add failed.

The tunnel cluster in the tunnel endpoint includes: a tunnel cluster number, a tunnel cluster type, The tunnel cluster management entity binds the interface, the number of tunnels, the tunnel cluster status, and the tunnel information database.

The tunnel endpoint batch processing includes: the tunnel cluster notifying the tunnel included in the tunnel information database one by one, switching the tunnel with protection, and printing on the device the centralized management of multiple identical paths provided by the present invention Method and system, on the two endpoint devices of the tunnel, create a tunnel cluster for multiple tunnels of the same path; bind the tunnel cluster to the detection management entity group; when the physical link is abnormal, the management entity group will detect the current state The tunnel cluster is notified, and the tunnel cluster completes batch processing according to the detection status. It can reduce the unnecessary T-MPLS OAM packets on the network. In the case of abnormal link conditions, the management layer is unified in the cluster mode. This prevents each MEG from printing the same warning and reduces the burden on the device. DRAWINGS

1 is a schematic structural diagram of an application scenario of a method for centralized management of multiple tunnels of the same path according to the present invention; FIG. 2 is a schematic flowchart of a method for centralized management of multiple tunnels of the same path according to the present invention; FIG. 3 is a schematic diagram of a current state of detection of a management entity group according to the present invention; Schematic diagram of the method of notifying the tunnel cluster;

FIG. 4 is a schematic structural diagram of a system for centralized management of multiple tunnels of the same path according to the present invention. detailed description

The basic idea of the present invention is to create a tunnel cluster for multiple tunnels of the same path on two endpoint devices of the tunnel; bind the tunnel cluster to the detection management entity group; when the physical link is abnormal, the management entity group will detect the current The status informs the tunnel cluster that the tunnel cluster completes the batch processing based on the detected status. In order to better understand the present invention, the application scenario of the present invention will be first introduced. FIG. 1 is a schematic structural diagram of an application scenario of a method for centralized management of multiple tunnels of the same path according to the present invention. As shown in FIG. 1, four tunnels between the device 101 and the device 108 are tunnel 1, tunnel 2, tunnel 3, and tunnel 4, respectively. Static tunnels pass through device 102, device 104, and device 106 at the same time. These tunnels are responsible for different services, such as: tunnel 1 is responsible for network telephony services, tunnel 2 is responsible for video data transmission services, tunnel 3 is responsible for fixed-line services, and tunnel 4 is responsible for point-to-point (BT) download services. However, the working paths of the four tunnels are identical and conform to the basic scenario of the present invention. Thus a tunnel cluster 1 can be created between device 101 and device 108. For a static tunnel, the MEG deployment only generates a corresponding response at the head node of the tunnel. Therefore, the tunnel cluster only needs to be deployed on the two endpoints of the corresponding tunnel, and the device 101 and the device 108.

The technical solutions of the present invention are further elaborated below in conjunction with the accompanying drawings and specific embodiments. FIG. 2 is a schematic flowchart of a method for centralized management of multiple tunnels of the same path according to the present invention. As shown in FIG. 2, the method specifically includes:

Step 201: Create a tunnel cluster for multiple tunnels of the same path on the two endpoint devices of the tunnel.

Specifically, the tunnel cluster includes: a tunnel cluster number, a tunnel cluster type, a tunnel cluster management entity binding interface, a tunnel number statistics, a tunnel cluster state, and a tunnel information database. The tunnel cluster number is used as an index of the tunnel cluster, and the tunnel device assigns a tunnel cluster number to the cluster when the tunnel cluster is created; the tunnel cluster type is used to specify the type of the tunnel cluster. There are two types of clusters, one for layer detection of tunnel clusters.

(TMP Detect), which is the tunnel cluster detected by this layer, and the other is the TMS Detect for the tunnel cluster, which is the tunnel cluster of the lower layer detection; the tunnel cluster management entity binding interface is used as the tunnel cluster. The entry of the notification of the status change. If the tunnel cluster type is for the T-MPLS Path (TMP), the specified detection mode needs to be bound in the cluster layer if the tunnel cluster type is for the segment layer (T-MPLS). Section, TMS), you need to be in the lower layer of the cluster The corresponding TMS is bound to the specified tunnel cluster; the number of tunnels is used to record how many tunnels have been bound in the current tunnel cluster; the tunnel cluster state is used to record the state of the current tunnel cluster, and the tunnel cluster There are four states, namely, signal failure alarm (SF), signal degradation alarm (SD), signal failure alarm (SF-SD) after signal degradation, and signal alarm normal (OK); the tunnel information database is used. The index of the tunnel information that has been bound in the current tunnel cluster is recorded, so that all the tunnels in the cluster are notified when the state of the tunnel cluster changes.

Step 202: Bind the tunnel cluster to the detection management entity group.

Specifically, the binding is specifically performed by the tunnel cluster management entity binding interface, and the tunneling cluster and the management entity group of each tunnel in the tunnel cluster are bound. Further, you need to configure the tunnel cluster type as required. If you need to respond to the TMP tunnel cluster, configure the local layer detection. If you need to respond to the TMS tunnel cluster, configure the lower layer detection. In the present invention, the two tunnel clusters are mutually exclusive and therefore need to be separately configured. The detection management entity group includes: a local layer detection module, configured to detect the current state of the connection path of the devices at both ends, and implement the OAM message transmission and reception between the devices through the protocol. The protocol may be a detection protocol such as TMPLS OAM or BFD. After the local layer detection module bound to the tunnel cluster has the corresponding relationship, the detection status can be correctly reported. The other is the lower layer detection module, which notifies the lower layer detection alarm for the lower layer detection alarm of the TMP. The lower layer detection entity of the current tunnel cluster is the TMS, and the tunnel cluster needs to respond to the alarm state sent by the lower layer detection module. The outbound interface of the TMS and the outbound interface corresponding to the tunnel are the same, and the TMS is registered through the tunnel cluster management entity binding interface. After the corresponding lower-layer notification relationship is established, the detection status of the lower layer can be correctly reported.

Step 203: When the physical link is abnormal, the management entity group notifies the tunnel cluster of the current detection state, and the tunnel cluster completes batch processing according to the detection state.

Specifically, the detection status is specifically sent in the form of alarm information. The tunnel cluster receives the detection status, that is, the alarm information, and then calculates the corresponding tunnel cluster alarm status, and then compares it with the existing alarm status. If the alarm status is the same as the current one, the tunnel cluster is not advertised. In the tunnel, the process ends. If the alarm status is different from the existing alarm status, the alarm status of the tunnel cluster is updated, and all tunnels in the tunnel cluster are advertised, the tunnel status is updated, and batch processing is completed. The batch processing specifically includes: the tunnel cluster notifies the tunnel included in the tunnel information database one by one, switches the tunnel with protection, and prints an alarm on the device.

Further, after the tunnel cluster is created, the method further includes: when a new tunnel needs to join the tunnel cluster, determine whether the existing tunnel information in the tunnel cluster matches, and if yes, add, otherwise the addition fails.

Specifically, when the tunnel joins the cluster, it first checks whether the tunnel cluster contains information about other tunnels. If there are no other tunnels in the cluster, directly join the tunnel to the cluster. If the tunnel already contains the tunnel, take the first tunnel and need it. The added tunnel compares the basic information of the two tunnels. If the mode (dynamic static), type (single bidirectional), outgoing interface, and next hop are the same, they are added to the tunnel cluster. For the dynamic tunnel, the tunnel is also required. Display route

(ERO) checks, if not, the addition fails, and the return user fails to be added. The above method can only determine whether the local end of the tunnel is consistent for a static tunnel, but cannot guarantee the same path. The specific topology needs to be guaranteed by the configurator.

Further, the tunnel cluster does not need to care about the current state of the tunnel in the process of updating the tunnel. When the tunnel is added to the tunnel protection group, the alarm state is directly updated to the current state of the tunnel cluster, and the tunnel is always maintained throughout the process. The alarm status of the cluster is the same. When the tunnel leaves the tunnel cluster, the tunnel cluster will cancel the current alarm state of the tunnel, and the initial value of the alarm state of the tunnel and the new alarm state will be re-notified by the MEG on the tunnel. After the tunnel obtains a new alarm status, the status is recorded. If the tunnel is not in the tunnel protection group and there is a corresponding upward advertised configuration, the tunnel advertises a new alarm state to the upper layer TMC. Before the tunnel is added to the tunnel cluster, if the state of the tunnel is different from the tunnel cluster, the tunnel cluster advertises the tunnel update alarm status. If the tunnel exists in the tunnel protection group, the tunnel notifies the protection group of the new alarm state, and the protection group calculates the new protection group switching state according to the corresponding protection policy. The row corresponds to the operation. If there is a corresponding upward advertisement configuration on the tunnel, the tunnel will notify the upper TMC of the new alarm status. Before the tunnel is added to the tunnel cluster, if the state of the tunnel is different from the tunnel cluster, the tunnel cluster advertises the tunnel update alarm status.

FIG. 3 is a schematic flowchart of a method for a management entity group to notify a tunnel cluster of a current detection state according to the present invention. As shown in FIG. 3, the method includes:

Step 301, when the physical link is abnormal, the detecting module on the link calculates the corresponding alarm and reports it, if it is the layer detecting module, step 302 is performed; if it is the lower layer detecting module, step 304 is performed;

Specifically, the abnormality may be a packet failure abnormality caused by fiber extraction, a closed port, or a lost route, or may be an abnormality caused by a packet signal degradation caused by other interferences and an error rate being too high.

Step 302: Determine whether the management entity group has a binding relationship with the tunnel cluster. If the binding is established, go to step 303. If there is no binding, process it according to the prior art.

Step 303: Determine whether the type of the tunnel cluster is detected by the layer. If yes, send the alarm information to the tunnel cluster, and end the process. If the tunnel cluster type is set to the lower layer detection, the tunnel cluster does not respond.

Step 304: The lower layer detecting module updates its own alarm state, and sends the alarm information to the tunnel cluster.

Specifically, after receiving the alarm, the TMS searches for the upward notification object of the TMS. When the TMS uplink notification interface is bound to the tunnel cluster management entity binding interface and the tunnel cluster type is the lower layer detection type, the alarm information is sent. Send to the tunnel cluster.

4 is a schematic structural diagram of a system for centralized management of multiple tunnels of the same path according to the present invention. As shown in FIG. 4, the system includes: a tunnel endpoint 41, a detection management entity group 42;

The tunnel endpoint 41 is configured to create a tunnel cluster for multiple tunnels of the same path, and the tunnel cluster on the tunnel endpoint completes batch processing according to the received detection state; Specifically, the tunnel cluster includes: a tunnel cluster number, a tunnel cluster type, a tunnel cluster management entity binding interface, a tunnel number statistics, a tunnel cluster state, and a tunnel information database. The tunnel cluster number is used as an index of the tunnel cluster. When the tunnel cluster is created, the tunnel endpoint 41 assigns a tunnel cluster number to the cluster; the tunnel cluster type is used to specify the type of the tunnel cluster. There are two types of clusters, one for the layer detection of the tunnel cluster, that is, the tunnel cluster for the layer detection, and the other for the lower layer detection of the tunnel cluster, that is, the tunnel cluster for the lower layer detection; An entity binding interface is used as an entry point for notification of a tunnel cluster state change. If the tunnel cluster type is for TMP, the specified detection mode needs to be bound in the cluster layer. If the tunnel cluster type is for TMS, Binding the specified tunnel cluster to the TMS corresponding to the lower layer of the cluster; the number of the tunnels is used to record the number of tunnels that have been bound in the current tunnel cluster; the tunnel cluster state is used to record the status of the current tunnel cluster. There are four states of the tunnel cluster, which are signal failure alarm (SF) and signal degradation alarm (SD). After the signal is degraded, The signal failure alarm (SF-SD) is generated, and the signal alarm is normal (OK); the tunnel information database is used to record the index of the tunnel information already bound in the current tunnel cluster, so as to notify the cluster when the state of the tunnel cluster changes. All the tunnels.

The detection status is specifically sent in the form of alarm information. After the tunnel cluster receives the detection status, that is, the alarm information, the corresponding tunnel cluster alarm status is calculated, and then compared with the existing alarm status. If the alarm status is the same as the current one, the tunnel in the tunnel cluster is not advertised. If the alarm status is different from the existing alarm status, update the alarm status of the tunnel cluster, and advertise all the tunnels in the tunnel cluster, update the tunnel status, and complete batch processing. The batch processing specifically includes: the tunnel cluster notifies the tunnel included in the tunnel information database one by one, switches the tunnel with protection, and prints an alarm on the device.

Further, the tunnel cluster does not need to care about the current state of the tunnel in the process of updating the tunnel. When the tunnel is added to the tunnel protection group, the alarm state is directly updated to the current state of the tunnel cluster, and the tunnel is always maintained throughout the process. The alarm status of the cluster is the same. When the tunnel leaves After the tunnel is clustered, the tunnel cluster will cancel the current alarm state of the tunnel, and the initial value of the alarm state of the tunnel and the new alarm state will be re-notified by the MEG on the tunnel. After the tunnel obtains a new alarm status, the status is recorded. If the tunnel is not in the tunnel protection group and there is a corresponding upward advertised configuration, the tunnel advertises a new alarm state to the upper layer TMC. Before the tunnel is added to the tunnel cluster, if the state of the tunnel is different from the tunnel cluster, the tunnel cluster advertises the tunnel update alarm status. If the tunnel exists in the tunnel protection group, the tunnel notifies the protection group of the new alarm state. The protection group calculates the new protection group switching state and performs corresponding operations according to the corresponding protection policy. If there is a corresponding upward advertisement configuration on the tunnel, the tunnel will notify the upper TMC of the new alarm status. Before the tunnel is added to the tunnel cluster, if the state of the tunnel is different from the tunnel cluster, the tunnel cluster advertises the tunnel update alarm status.

The detection management entity group 42 is configured to bind the tunnel cluster. When the physical link is abnormal, the management entity group 42 sends the current detection status to the tunnel endpoint 41.

Specifically, the binding is specifically binding the management entity group of each tunnel in the tunnel cluster and the tunnel cluster by using the tunnel cluster management entity binding interface. Further, you need to configure the tunnel cluster type as required. If you need to respond to the TMP tunnel cluster, configure the local layer detection. If you need to respond to the TMS tunnel cluster, configure the lower layer detection. In the present invention, the two tunnel clusters are mutually exclusive and therefore need to be separately configured. The detection management entity group includes: a layer detection module, configured to detect the current state of the connection path of the devices at both ends, and implement the OAM packet transmission and reception between the devices through the protocol. The protocol may be a detection protocol such as TMPLS OAM or BFD. After the local layer detection module bound to the tunnel cluster has the corresponding relationship, the detection status can be correctly reported. The other is the lower layer detection module, which notifies the lower layer detection alarm for the lower layer detection alarm of the TMP. The lower layer detection entity of the current tunnel cluster is the TMS, and the tunnel cluster needs to respond to the alarm state sent by the lower layer detection module. The outbound interface of the TMS and the outbound interface corresponding to the tunnel are the same, and the TMS is registered through the tunnel cluster management entity binding interface. After the corresponding lower-layer notification relationship is established, the detection status of the lower layer can be correctly reported. Further, the tunnel endpoint 41 is further configured to determine, when the new tunnel needs to join the tunnel cluster, whether the new tunnel joins the existing tunnel information in the tunnel cluster, and if yes, join, Otherwise the addition fails.

Specifically, when the tunnel joins the cluster, it first checks whether the tunnel cluster contains information about other tunnels. If there are no other tunnels in the cluster, directly join the tunnel to the cluster. If the tunnel already contains the tunnel, take the first tunnel and need it. The added tunnel compares the basic information of the two tunnels. If the mode (dynamic static), type (single bidirectional), outgoing interface, and next hop are the same, they are added to the tunnel cluster. For the dynamic tunnel, the tunnel is also required. The display route is checked. If it is not the same, the addition fails. The return user fails to be added. The above method can only determine whether the local end of the tunnel is consistent for a static tunnel, but cannot guarantee the same path. The specific topology needs to be guaranteed by the configurator.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Claims

Claim

A method for centralized management of multiple tunnels of the same path, wherein the method includes:

The method according to claim 1, wherein after the tunnel cluster is created, the method further includes:

When a new tunnel joins the tunnel cluster, it is judged whether it matches the existing tunnel information in the tunnel cluster. If it matches, it joins, otherwise the addition fails.

The method according to claim 1 or 2, wherein the tunnel cluster comprises: a tunnel cluster number, a tunnel cluster type, a tunnel cluster management entity binding interface, a tunnel number statistics, a tunnel cluster state, and a tunnel information. database.

4. The method according to claim 3, wherein the tunnel cluster type comprises: a tunnel cluster detected by the layer, and a tunnel cluster detected by the lower layer.

The method according to claim 1 or 2, wherein the batch processing comprises: the tunnel cluster notifying the tunnel included in the tunnel information database one by one, switching the tunnel with protection and printing on the device Alarm.

A system for centrally managing a plurality of tunnels of the same path, wherein the system includes: a tunnel endpoint, a detection management entity group;

The system according to claim 6, wherein the tunnel endpoint is further configured to determine, when the new tunnel needs to join the tunnel cluster, the new tunnel and the cluster cluster Whether some tunnel information matches, if it matches, join, otherwise the addition fails.

The system according to claim 6 or 7, wherein the tunnel cluster in the tunnel endpoint comprises: a tunnel cluster number, a tunnel cluster type, a tunnel cluster management entity binding interface, a tunnel number statistics, and a tunnel cluster. Status, tunnel information database.

9. The system according to claim 8, wherein the tunnel cluster type comprises: a tunnel cluster detected by the layer, and a tunnel cluster detected by the lower layer.

The system according to claim 6 or 7, wherein the tunnel endpoint batch processing comprises: the tunnel cluster notifying the tunnel included in the tunnel information database one by one, switching the tunnel with protection, and the device Print an alarm on it.