KR20090013098A - Service Search and Service Failure Detection Method Using UPI - Google Patents

Abstract

The present invention relates to a service discovery and service failure detection method using UPnP, and provides additional functions such as flexible service discovery, service violation operation, and internal service query, and each NMS may independently discover a desired service. It allows you to apply for any service violation report.

In the WBEM-based network management, an object of the present invention is to initialize a UPnP provider including a control point and a device function to advertise or discover an available service, store a newly discovered service in a CIM repository, and UPnP provider. When a user saves a service to the CIM repository, the process of applying for event notification of the current service, if there is a change in the stored service, the NMS notifies by the UPnP event notification mechanism, and the UPnP provider is responsible for changing the service. It includes calling the problem manager to check whether the service change has a negative effect on the current service.

Description

Service discovery and service fault detection method using UF EN {Mathod for service discovery and Fault Management in UPnP based}

The present invention relates to a technique for fast service discovery and service failure recovery using UPnP (Uinversal Plug and Play) in WBEM-based network management, and more specifically, flexible service discovery and service violation operation, internal service query and The same additional function is provided, and each NMS can independently discover the desired service and subscribe to the service for reporting any service violation, and use the UPnP service discovery and service failure detection method in WBEM-based network management. It is about.

[SLP (Service Location Protocol)]

SLP can eliminate the need for the user to know the name of the network host. With SLP, the user only needs to describe the service of interest. Based on the description of the service, the SLP may pass the URL of the requested service to the user. SLP is largely defined by three kinds of agents. A software entity looking for the location of one or more messages is called a user agent (UA) 102 and a software entity advertising the location of one or more services is called a service agent (SA). Finally, a software entity that acts as a repository for service location information is called a directory agent (DA). SA and UA prioritize to indicate available DA.

Simple Network Management Protocol

The SNMP traps basically asynchronously notify the network management system 201 from the SNMP agent 202. The trap PDU (Protocol Data Unit) consists of enough information about the failure that an SNMP trap can represent. Like other messages in SNMP 203, traps also send information using UDP 204. A trap is a bundle of data defined in a managed information base (MIB).

[UPnP (Universal Plug and Play) Structure]

The UPnP architecture relies on P2P networking of PCs, applications, and wireless devices. It is a distributed architecture based on TCP / IP, UDP and HTTP. This is a standardized and defined collection for using networked devices in a plug-and-play method without manual configuration or configuration. It is designed to provide auto-discovery and "invisible" networking for a variety of devices from a variety of vendors without any configuration. That is, device 302 can actively participate in the network, obtain an IP address, and know the characteristics or status of other devices.

[Universal Plug and Play (UPnP)]

UPnP is a collection of computer network protocols published by the UPnP forum. The purpose of UPnP is to make it easy to use the network at home or in common areas and to provide seamless connectivity. UPnP has established the UPnP device control protocol in its Internet-based communication standard.

The fault management system based on SNMP trap notification technology has limitations in defining faults in the physical and multi-protocol label switching (MPLS) layers. And, the part about failure recovery is not easy to deal with by SNMP failure detection mechanism. In addition, there is currently no MOF available for service recovery or UPnP in CIM.

Fast fault recovery is achieved by the immediate detection of a fault that occurs. In some cases, in order to provide fast failover / blocking, it is sometimes more difficult to define the cause of the failure. In addition, service class failures such as Service Level Agreements (SLAs) may be due to physical device failures.

The present invention is to solve the conventional problems as described above, the object of the present invention is to provide additional functions, such as flexible service discovery, service violation operation, internal service query, each NMS is independently of the desired service It is to provide a service discovery and service failure detection method using UPnP in WBEM-based network management that can discover or subscribe to any service violation report.

In order to achieve the object of the present invention as described above, the service discovery and service failure detection process using UPnP in the WBEM-based network management according to the present invention, in the WBEM-based network management, the control point (Control Point) and the device The UPnP provider including a client function may include a first process of initializing to advertise or discover an available service, a second process of storing the newly found service in a CIM repository, and a service of the UPnP provider in the CIM repository. A third process of subscribing for event notification of the current service when storing the service, a fourth process of NMS notifying by the UPnP event notification mechanism if there is a change in the stored service, and UPnP for the change of the service. The provider calls the WBEM server's fault manager to indicate that the service change is currently in use. And a fifth step of determining with respect to how the adverse effect on the service is characterized in that is made.

According to the present invention as described above, the use of UPnP in network management supplements the disadvantages of network management by providing functions of service management such as service discovery, continuous monitoring, return of service technology, and acquisition of control value without service call.

The UPnP-based service discovery and service management architecture proposed in the present invention can be applied in network management, and this application can detect and manage a drop in SLA to provide fast operation of service violations and to provide quality DiffServ-over-MPLS. Make it possible.

Hereinafter, with reference to the accompanying drawings an embodiment of the present invention will be described in more detail.

4 is a functional block diagram of the UPnP-based WBEM server structure for TE between ASs. UPnP provider 401 is a new model integrated in the WBEM server. The UPnP provider 401 includes control point 402 and device 423 functions, and initializes the available services (SSDP: search, SSDP: advertisement) for advertisement or discovery.

The newly found service is stored in the CIM repository 430. If a service already exists in the CIM repository 430, an expiration time (ExpireTime), which is one of service attributes, is newly updated.

When the UPnP provider 401 stores a service, it subscribes for event notification of the current service.

Thus, if there are any changes in the service, the NMS will use the UPnP event notification mechanism to detect this, and the UPnP provider 401 will call the WBEM server's fault manager to see if this change has a negative impact on the current service. Check against.

Herein, the UPnP provider 401 will be described in more detail. The UPnP provider 401 is used to store a service discovery, a subscription for service event notification, and a service description.

When an SSDP discovery (advertisement or response) message arrives, information about the service description is obtained from the server. Then, it is checked whether or not the information already exists in the CIM storage 430.

If the found information is not yet stored information, the UPnP provider 401 subscribes for a UPnP event and stores all the information about the service description in the CIM repository 430. This information includes service ID, service location, type of service, service life time and event subscription ID of the service.

This service description information can then be used through the SLA provider to establish SLA negotiations. This is because the service information is updated periodically. If the found service already exists in the repository, UPnP provider 401 only plays the service end time.

If the parameters of a service have been changed via the service provider NMS, a UPnP event notification message will be sent to all subscribed control points 402. Of course, if the service is not needed by anyone, the UPnP provider 401 will cancel the application.

5 is a diagram illustrating a UPnP-based NMS structure and functionality according to an embodiment of the present invention. In order to support active service and resource discovery, the UPnP provider 401 structure is included in the WBEM.

The UPnP provider 401 is newly added among the providers of the NMS server 404, and the functionality of the client advertising the control point 402 and the provided service 403 can be supported.

When adding a new service to the NMS 404, the UPnP discovery protocol allows the NMS to advertise 405 to the control point 402 of the network.

Similarly, when a control point 401 is added to the network, the UPnP discovery protocol allows the control point 402 to search for a device in the network 406 and return a result 407.

In both cases, an important exchange is a search message containing the necessary information of the service or device.

Advertise multicasts these discovery messages through standard addresses and ports. The control point 402 always watches for messages coming through this port.

When a service is about to be removed from the network, it multicasts the message 'ssdp: byebye'.

If the NMS 404 is suddenly deleted from the network, it may not be possible to multicast the message. If so, include the expiration value in the 'CACHE-CONTROL' header in the search message.

After the control point 402 is included in the network, the multicast request is sent using the 'M-SEARCH' method (408).

The UPnP description for the service is divided into two logical parts. The first NMS description 409 represents the physical and logical container, and the second service description 410 represents the characteristic.

The description is in XML syntax and is based on the standard UPnP device template. The UPnP service description contains a list of commands, actions for service responses, or parameters for each action.

As with discovery, the description plays a significant role in the interoperability of the control point 402 and the NMS 403 using different versions of UPnP networking.

To invoke an action in the NMS 403 service, the control point 402 sends an appropriate action request 411 to the service's control URL. In response, the service returns any result 413.

The effect of this behavior is to change the variables that represent the state of the service. To determine the value of the variable in its current state, control point 402 examines the service.

In order to receive an event or notification message associated with the service, UPnP provider 401 makes a subscription request 414 for the service, and returns the proof ID 415. The event message contains the names of one or more state variables and the values of those variables now in XML format.

The prover can send or receive all event messages.

If the responder does not receive a response to the event message, the publisher continues to send the event message to the prover until the attestation timeout expires (417).

If an event message is received, the UPnP provider 401 calls the fault management module's checkForSLADegradation () function to check for service degradation.

Each domain's NMS is equipped with WBEM server and client functions for traffic engineering between domains. However, UPnP provider 401 uses its own protocol stack for sending or receiving UPnP messages for event notification.

MOF is implemented in C ++ objects and provided by an instance provider. The UPnP control point 402 functionality was implemented using the libupnp API to integrate in the WBEM architecture.

Currently implemented UPnP provider 401 has service discovery and SLA drop operation functions, and external control functions including control functions of control point 402 such as StartCPService (), StopCPService (), and DeleteService (). There is a function that can be inherited from a provider.

In addition, UPnP provider 401 inherits from instance provider and association provider to provide enumeration and association traversal of newly created QoS_UpnpNMSServiceSetting class. Receive.

  The UPnP provider also provides device functionality, such as advertising services, sending event notification messages to subscribed providers, and the like.

Referring to Figure 6 with respect to the procedure of such a function as follows.

FIG. 6 is a flow chart of a UPnP service discovery and event mechanism that performs the functions of FIG. 5. Due to the same functions as FIG. 5, while searching for a new service, UPnP provider 401 may advertise its service. have.

When the UPnP provider 401 receives the event notification message, the UPnP provider 401 calls the fault manager module's checkForSLADegradation () function.

All information about the service is stored in the CIM repository 430. Thus, other UPnP providers 401 may access for further use, such as SLA negotiation, connection establishment, and the like.

UPnP's event notification mechanism can also be used in SLA / SLS management, particularly for defining service management and service violation operations.

The UPnP provider discovers the service, stores the proof for service event notification, and the service description.

The service search and service failure detection method using UPnP according to the present invention described above is not limited to the above-described embodiment, and in the field to which the present invention belongs without departing from the gist of the present invention as claimed in the following claims. Anyone with ordinary knowledge has the technical spirit to the extent that various changes can be made.

1 illustrates a typical SLP structure.

2 illustrates a general SNMP architecture.

3 is a diagram illustrating a general UPnP structure.

4 is a functional block diagram of a UPnP-based WBEM server structure according to the present invention.

5 is a diagram illustrating a UPnP-based NMS structure and functionality according to the present invention.

6 is a process diagram for implementing the functions of FIG.

Claims (7)

In the service discovery and service failure detection method using UPnP in WBEM-based network management that performs flexible service discovery, service forgery and internal service query functions, The UPnP provider including a control point and a client function may include a first step of initializing to advertise or discover an available service; A second step of storing the newly found service in a CIM repository; A third step of subscribing for event notification of a current service when the UPnP provider stores a service in the CIM repository; A fourth step of NMS notifying a UPnP event notification mechanism if there is a change in the stored service; In response to the service change, the UPnP provider calls a failure manager module of the WBEM server to confirm whether the service change negatively affects the currently used service. Service search and service failure detection method using. The method of claim 1, The UPnP provider is a service discovery and service failure detection method using the UPnP, characterized in that for performing the function of storing the service (subscribe) for the service discovery, service event notification (subscribe) and service description. The method of claim 1, If the service stored in the CIM repository already exists in the second process in the second process, an expiration time (ExpireTime), which is one of service attributes, is newly updated. Detection method. The method of claim 1, In the third process, the information about the service description stored in the CIM repository is periodically updated, and UPnP is characterized by including a service ID, a service location, a service type, a service life time, and an event application ID of the service. Service search and service failure detection method using. The method of claim 4, wherein The service description information is a service search and service failure detection method using UPnP, characterized in that it is used through the SLA provider to establish the SLA negotiation. The method of claim 1, In addition to the third process, if a parameter of a service has been changed through the service provider NMS, a UPnP event notification message is sent to all the applied control points. If the service is not needed by anyone, the UPnP provider applies for the application. Service search and service failure detection method using UPnP, characterized in that it further comprises the step of canceling. The method of claim 1, The fifth process is a service search and service failure detection method using UPnP, characterized in that to determine whether the service level agreement (SLA) has dropped by calling the failure manager module of the WBEM server in response to the event notification message.

Images