Abstract: A service request for a managed computer system is received and once a primary authorization for same has been given, a secondary authorization management system (SAMS) determines whether or not the service request requires secondary authorization. This determination is made according to a context of the managed computer system and an authorization profile for the received service request. If needed, the SAMS resolves the secondary authorization request and returns the resolution decision.

Abstract: Data is often populated into Configuration Management Databases (CMDBs) from different sources. Because the data can come from a variety of sources, it may have inconsistencies—and may even be incomplete. A Normalization Engine (NE) may be able to automatically clean up the incoming data based on certain rules and knowledge. In one embodiment, the NE takes each Configuration Item (CI) or group of CIs that are to be normalized and applies a rule or a set of rules to see if the data may be cleaned up, and, if so, updates the CI or group of CIs accordingly. In particular, one embodiment may allow for the CI's data to be normalized by doing a look up against a Product Catalog and/or an Alias Catalog. In another embodiment, the NE architecture could be fully extensible, allowing for the creation of custom, rules-based plug-ins by users and/or third parties.

Abstract: Data is often populated into Configuration Management Databases (CMDBs) from different sources. Because the data can come from a variety of sources, it may have inconsistencies—and may even be incomplete. A Normalization Engine (NE) may be able to automatically clean up the incoming data based on certain rules and knowledge. In one embodiment, the NE takes each Configuration Item (CI) or group of CIs that are to be normalized and applies a rule or a set of rules to see if the data may be cleaned up, and, if so, updates the CI or group of CIs accordingly. In particular, one embodiment may allow for the CI's data to be normalized by doing a look up against a Product Catalog and/or an Alias Catalog. In another embodiment, the NE architecture could be fully extensible, allowing for the creation of custom, rules-based plug-ins by users and/or third parties.

Abstract: Disclosed are methods and systems to provide coordinated identification of data items across a plurality of distributed data storage repositories (datastores). In one disclosed embodiment, a single configuration management database (CMDB) controls identification rights for all CIs as they are first identified in a master/slave relationship with all other CMDBs in the distributed environment. In a second embodiment, a plurality of CMDBs divide identification rights based upon coordination identification rules where certain CMDBs are assigned authoritative identification rights for CIs matching the rules of a particular CMDB in the distributed environment. In a third embodiment, one or more of the plurality of CMDBs may also have advisory identification rights for CIs which do not already have an identifiable unique identity and can coordinate with an authoritative CMDB to establish an identity for CIs.

Abstract: This disclosure relates generally to the field of Configuration Management Databases (CMDBs). One embodiment of a user interface embodying the present invention is an extension of the process for creating CMDB classes and is therefore readily available for use by someone with knowledge of CMDB administration. The CMDB administrator is thus relieved from having to understand in detail the technologies and interfaces used by the Management Data Repository (MDR) sources. The result of setting up a relation from a CMDB data structure to an MDR data structure by a CMDB administrator may be represented by one or more new CMDB class(es) for the MDR data. The related MDR may then be accessed by an existing CMDB application using already existing CMDB interfaces. The instances of the new relationships and classes thus appear as if they were native instances stored in the CMDB.

Abstract: A mechanism is disclosed for enabling new task types to be added to a management server without changing the source code of the management server. To add a new task type, a descriptor file is provided. This descriptor file specifies all of the particulars of the new task type. Any mechanisms referred to in the descriptor file are also provided. At startup time, the management server processes the descriptor file and forms associations between the various aspects of the task type with the mechanisms that are responsible for implementing those various aspects. At runtime, these associations are used to manage the definition and execution of tasks of that task type. By doing so, the management server enables new task types to be added, and enables task instances of the new task types to be defined and executed, without changing the source code of the management server.

Abstract: Techniques are described to allow the deprecation of classes in an object-oriented data model, such as a CDM for a CMDB. When a class is deprecated and replaced by another existing or new class, data associated with instances of the deprecated class may be migrated to the replacement class. A mapping between the deprecated class and its replacement class may be provided to allow existing applications to continue to access data using the deprecated class without change until the deprecated class is finally deleted or the application is updated to use the replacement class. New applications written to use the object-oriented data model after the deprecation may use the replacement class to access data instances created using the original data model.

Abstract: Data is often populated into Configuration Management Databases (CMDBs) from different sources. Because the data can come from a variety of sources, it may have inconsistencies—and may even be incomplete. A Normalization Engine (NE) may be able to automatically clean up the incoming data based on certain rules and knowledge. In one embodiment, the NE takes each Configuration Item (CI) or group of CIs that are to be normalized and applies a rule or a set of rules to see if the data may be cleaned up, and, if so, updates the CI or group of CIs accordingly. In particular, one embodiment may allow for the CI's data to be normalized by doing a look up against a Product Catalog and/or an Alias Catalog. In another embodiment, the NE architecture could be fully extensible, allowing for the creation of custom, rules-based plug-ins by users and/or third parties.

Abstract: This disclosure relates generally to the field of Configuration Management Databases (CMDBs). One embodiment of a user interface embodying the present invention is an extension of the process for creating CMDB classes and is therefore readily available for use by someone with knowledge of CMDB administration. The CMDB administrator is thus relieved from having to understand in detail the technologies and interfaces used by the Management Data Repository (MDR) sources. The result of setting up a relation from a CMDB data structure to an MDR data structure by a CMDB administrator may be represented by one or more new CMDB class(es) for the MDR data. The related MDR may then be accessed by an existing CMDB application using already existing CMDB interfaces. The instances of the new relationships and classes thus appear as if they were native instances stored in the CMDB.

Abstract: A mechanism is disclosed for enabling new task types to be added to a management server without changing the source code of the management server. To add a new task type, a descriptor file is provided. This descriptor file specifies all of the particulars of the new task type. Any mechanisms referred to in the descriptor file are also provided. At startup time, the management server processes the descriptor file and forms associations between the various aspects of the task type with the mechanisms that are responsible for implementing those various aspects. At runtime, these associations are used to manage the definition and execution of tasks of that task type. By doing so, the management server enables new task types to be added, and enables task instances of the new task types to be defined and executed, without changing the source code of the management server.

Abstract: In a computer network system, a mechanism is described for enabling system/network management system to operate in a network address translation environment. Embodiments of the invention allow users to define and specify logical identifiers for agents connecting to the system management network and wishing to communicate with a server layer in the network. The logical identifiers enable the server layer to resolve the addresses of connecting agents to the IP addresses stored and recognized by the server layer.

Abstract: Apparatus, methods and computer program products are disclosed that allow the user of an application to customize the graphical user interface (GUI) for the user's needs. The invention engages in a dialog with the user to determine which GUI features are desired by the user. Depending on the user's answers a model representation of the GUI (a configuration) is saved. The invention provides the user with the ability to select and use the saved configurations.

Abstract: Apparatus, methods and computer program products are disclosed for interfacing a client based, multi-threaded API that is written in a first programming language (such as JAVA) with a server based, single-thread API that is written in a second programming language (such as C or C++). The invention uses an object factory to generate a server object for each client. Programmed-method invocations performed on logical objects at the client are communicated to the server object. The server object processes the programmed-method invocation to perform the requested service such as providing network management information services.

Abstract: Techniques for providing a client program with remote access to information management services of a computer network are disclosed. An application programming interface can be provided to enable the client to remotely access the information management services. By using the application program interface the client can remotely invoke appropriate methods on the server to access the information management services. The application programming interface can be implemented as a thin-class. In addition, the communication between the server and the client programs is transparent to the applications that utilize the application programming interface.

Abstract: A network manager automatically sends certain requests in response to selected events generated by network devices. When a device generates an event, the network manager can send stop requests to the device; send different event requests to the device; or send the same event request to the device, but over a different path. The stop request is sent to cut down on network management traffic. The different event requests can be sent to determine why the event was generated by the device in the first place. The same event request can be sent over a different path to determine whether a problem exists with the device itself or with the path to the device.

Abstract: A network management system allows a user to generate a customized event request for network accessible devices. In response to a user request the system displays a list of devices, and a list of predefined event requests. In response to user customization commands the system generates customized event requests by associating various ones of the displayed devices with various ones of the displayed predefined event requests. At most one predefined event request is associated with each device. The network management system periodically sends the customized event requests to the devices with which they have been associated, and receives response messages from those devices. A display or report may be generated showing the status of the responding devices, using the information in the received response messages. A predefined event request may be formed by storing one or more variable/threshold pairs in a predefined event request data structure.

Abstract: The present invention is a method, apparatus and computer program product for discovering the nodes and other network information of a computer network (the target network) from a device or node that is not part of the target network (the discovery node) when the discovery node uses a different network protocol than the target network. The invention involves identifying one or more additional nodes (the target nodes) that have access to network information about the target network and with whom the discovery node can communicate. This target network information can include the identification of nodes which are present on the network, the topology of the network, the services provided by the network or by the nodes on the network, status information as the network changes such as information about network and node events or alarms, etc.

Abstract: Disclosed herein are methods and apparatus for discovering devices on a network. Active devices can be discovered in ARP tables from routers on the network. Pings can then be sent to the active devices for verification, or pings can be sent to devices at other addresses on the network. Devices can also be discovered by sending a batch of pings to addresses on the network and monitoring responses from those addresses over an interval. After the interval elapses, another batch of pings can be sent. The devices can be discovered by a host on the network or by a network manager. The network manager can add the discovered devices to a network topology database.

Abstract: Elements on a network are selectively unmanaged by a network manager. An element is not managed when it is placed in a Pending state. Traps and events from the element in the Pending state are handled but not processed. The element can be automatically placed in the Pending state if a trap or event has been generated by that element.

Abstract: Elements on a network are selectively unmanaged by a network manager. An element is not managed when it is placed in a Pending state. Traps and events from the element in the Pending state are handled but not processed. The element can be automatically placed in the Pending state if a trap or event has been generated by that element.