Abstract: A method is disclosed for transmitting system management requests to computer systems along a network path using a network control protocol, such as RSVP. For example, an originating node may send a single system management request along a path to a destination node using a network control protocol. Each computer system along the network path may analyze the network control protocol message to determine whether the message contains a system management request. If a system management request is found in the message, the computer system may perform the system management function identified in the request, and respond to it.

Abstract: A system and method provides a managed resource instrumentation application on a network device to instrument a managed network device component. The managed resource instrumentation application includes an application programming interface (API) that is common to two or more management agents that are implemented on the device, the management agents being configured to process network management instructions that are in different respective protocols. The API is common to the management agents, to provide access by the management agents to the managed network device component via the common API.

Abstract: The network device is described that comprises an allocator to adaptively allocate respective event message rate limits to client network devices that is in communication with an event-based system logging server to send event messages to the logging server for processing. The adaptively allocated event message rate limits are communicated to the client network devices so that limiting of a global rate of event messages received by the logging server comprises limiting the respective rates at which the client network devices can transmit event messages to the logging server. Measurement of respective event message rates comprises a count of event messages actually received by the logging server from the corresponding client device within a defined time window.

Abstract: In an embodiment, a method comprises initiating a monitoring session for a communication path including creating and storing monitoring session state data; sending, to a first responder computer of the communication path, a first request to initiate a first state servlet that is configured to monitor continuously during the monitoring session one or more characteristics of one or more processes that the first responder computer may perform; sending, to the first responder computer, monitoring instructions to monitor the one or more characteristics of the one or more processes; while the monitoring session is active and the first responder computer is in the communication path, receiving and collecting monitored information from the first responder computer; in response to determining that the first responder computer is not in the communication path or that the monitoring session has become inactive, automatically and autonomously ending the monitoring session.

Abstract: Methods and apparatus for efficient decentralized information dissemination in a network are discussed herein. The decentralized information dissemination techniques are based on non-deterministic protocols such as gossip-based protocols. An example method for dissemination of information in a network can include: selecting a destination node from a set of neighboring nodes using a gossip-based protocol; transmitting a message to the destination node; and receiving a feedback response from the destination node. The feedback response can include a level of usefulness of the message to the destination node. In addition, each neighboring node in the set can be associated with a selection weight, and the selection weights can be related to a probability of selecting each neighboring node in the set as the destination node using the gossip-based protocol. Upon receiving the feedback response, the method can include adjusting the selection weight of the destination node based on the feedback response.

Abstract: A method and apparatus for self-monitoring to identify an occurrence of a threshold and rebooting in response to the occurrence of the threshold is provided. In an embodiment, a data processing apparatus comprises one or more processors; logic coupled to the one or more processors and comprising one or more stored sequences of instructions which, when executed by one or more processors, cause the one or more processors to obtain a threshold associated with the apparatus; self-monitor the apparatus to identify an occurrence of the threshold; and self-reboot the apparatus responsive to the occurrence of the threshold.

Abstract: A system and technique for visualizing changes and trends in performance data over a network path are disclosed. More specifically, a network management system generates and displays an object that represents data for one or more performance characteristics related to a plurality of network nodes in the network path. The object is dynamically updated to include updated data related to the one or more performance characteristics. In one embodiment, a representation of the updated data is superimposed over representations of previously collected data relating to the performance characteristics. The transparency of the representations associated with the previously collected data may reflect a difference in time between the current time and the time at which the data was collected. In another embodiment, a timeline is included in the object that tracks the state of the object at one or more previous points in time.

Abstract: A method, device, and computer-readable medium are disclosed for automatically activating a probe configured to generate test network traffic in response to evaluating a policy that accounts for aggregated information that describes traffic that was processed by a network device. The method includes storing a policy, and evaluating the policy based at least in part on an item of received aggregated information that describes a set of packets that were processed by the network device and sent or received on a network. The probe is activated at least partially in response to evaluating the policy. The probe generates test packets to emulate one or more applications, services, or devices communicating on the network.

Abstract: Techniques in a data processor for translating an object-oriented data model to a YANG data model are described. In one embodiment, for example, a translator system is described for translating data representing an object-oriented data model construct to a data representing one or more YANG data model statements, the translator system comprises: an object-oriented data model construct identifier for identifying data representing an object-oriented data model construct; an object-oriented data model construct-to-YANG data model statement translator for translating the identified data to translated data representing one or more YANG data model statements; and a storing module for storing the translated data in one or more non-transitory computer-readable media.

Abstract: In an embodiment, a method comprises initiating a monitoring session for a communication path including creating and storing monitoring session state data; sending, to a first responder computer of the communication path, a first request to initiate a first state servlet that is configured to monitor continuously during the monitoring session one or more characteristics of one or more processes that the first responder computer may perform; sending, to the first responder computer, monitoring instructions to monitor the one or more characteristics of the one or more processes; while the monitoring session is active and the first responder computer is in the communication path, receiving and collecting monitored information from the first responder computer; in response to determining that the first responder computer is not in the communication path or that the monitoring session has become inactive, automatically and autonomously ending the monitoring session.

Abstract: In one implementation, a protocol for dissemination of information, such as configuration information, in a computer network is provided. The dissemination of the payload or content is separated from the dissemination of the fact that a new revision of that payload or content exists. The availability of a new revision of the payload or content is signaled prior to the transfer. Only nodes receiving the notice of existence of the new revision and requesting the new revision receive the payload or content. While the payload or content revision notifications potentially arrive at a receiving node from many different neighbors, the payload or content is only requested once by the receiving node and transmitted once over links between neighbors. This scheme may result in low end-to-end delays and in resiliency.

Abstract: A method, device, and computer-readable medium are disclosed for automatically activating a probe configured to generate test network traffic in response to evaluating a policy that accounts for aggregated information that describes traffic that was processed by a network device. The method includes storing a policy, and evaluating the policy based at least in part on an item of received aggregated information that describes a set of packets that were processed by the network device and sent or received on a network. The probe is activated at least partially in response to evaluating the policy. The probe generates test packets to emulate one or more applications, services, or devices communicating on the network.

Abstract: Methods and apparatus for efficient decentralized information dissemination in a network are discussed herein. The decentralized information dissemination techniques are based on non-deterministic protocols such as gossip-based protocols. An example method for dissemination of information in a network can include: selecting a destination node from a set of neighboring nodes using a gossip-based protocol; transmitting a message to the destination node; and receiving a feedback response from the destination node. The feedback response can include a level of usefulness of the message to the destination node. In addition, each neighboring node in the set can be associated with a selection weight, and the selection weights can be related to a probability of selecting each neighboring node in the set as the destination node using the gossip-based protocol. Upon receiving the feedback response, the method can include adjusting the selection weight of the destination node based on the feedback response.

Abstract: A method, a system, and an apparatus are provided for organizing management information bases (MIB) in a network. A table, associated with an overlay MIB structure, is populated with entries. Each entry defines an object identifier (OID) of a mount point and an OID of a mount target. A selection is made between the existing OIDs and OIDs corresponding to a MIB overlay. When management communication refers to managed objects, they can be referred to in terms of OIDs defined by the MIB overlay, as an alternative to their existing OIDs. An agent infrastructure is defined to support the MIB overlay structure.

Abstract: The network device is described that comprises an allocator to adaptively allocate respective event message rate limits to client network devices that is in communication with an event-based system logging server to send event messages to the logging server for processing. The adaptively allocated event message rate limits are communicated to the client network devices so that limiting of a global rate of event messages received by the logging server comprises limiting the respective rates at which the client network devices can transmit event messages to the logging server. Measurement of respective event message rates comprises a count of event messages actually received by the logging server from the corresponding client device within a defined time window.

Abstract: A method, device, and computer-readable medium are disclosed for automatically activating a probe configured to generate test network traffic in response to evaluating a policy that accounts for aggregated information that describes traffic that was processed by a network device. The method includes storing a policy, and evaluating the policy based at least in part on an item of received aggregated information that describes a set of packets that were processed by the network device and sent or received on a network. The probe is activated at least partially in response to evaluating the policy. The probe generates test packets to emulate one or more applications, services, or devices communicating on the network.

Abstract: In one embodiment, a network device is described as including a rate monitor to monitor an actual individual message rate of event messages sent from each one of a plurality of sending devices operatively in communication with the network device, an allocator to allocate an individual message rate limit to each of the plurality of sending devices, and a communication module to communicate a rate limit instruction to at least one of the sending devices, the rate limit instruction to limit the transmission rate of event messages.

Abstract: A method is disclosed for managing a network entity that is initiated by the network entity, the method comprising the computer-implemented steps performed at the network entity of monitoring the network entity; periodically evaluating one or more specified conditions at the managed network entity; when one or more of the specified conditions are satisfied, then gathering specified information from the managed network entity, preparing a message that includes the specified information and the specified conditions that were satisfied, and sending the message to a management point.

Abstract: An apparatus is disclosed for creating and storing policy data records comprising data identifying network policy decisions. After a data packet is received, a network policy decision is made based on information in the packet and one or more network policies. A policy data record identifying the network policy decision is created, and the policy data record is stored.

Abstract: Techniques for system (syslog) messages are provided. As syslog messages are generated a dual sequence number is maintained for each syslog message. The first sequence number is consistent for each syslog message and represents a total number of syslog messages received. The second sequence number is maintained as a total message count for a specific session for which a syslog message is assigned.