Abstract: A technique is disclosed for detecting a memory leak. A request to allocate or deallocate memory is received. In a memory leak table that is bounded in size, an entry that is associated with a call path with which the request is associated is accessed. The entry in the memory leak table is updated.

Abstract: A technique is disclosed for detecting a memory leak. A request to allocate or deallocate memory is received. In a memory leak table that is bounded in size, an entry that is associated with a call path with which the request is associated is accessed. The entry in the memory leak table is updated. The entry is accessed and a value stored in the entry is read. It is determined if the value indicates that a memory leak may be present.

Abstract: A fabric access device having a plurality of system interfaces is disclosed. The fabric access device includes a fabric interface configured to receive data from and send data to a switch fabric and a routing circuit configured to send data received on one of the plurality of system interfaces to the switch fabric via the fabric interface. The inclusion of a plurality of system interfaces facilitates a variety of configurations that provide different levels of redundancy and optimize use of available bandwidth.

Abstract: An actuating device for a connector having a push-pull coupling mechanism is disclosed. The device comprises a release pull integrated with the connector and configured to operably engage the push-pull coupling mechanism, and a grip, coupled to the release pull and remote from the push-pull coupling mechanism, to which force may be applied to cause the release pull to actuate the push-pull coupling mechanism.

Abstract: A header value or label, referred to herein as a source station identification (SSID), is added to an encapsulated packet header, such as by adding the SSID as a label to the bottom of a stack of MPLS labels. The SSID comprises a unique identifier that identifies the PE that originated the packet. In some embodiments, the IP address of the originating PE may be used as the SSID for that PE. The PE receiving this packet can then associate the source Ethernet MAC address of received TLS traffic, e.g., with the originating PE. Given the SSID of the originating PE, the receiving PE is able to determine which LSP to use to send Ethernet traffic to the station with the learned MAC address.

Abstract: Optimizing statistics counter use is disclosed. A total number of counter bits to be used to track two or more statistics is determined. The total number of counter bits is allocated among the two or more statistics to provide for each statistic a counter comprising the number of bits allocated for that statistic, the allocation being such that each counter overflows at a rate desired for that counter. The overflow rates may be balanced, such that each counter overflows at approximately the same rate.

Abstract: Extending the Spanning Tree Protocol (STP) to provide redundancy and avoid loops in the topology of a service domain associated with a data transport service provided using a provider core network is disclosed. At each of a plurality of provider edge (PE) devices participating in the service, each participating PE being connected to the other participating PE's by a full mesh of mesh connections through the provider core network, a virtual core port is defined that encompasses all of the ports on the PE that are associated with a mesh connection from the PE to one or more other PE's participating in the service. Each participating PE is configured to associate the virtual core port with a virtual core bridge representing the core of the service domain. It is ensured that the virtual core bridge becomes and remains the root bridge.

Abstract: An ejector assembly for securing a modular assembly in a chassis is disclosed. The ejector assembly comprises a cam rotatably mounted to the modular assembly such that the cam may be rotated about a first axis, the cam having a control member configured to extend into a recess in the chassis when the cam is rotated into a locked position and be retracted from the recess when the cam is rotated into a retracted position; and a lever rotatably mounted to the modular assembly such that the lever may be rotated about a second axis substantially parallel to the first axis, the lever being configured to cause the cam to be rotated about the first axis when a force is applied to the lever to rotate the lever about the second axis, whereby the cam may be rotated into either the locked position or the retracted position by operating the lever.

Abstract: An actuating device for a connector having a push-pull coupling mechanism is disclosed. The device comprises a release pull integrated with the connector and configured to operably engage the push-pull coupling mechanism, and a grip, coupled to the release pull and remote from the push-pull coupling mechanism, to which force may be applied to cause the release pull to actuate the push-pull coupling mechanism.

Abstract: Loss of connectivity on a mesh or virtual service (e.g., VPLS) may be restored by “stitching” a healing path between an affected node and another node on the mesh. Connectivity may be restored by “healing” or constructing an alternate path or connection between the affected node and the mesh. Unidirectional and bidirectional healing may be used to establish an alternate path between the affected node and the mesh. Additionally, removing pseudowire connections between the affected node and the mesh may be performed, while establishing a spoke connection to an advertised, available router to create a new connection or “stitch” together existing connections.

Abstract: A fabric access device comprising a plurality of system interfaces is disclosed. The fabric access device includes a fabric interface configured to receive data from and send data to a switch fabric and a routing circuit configured to send data received on one of the plurality of system interfaces to the switch fabric via the fabric interface. The inclusion of a plurality of system interfaces facilitates a variety of configurations that provide different levels of redundancy and optimize use of available bandwidth.

Abstract: Echo messaging for operation, administration, and management of a service-based distribution path and associated services are disclosed. Service-based distribution paths or transport tunnels include services mapped or bound to a path associated with the transport tunnel. Echo messaging provides OAM capabilities to monitor the operational state of a service-based distribution path, including determining configuration, connectivity, and other characteristics of the path and associated services that transport data. OAM functions provided by echo messaging enable OAM functions despite service volume along a core network, path or set of paths.

Abstract: Optimizing statistics counter use is disclosed. A total number of counter bits to be used to track two or more statistics is determined. The total number of counter bits is allocated among the two or more statistics to provide for each statistic a counter comprising the number of bits allocated for that statistic, the allocation being such that each counter overflows at a rate desired for that counter. The overflow rates may be balanced, such that each counter overflows at approximately the same rate.

Abstract: Inserting an address used for performing such OAM functions in an efficient way that is transparent to a customer or service using the network path is disclosed. An indication is received that a source address reserved for OAM use should be inserted into a forwarding database associated with a provider network node, such as a provider edge device, that receives the indication. The address is inserted into a forwarding database as an address that is owned by the provider network node and reserved for OAM use. One or more other nodes associated with a service with which the OAM address is associated may be notified that the OAM address is associated with the provider network node that owns the OAM address.

Abstract: Network data switching comprises receiving a cell, associating the cell with a destination port, selecting an egress link that has been soft configured to be associated with the destination port, and switching the cell to the selected egress link. A network switch comprises an ingress link configured to receive a cell and a switch circuit coupled to the ingress link, configured to associate the cell with a destination port, to select an egress link that has been soft configured to be associated with the destination port, and to switch the cell to the egress link.

Abstract: Setting packet queue priority is disclosed. A rate associated with a packet queue is measured. The rate is compared to a quality of service metric. A priority is assigned to the packet queue based at least in part on the comparison between the rate and the quality of service metric. The quality of service metric may be a committed information rate and/or a peak information rate.

Abstract: Network traffic switching comprises: for each of a plurality of classes, specifying a next queue to be serviced within that class, selecting from among the classes a next class to be serviced, and sending data from the next queue of the next class to be serviced via an egress link.

Abstract: An enhanced virtual router redundancy protocol for manipulating master router selection within a network is disclosed. The use of a priority control policy and priority values take into account conditions and events that may distinguish the desirability of one router over another. Explicit priority values may be specified to indicate the explicit priority of a particular router instance. Delta priority values may also be used to decrement or increment a base priority for a particular router instance. By using priority control policies and priority values to select a master, improved control, management, and manipulation of virtual router instances over a particular IP core network interface are enabled.

Abstract: A technique for providing access to a switch fabric is disclosed. In some embodiments, the method comprises determining scheduling information specific to a link to the switch fabric and using the scheduling information to determine a queue to be serviced by the link. In some embodiments, the system comprises a buffer configured to store a plurality of queues, and a queue scheduler associated with the buffer and configured to determine scheduling information specific to a link to the switch fabric and use the scheduling information to select one of the plurality of queues as the next queue to be serviced by the link.

Abstract: Multicasting network packets is disclosed. A total number of copies of a frame, t, to be sent is determined. A number of copies of the frame, m, which is less than a total number of copies of the frame, t, to be made during a current iteration is determined. M copies of the frame are made. The m copies of the frame are then sent to their destinations. The original input frame is provided as output with an indication that the frame should be returned for further processing. Processing of the frame is discontinued during an interval in which other frames are processed. The process is repeated until t copies have been sent.