Nicira, Inc.

Cisco Technology, Inc.

Pradeep Singh

Hua Wang

Anupam Chanda

Ziyou Wang

Jin Liu

Bolt Zhang

Jianjun Shen

Ariel Tubaltsev

HsinYi Shen

David Delano Ward

Suresh Katukam

Dongling Duan

David A Friedman

Joseph Marchionni

Chakravarthi S Chigurupati

Tejas H Rajkotia

Bharat M Bhojwani

For a network with host machines that are hosting virtual machines, a method for facilitating BUM (broadcast, unknown unicast, and multicast) traffic between a hardware switch (e.g., ToR switch) and the host machines is provided. The network has a set of host machines configured as a cluster of replicators for replicating BUM traffic from the hardware switch to the host machines. A set of network controllers establishes failure-detection tunnels for links between the hardware switch and the replicator cluster. The replicator cluster informs the set of controllers of a change in the membership of the replicator cluster to initiate an update to the active failure-detection sessions. The set of network controllers communicates with the replicator cluster and a ToR switch to establish bidirectional forwarding detection (BFD) sessions between one or more replicator nodes in the replicator cluster and the ToR switch.

Automatic setup of failure detection sessions

Some embodiments provide, for a first controller application, a method for configuring a managed hardware forwarding element (MHFE) to implement one or more logical networks. The method of some embodiments receives logical network data that defines at least one logical forwarding element of a logical network to be implemented by the MHFE. The method then identifies a set of tables of a database instance that is instantiated on the MHFE in order to distribute the logical network data to the MHFE. In some embodiments, the method monitors the identified set of tables in order to determine whether a second controller application updates any one of the set of tables. The method distributes the logical network data to the MHFE so long as none of the tables in the set of tables is updated by the second controller application.

Preventing concurrent distribution of network data to a hardware switch by multiple controllers

Disclosed are, inter alia, methods, apparatus, and means for stateful switching between reliable transport modules for communicating with an external peer without losing the transport layer connection. Primary and standby reliable transport protocol modules each maintain state concerning the reliable transport connection (e.g., data, segmentation, acknowledgements) such that if the primary or standby reliable transport protocol module fails, the other can resume by itself such that the communication with the peer transport application does not need to be restarted. Also, by the communications subsystem of a device providing copies of received reliable transport protocol messages directly to both the primary and standby reliable transport protocol modules, upon failover, the communications subsystem does not need to be reconfigured for resuming operations as, for example, the standby reliable transport protocol module will already be receiving these packets.

Stateful switching between reliable transport modules for communicating with an external peer without losing the transport layer connection

A system for resolving address conflicts in a network. In an illustrative embodiment, the system includes an address-configuration module that is adapted to assign addresses to one or more devices. The one or more devices are connected to the network via device interfaces. An interface-monitoring module communicates with the address-configuration module. The interface-monitoring module is adapted to determine when an address conflict involving plural addresses occurs and to provide a signal in response thereto. A conflict-resolution module is adapted to selectively terminate one or more device interfaces associated with the plural addresses in response to the signal and based on the addresses and one or more predetermined precedence rules. In a more specific embodiment, the system further employs a user interface for facilitating selectively adjusting the precedence rules. The conflict-resolution module runs one or more routines adapted to implement a static conflict-resolution policy, a longest-prefix policy, a highest-address policy, and/or an interface-identifier policy for Internet Protocol (IP) addresses.

System and method for resolving address conflicts in a network

Methods and apparatus for substantially minimizing single points of failure for circuit paths in networks with mixed protection schemes are disclosed. According to one aspect of the present invention, a method for routing circuit paths between a source and a destination of a network includes identifying a first available circuit path between the source and the destination. The first available circuit path includes a first plurality of links which each have an associated protection type. The method also includes determining a number protection changes associated with the first plurality of links, and assigning a first metric to the first available path that is based on the number of protection changes. Finally, the method includes identifying a selected available path to be used to pass information between the source and the destination based at least in part on the first metric.

Minimizing single points of failure in paths with mixed protection schemes

A method and system for representing a low-order connection and a high-order connection as a single circuit are disclosed. The method includes creating a low-order listener object and a low/high listener object at each end point of the low-order connection and creating a high-order listener object at each end point of the high-order connection. The low/high listener objects are matched with the high-order listener objects to create a link and represent the connections as a single circuit.

Method and system for discovering circuits in a network

Growing community of inventors

San Jose, CA, United States of America

Pradeep Singh