Abstract: Methods for directly mapping attributes from controllers connected to a switch are provided. Multicast protocols are combined with policy management and applied to systems with multiple host and source devices. Unnecessary data packets are dropped, while only necessary data packets are forwarded. These methods decrease the amount of resources needed to process requests received during multicasting, provide enhanced security options, and further integrate networking solutions.

Abstract: Examples disclosed herein relate to establishing a layer 3 (L3) Multi-Chassis Link Aggregation Group (MC-LAG). In an example, a common IP address and a common MAC address may be associated with a primary network device and a secondary network device. A layer 3 MC-LAG may be established in a multi-homing configuration between the primary network device and the secondary network device to provide a redundant L3 connectivity to a core network device in a network. A dedicated communication link may be established between the primary network device and the secondary network device, for the primary network device and the secondary network device to share network packets.

Abstract: A system that can facilitate multicast forwarding via overlay multicast tunnels in a network is provided. The system can operate as a tunnel endpoint and form a first overlay tunnel with a peer tunnel endpoint. During operation, the system can receive a join request for a multicast group via the first overlay tunnel. The system can identify a local port of the system associated with the multicast group indicated by the join request. The system can then include the first overlay tunnel in a set of forwarding tunnels for multicast traffic of the multicast group based on the join request. Subsequently, the system can forward the multicast traffic of the multicast group received from a local port via the set of forwarding tunnels, thereby limiting forwarding of the multicast traffic to tunnel endpoints associated with the multicast group.

Abstract: An example system includes a network having interconnected routers, a multicast source to transmit data through the network for receipt by receiving host devices, and a query agent communicatively coupled to each router of the interconnected routers. The query agent is to generate a flow tree for multicasting of the data from the multicast source to each receiving host device, the flow tree including flow paths from the multicast source to each receiving host device, query each router to determine a status of at least one flow path of the flow tree, and identify an anomaly associated with at least one router in the at least one flow path.

Abstract: Examples disclosed herein relate to a method comprising determining, at a first node, that a first Ethernet Ring Protection Switching (ERPS) port of the first node is down, wherein the first node belongs to a plurality of network nodes connected in a ring topology to form an access network. The method includes determining, at the first node, that the first ERPS port is a querier port, wherein the first ERPS port is paired with a second ERPS port of the first node and marking the second ERPS port as the querier port of the first node. The method includes transmitting, by the first node, an IGMP join message to a device part of a multicast implementation, wherein the multicast implementation covers at least the ring topology.

Abstract: Some examples relate to managing multicast scaling. In an example, a network device may receive a message to join an IP multicast group. The network device may determine whether a hardware filter is available on the network device to process the received message. In response to the determination that the hardware filter is not available on the network device to process the received message, a current state of an IP multicast protocol associated with the message may be changed to an error state on the network device, wherein in the error state, further messages to join the IP multicast group that are received by the network device are discarded by the network device.

Abstract: Examples include receiving a first multicast packet, determining a first flow for the first multicast packet based on a multicast protocol, and storing the first flow in a routing table. The first flow comprises a first source and a first group and the routing table comprises a second flow with a second source and the first group. Additionally, examples include programming the first flow into a hardware memory resource and programming a summary flow into the hardware memory resource. The hardware memory resource comprises the second flow and the summary flow comprises a wild character that matches the first source and the second source.

Abstract: Examples include generating a Precision Time Protocol (PTP) packet for a first nexthop in an Equal Cost Multi-Path set and sending the PTP packet to the first nexthop. Examples also include receiving a response from the first nexthop that identifies a time delay associated with a route to the first nexthop and updating the ECMP based on the time delay.

Abstract: The present disclosure relates a system and a method of identifying network faults residing in an underlay network. A network device initiating fault analysis sends a query message through a tunnel interface present in an overlay network configured over an underlay network. While an upstream interface is a network tunnel having a first end and a second end, a unicast path from the first end to the second end of the network tunnel is traced using a network route analysis function. Reachability of a destination device present on the second end of the network tunnel is identified. An output of the network route analysis function is appended with an output of a multicast network connectivity determining function to produce an appended result that is shared with the network device initiating fault analysis.

Abstract: Systems and methods are provided for performing a node-level redundant failover-type process with respect to the protocol-independent multicast (PIM) functionality in a multi-chassis environment. When a PIM-related failure occurs on a first network device, but otherwise it remains operational, a second network device is configured to assume responsibility for performing PIM data traffic forwarding. Upon detecting the PIM-related failure of the first network device, the second network device sends a PIM-DR failover event signal to the second network device's PIM module by loading multicast route states used by the first network device into the PIM data traffic forwarding hardware of the second network device. Upon the second network device assuming responsibility, the first network device disables its PIM data traffic forwarding functionality.

Abstract: Examples include generating a Precision Time Protocol (PTP) packet for a first nexthop in an Equal Cost Multi-Path set and sending the PTP packet to the first nexthop. Examples also include receiving a response from the first nexthop that identifies a time delay associated with a route to the first nexthop and updating the ECMP based on the time delay.

Abstract: Examples disclosed herein relate to monitoring flow activity on a network device. In an example, a neighbor table is maintained on a network device. The neighbor table may include a record of a neighbor network device and a hit bit corresponding to the neighbor network device. The hit bit may be used to represent a flow activity of the neighbor network device. A determination may be made whether a status of the hit bit corresponding to the neighbor network device is inactive. If the status of the hit bit is inactive, a flow entry corresponding to the neighbor network device may be deleted from an ASIC table on the network device.

Abstract: In some examples, a network device includes an interface, and a processor to apply a restriction on multicast communication associated with an entity on the interface. The restriction on multicast communication includes detecting, on the interface, a multicast communication pattern associated with the entity, indicating, based on the multicast communication pattern on the interface violating a threshold, that the entity is malicious, and blocking processing of the multicast communication associated with the entity in response to indicating that the entity is malicious.

Abstract: Examples disclosed herein relate to a method comprising determining, at a first node, that a first Ethernet Ring Protection Switching (ERPS) port of the first node is down, wherein the first node belongs to a plurality of network nodes connected in a ring topology to form an access network. The method includes determining, at the first node, that the first ERPS port is a querier port, wherein the first ERPS port is paired with a second ERPS port of the first node and marking the second ERPS port as the querier port of the first node. The method includes transmitting, by the first node, an IGMP join message to a device part of a multicast implementation, wherein the multicast implementation covers at least the ring topology.

Abstract: Examples disclosed herein relate to establishing a layer 3 (L3) Multi-Chassis Link Aggregation Group (MC-LAG). In an example, a common IP address and a common MAC address may be associated with a primary network device and a secondary network device. A layer 3 MC-LAG may be established in a multi-homing configuration between the primary network device and the secondary network device to provide a redundant L3 connectivity to a core network device in a network. A dedicated communication link may be established between the primary network device and the secondary network device, for the primary network device and the secondary network device to share network packets.

Abstract: An example system includes a network having interconnected routers, a multicast source to transmit data through the network for receipt by receiving host devices, and a query agent communicatively coupled to each router of the interconnected routers. The query agent is to generate a flow tree for multicasting of the data from the multicast source to each receiving host device, the flow tree including flow paths from the multicast source to each receiving host device, query each router to determine a status of at least one flow path of the flow tree, and identify an anomaly associated with at least one router in the at least one flow path.

Abstract: Examples herein are directed to centralized database based multicast converging. For instance, in various examples centralized database based multicast converging can include starting a restart timer having a value greater than a time to validate stored entries in a centralized database, sending data packets at least to hosts on the network corresponding to the stored entries in the centralized database to maintain service to the hosts while the restart timer is running, sending query packets to validate a host corresponding to an entry of the stored entries in the centralized database, and responsive to the restart timer expiring, sending data packets to a converged group of hosts including at least the validated host.

Abstract: A network device may send, to a Dynamic Host Configuration Protocol (DHCP) server, a request for an Internet Protocol version 6 (IPv6) address to be assigned to a management port of the network device, wherein IPv6 is disabled at the network device, and may receive a message that includes information associated with a network management system (NMS) and IPv6 configuration information for enabling IPv6 processing on the management port. In response to receiving the IPv6 configuration information, the network device may enable IPv6 processing on the management port of the network device and may register with the NMS based at least in part on the information associated with the NMS. The network device may, in response to receiving one or more configuration commands sent from the NMS to the management port of the network device, configure the network device according to the one or more configuration commands.

Abstract: Examples include receiving a recovery signal and access data stored in a centralized database of a networking device in response to the recovery signal. The data may include information associated to past multicast flow events. Additionally, examples include determining a rebuilt multicast flow event based on the data, generating a multicast protocol packet based on the multicast flow event, and sending the multicast protocol packet to a multicast protocol in the networking device.

Abstract: Systems and methods are provided for performing a node-level redundant failover-type process with respect to the protocol-independent multicast (PIM) functionality in a multi-chassis environment. When a PIM-related failure occurs on a first network device, but otherwise it remains operational, a second network device is configured to assume responsibility for performing PIM data traffic forwarding. Upon detecting the PIM-related failure of the first network device, the second network device sends a PIM-DR failover event signal to the second network device's PIM module by loading multicast route states used by the first network device into the PIM data traffic forwarding hardware of the second network device. Upon the second network device assuming responsibility, the first network device disables its PIM data traffic forwarding functionality.