Juniper Networks, Inc.

Manoj Sharma

Nitin Singh

Shrinivasa Kini

Sunesh Rustagi

Tapraj Singh

Poorna Pushkala Balasubramanian

John E Drake

Kaushik Ghosh

Anil Lohiya

Ramasamy Ramanathan

Hassan C Hosseini

Xiaomin Wu

Cunzhi Lu

Siddharth A Tuli

Wen Lin

Pankaj Shukla

Arijit Paul

Ryan Bickhart

Vikas Vidyalal Patel

Deepak Ojha

Anshu Verma

Prasantha Kumar Gudipati

Sudha Madhavi Yeevani

A provider edge device, capable of accessing a first type of memory and a second type of memory, may determine a network address associated with a customer edge device. The provider edge device may determine whether the customer edge device is categorized as a leaf device in an Ethernet Tree service provided by the provider edge device. The provider edge device may selectively store the network address in the first type of memory or the second type of memory based on determining whether the customer edge device is categorized as a leaf device in the Ethernet Tree service.

Conserving memory resources of a provider edge device in an ethernet tree

A device may receive, from a first device associated with a first LAN, network traffic destined for a second LAN. The device may provide the first LAN with access to a core network. The device may not provide the second LAN with access to the core network. The device may identify, based on the network traffic, a Layer 3 address associated with a second device. The second device may be associated with the second LAN. The device may determine that the first device is categorized as a leaf device within an Ethernet Tree provided by the device. The device may determine, based on the Layer 3 address, that the second device is categorized as a leaf device within the Ethernet Tree. The device may drop the network traffic based on determining that the first device and the second device are categorized as leaf devices within the Ethernet Tree.

Processing inter-VLAN traffic in an ethernet tree

A provider edge (PE) device may receive an indication to perform a designated forwarder (DF) election associated with a network segment that includes the PE device, one or more other PE devices, and a client edge (CE) device. The PE device, the one or more other PE devices, and the CE device may be associated with an Ethernet virtual private network (EVPN) that includes a group of EVPN instances (EVIs). The PE device may perform the DF election in order to determine election information associated with the PE device. The election information may include information associated with a particular EVI, of the group of EVIs, for which the PE device is to act as a DF. The PE device may provide the election information to the CE device to cause the CE device to provide traffic, associated with a particular VLAN included in the particular EVI, to the PE device.

Adaptive load balancing for single active redundancy using EVPN designated forwarder election

The disclosed apparatus may include (1) a database that maintains a pruned multicast distribution tree and (2) a processing unit having access to the database, wherein the processing unit (A) receives at least one type 3 Ethernet Segment Identifier (ESI) message from at least one other bridge domain on at least one other computing device, (B) determines, based at least in part on the type 3 ESI message, whether the other bridge domain on the other computing device has a root access concentrator, and then (C) populates, in response to determining whether the other bridge domain has a root access concentrator, the pruned multicast distribution tree to account for the other bridge domain such that the bridge domain on the computing device is able to determine whether to send multicast traffic to the other bridge domain. Various other apparatuses, systems, and methods are also disclosed.

Apparatus, system, and method for preventing multicast traffic drops at egress routers in E-Tree topologies

In one example, a method includes configuring a first provider edge (PE) router of a Provider Backbone Bridging (PBB) Ethernet Virtual Private Network (EVPN) to join an Ethernet Segment in active-active mode with at least a second PE router that is operating as a designated forwarder for the Ethernet Segment; receiving, by the first PE router from a remote PE router and prior to the first PE router performing Media Access Control (MAC) learning of a customer-MAC (C-MAC) address that is reachable via a backbone-MAC (B-MAC) address associated with the Ethernet Segment, a network packet that includes the C-MAC address; and in response to determining that the C-MAC address has not been learned by the first PE router and the B-MAC address included in the network packet is associated with the Ethernet Segment, forwarding, by the first PE router, the network packet to a destination identified by the C-MAC address.

Traffic black holing avoidance and fast convergence for active-active PBB-EVPN redundancy

In general, techniques are provided for described herein that extend existing Ethernet Virtual Private Network (EVPN) protocol signaling mechanisms so that local, multi-homing PEs couple to an Ethernet segment can definitively convey their primary/backup designated forwarder (DF) status to any remote PE of the EVPN. In one example, this is accomplished by utilizing a new extended community attribute to each Ethernet A-D per EVI route advertised by each of the multi-homing PEs to specifically carry the advertising PE's primary or backup status. As such, any receiving remote PE need not rely on the arrival of individual MAC routes from a new primary PE and withdrawal of MAC routes from a former primary PE to update its forwarding information.

Update of MAC routes in EVPN single-active topology

In general, techniques for facilitating a distributed network (L3) subnet by which multiple independent control planes of network devices connected to physically separate L2 networks provide L2 reachability to/from a single L3 subnet. In some examples, a shared L2 network physically situated to connect a plurality of physically separate L2 networks 'stitches' the L2 networks together within the respective, independent control planes of switches such that the control planes bridge L2 traffic for a single bridge domain for the separate L2 networks to the shared L2 network and visa-versa. Each of the independent control planes may be configured with a virtual IRB instance associated with the bridge domain and with a common network subnet. Each of the virtual IRBs provides a functionally similar routing interface for the single bridge domain for the separate L2 networks and allows the shared network subnet to be distributed among the independent control planes.

Distributed network subnet

Techniques are described for providing fast convergence in the event of a link failure in an all-active multi-homed Ethernet virtual private network. A provide edge (PE) network device may pre-configure an interface next hop and secondary next hops. The secondary next hops may be logical links to other PE network devices in the same Ethernet segment. In the event of a link failure in the interface next hop between the PE network device and a customer edge (CE) network device, the PE network device may be configured to forward data traffic to the CE network device using the secondary next hops. In the event of a link failure between the PE network device and a core network, the PE network device may be configured to send an out-of-service message to the CE network device that instructs the CE network device to stop sending traffic to the PE network device.

Fast convergence on link failure in multi-homed Ethernet virtual private networks

In general, techniques are described for providing control plane messaging in an active-active (or all-active) configuration of a multi-homed EVPN environment. In some examples, the techniques include receiving a control plane message comprising at least one address that identifies that second PE network device. The techniques may include configuring, based at least in part on the control plane message, a forwarding plane of a first PE network device to identify network packets having respective destination addresses that match the at least one address. The techniques may include determining that at least one address of the network packet matches the at least one address that identifies the second PE network device. The techniques may include, responsive to the determination, skipping a decrement of the Time-To-Live (TTL) value of the network packet, and forwarding the network packet to the second PE network device.

Growing community of inventors

Cupertino, CA, United States of America

Manoj Sharma