Tag :: TOI

EVPN MPLS VPWS (RFC 8214) provides the ability to forward customer traffic to / from a given attachment circuit (AC) without any MAC lookup / learning. The basic advantage of VPWS over an L2 EVPN is the reduced control plane signalling due to not exchanging MAC address information. In contrast to LDP pseudowires, EVPN MPLS VPWS uses BGP for signalling. Port based and VLAN based services are supported.

In network deployments, where the border leaf or Superspine act as PEG and it is in the transit path to other multicast VTEPs, the multicast stream will not pass since the border leaf will decapsulate the packet even if it doesn't have a receiver. This transit node is called the Bud Node. The device should be able to send decapsulated packets to any local receivers as well as send the encapsulated packets to other VTEPs

Multihoming in EVPN allows a single customer edge (CE) to connect to multiple provider edges (PE or tunnel endpoint). These PE devices are all connected to the same Ethernet-Segment (ES). Multihoming is activated by assigning a unique Ethernet Segment Identifier (ESI) and ES-Import Route Target (RT) which enables all the PEs connected to the same multihomed site to import the Type 4 ES routes

In EVPN, an overlay index is a field in type-5 IP Prefix routes that indicates that they should resolve indirectly rather than using resolution information contained in the type-5 route itself. Depending on the type of overlay index, this resolution information may come from type-1 auto discovery or type-2 MAC+IP routes. For this feature the gateway IP address field of the type-5 NLRI is used as the overlay index, which matches the target IPv4 / IPv6 address in the type-2 NLRI. Other types of overlay index are described in RFC9136, but these are currently unsupported.

In EOS 4.22.0F, EVPN VXLAN all active multi homing L2 support is available. A customer edge (CE) device can connect to

Ethernet VPN (EVPN) networks normally require some measure of redundancy to reduce or eliminate the impact of outages and maintenance. RFC7432 describes four types of route to be exchanged through EVPN, with a built-in multihoming mechanism for redundancy. Prior to EOS 4.22.0F, MLAG was available as a redundancy option for EVPN with VXLAN, but not multihoming. EVPN multihoming is a multi-vendor standards-based redundancy solution that does not require a dedicated peer link and allows for more flexible configurations than MLAG, supporting peering on a per interface level rather than a per device level. It also supports a mass withdrawal mechanism to minimize traffic loss when a link goes down.

EVPN gateway support for all-active (A-A) multihoming adds a new redundancy model to our multi-domain EVPN solution. This deployment model introduces the concept of a WAN Interconnect Ethernet Segment identifier (WAN I-ESI).

The EVPN Gateway Data Center Interconnect (DCI) feature supports multihoming redundancy. This deployment model leverages a virtual Interconnect Ethernet Segment Identifier (I-ESI) to form an overlay ECMP across the EVPN DCI gateways. Recently, EOS added new features for managing the I-ES that improve traffic handling and convergence in certain failure scenarios:

This feature adds the ability for an L3 default gateway TEP in a Centralized Gateway topology to advertise its SVI virtual IP addresses to VARP MAC bindings and primary addresses to System MAC bindings using EVPN type-2 routes for EVPN VXLAN overlays. Two new commands, redistribute router-mac virtual-ip[next-hop vtep primary] and redistribute router-mac system ip are introduced to enable the redistributions. This would help the L2 TEP on the network to learn the default gateway IP without flooding an ARP request for the gateway IP. This feature is only intended for Centralized Gateway Topologies.

Smart System Upgrade (SSU) provides the ability to upgrade the EOS image with minimal traffic disruption. SSU is supported on a standalone L2 only VTEP with EVPN-VXLAN configuration with the following scale is supported as of EOS 4.32.0F with the following scale numbers

Typical WiFi networks utilize a single, central Wireless LAN Controller (WLC) to act as a gateway between the wireless APs and the wired network. Arista differentiates itself by allowing the wireless network to utilize a distributed set of aggregation switches to connect APs to the wired network. This feature allows a decentralized and distributed set of aggregation switches to bridge wireless traffic on behalf of the set of APs configured to VXLAN tunnel all traffic to those aggregation switches, or their “local” APs.

If any two policies use the same filter interface and the same priority, then an additional dynamic policy will be created to ensure the delivery of packets matching both of the original policies. There is a limit on how many overlap policies can be created and it is configurable with a range between 0 to 10 with a default value of 4. Currently, we exclude policies configured as inactive in the overlap policy limit calculation. With this new feature, we exclude policies that have an expired duration from the overlap policies limit calculation.

This enhancement is to display the number of packets that were ECN (Explicit Congestion Notification) marked by the

Administrative Groups (AG) provide a way to associate certain attributes or policies with links, enabling network administrators to control the routing decisions based on specific criteria. Extended Administrative Groups (EAG) are an extension of AG which allow a larger range of admin groups to be utilized for various Traffic Engineering (TE) purposes within a network. EAGs are defined in a new sub-TLV for IS-IS link attributes, separate to AGs, however they are considered as one within EOS. The EAG feature in EOS allows the range of administrative color to be increased from 0-31 to 0-127.

Use an External Certification Authority (ECA) to ensure secure communication and authentication with CloudVision..By default, Streaming Agent and other applications communicate with CloudVision using mutual-TLS certificates signed by a local certificate authority (CA). You now have the option to integrate CloudVision with Venafi,  an external CA, to sign and verify these certificates.

Starting EOS 4.15.0F, EOS can monitor (for long durations) low error rate errors on all fabric links. It

The 7250X and 7300 series use an optimized internal CLOS design with multiple port ASICs interconnected via Fabric

With the 18.0 release, Access Points (AP) can also use LAN2 as the Uplink Port. If both the LAN Ports are available as Uplink, the AP monitors both ports equally. Only on the first AP boot will AP consider LAN1 as the default Uplink, and LAN2 will be the failover. If LAN1 and LAN2 are connected and LAN1 fails to receive any packets, the AP can fail over to LAN2 as the Uplink Port and will continue to operate on the same uplink even if LAN1 is active again. 

With the 19.0 release, Access Points (AP) can seamlessly switch between LAN 1 and LAN 2 as the Uplink Port without disturbing the client connectivity and without any reboot. For the list of enhancements done for the same feature in the previous release, see the 18.0 TOI.

Fair Adaptive Dynamic thresholds (FADT) provides efficient allocation of shared packet buffer resources amongst various virtual output queues. FADT is useful when queues are getting congested and buffer resources should be allocated in a way tdat prioritizes certain queues while avoiding starvation of lower priority queues. the scheme works on each incoming packet by calculating instantaneous queue threshold based on available free resources. Queue buffer threshold is calculated as:

The 7280E and 7500E series are Virtual Output Queues (VOQs) based multi chip systems where there is a VOQ for all the

Fallback PBR policy enables an alternate policy to be active when PBR policy attached to an interface is being

Fast poll counters allow for rapid collection of a basic set of MAC counters on supported platforms at a very high frequency.

This feature is to permit rapid restoration of outbound traffic on ECMP groups that have a mix of ports from Supervisor1(Linecard1) and Supervisor2(Linecard2) cards. In the context of the supported platforms, these are referred to as Uplink ports and have names starting with Eth1/ or Ethernet1/ (Linecard1) and Eth2 or Ethernet2/ (Linecard2).

This feature is to permit rapid restoration of outbound traffic on LAG (port-channel) groups that have a mix of ports from Supervisor1(Linecard1) and Supervisor2(Linecard2) cards. In the context of the supported platforms, these are referred to as Uplink ports and have names starting with Eth1/ or Ethernet1/ (Linecard1) and Eth2 or Ethernet2/ (Linecard2).

This document describes the Fec Dampening feature. When hardware FEC / ECMP resources usage go above the platform limit, Ale (HW Abstraction layer ) deletes some routes in the anticipation of freeing up some more hardware FEC resources to allow newly created FEC to get programmed. The above logic of deleting/unprogramming the route may lead to unnecessary traffic drop in the following cases of transient FEC resources overflow.

The FEC (Forward Error Correction) traffic analyzer is designed to estimate the performance of the FEC layer, identify error statistics, and the source of correlated errors on physical interfaces.

In the 17.0 release, CV-CUE introduces FEED. FEED is a network dashboard that presents a timeline view of all the detected anomalies in the network. CV-CUE curates the FEED by continuously monitoring and proactively detecting anomalies in the network. It also analyzes the cause of the anomaly and provides dynamic suggestions to mitigate the issue. The administrator can analyze the issue, the AI-based recommended action, and then decide on the best approach to mitigate the issue. Feed also lets administrators go back in time and understand anomalies that occurred in the past.

FIB compression allows us to program routes into the hardware more efficiently. Routes are programmed in the route

This feature introduces a per-VRF table “FIB route count” for hardware FIB tables, and associated actions.

The Filter managed service action filters packets on the Service Node (SN) interface and supports optional VLAN tagging. Utilizing ACL rules, the system forwards or drops matched traffic. Traffic tagged with a VLAN exits the interface (Tx) after processing through the action chain. VLAN tagging specifically facilitates traffic steering in Switch-less SN deployments, where the forwarding plane relies on VLANs. This configuration produces no functional impact when the SN connects directly to a DMF switch within the fabric.

Filtered mirroring allows certain packets to be selected for mirroring, rather than all packets ingressing or egressing a mirror source port.

MPLSoGRE Filtered Mirroring is a specialized version of Mirroring to GRE Tunnel and Filtered Mirroring in which

Directed broadcast ACL allows inbound broadcast IP packets with source IP address as one of the permitted hosts and denies the rest of the directed broadcast traffic. Destination broadcast address of the IP packet should be the broadcast address of an interface with directed broadcast enabled. This feature gives a global command to configure sets of the permitted hosts via field-set. 

With the 19.0 release, you can apply filters to report data before generating or scheduling a report. Previously, you manually filtered out the relevant data from the generated report. Applying filters before generating a report helps streamline the data, speeds up report generation, and improves its readability. With filters, you can create a customized report based on your specific needs.

Organizations may have multiple access points (APs) of different models operating with various firmware versions. As an organization, you may want to designate a specific version as a compliant firmware version for a certain model. Assigning a compliant firmware version helps network administrators identify non-compliant AP models by generating notification alerts.

Systems with support for Arista secure boot protect against tampering of the BIOS firmware & Aboot by write-protecting the BIOS SPI flash before EOS is loaded (refer to the “Security model” section in the secure boot TOI for details). While effective at protecting against unauthorized changes made from EOS, such a mechanism has limitations. For example, it is ineffective at protecting against physical reprogramming of the contents of the BIOS SPI flash, tampering through privileged serial console access, undiscovered security vulnerabilities in BIOS upgrade mechanism, etc.

This document describes the CLI introduced to reallocate ECMP FEC banks on different levels in a hierarchical FEC configuration. Users may run out of entries on a certain level with other levels having little to no usage, and this CLI reconfigures the ECMP FEC entries to meet the requirements of the user.

Disabling the flooding of broadcast, multicast, and unknown unicast traffic into the VXLAN fabric can significantly reduce bandwidth consumption in the VXLAN underlay. This is particularly beneficial in use cases where such traffic is unnecessary. This feature, exclusively supported with EVPN, allows for the selective flooding of ARP and/or ND traffic, offering further control over bandwidth usage.

With the 16.0 release, CloudVision Cognitive Unified Edge (CV-CUE) introduces the following enhancements to Floor Plans: 

Control Plane Policing (CoPP) classifies control plane traffic into different classes (e.g., IGMP, LLDP, PTP, OSPF) and applies rate limiting per class using queues shaping. However, when multiple flows within the same CoPP class share a single queue, a single high-rate flow can consume the entire allocated bandwidth for that class, starving other legitimate flows.

Latency and drop information help determine if there is a loss in a particular flow and where the loss occurred. A Service Node action configured as a DANZ Monitoring Fabric (DMF) managed service has multiple separate taps or spans in the production network and can measure the latency of a flow traversing through any pair of these points. It can also detect packet drops between any two points in the network if the packet only appears on one point within a specified time frame, currently set to 200ms.

Latency and drop information help determine if there is a loss in a particular flow and where the loss occurred. A Service Node action configured as a DANZ Monitoring Fabric (DMF) managed service has multiple separate taps or spans in the production network and can measure the latency of a flow traversing through any pair of these points. It can also detect packet drops between any two points in the network if the packet only appears on one point within a specified time frame, currently set to 200ms.

This feature provides a way to distinguish groups of flows within encrypted GRE tunnels. That enables downstream forwarding devices to process multiple flows in parallel while maintaining packet order within individual flows. Parallel processing offers the opportunity for significant aggregate throughput improvement.

This feature provides a way to distinguish groups of flows within encrypted IPsec tunnels. That enables downstream forwarding devices to process multiple flows in parallel while maintaining packet order within individual flows. Parallel processing offers the opportunity for significant aggregate throughput improvement

Receive Side Scaling (RSS) which is also known as multi queue receive, distributes network receive flows across NIC card multiple hardware queues.

The agent DmaQueueMonitor provides visibility into packets coming up to the CPU via CPU queues. Packets are continuously sampled on monitored queues and kept available for reporting when a CPU congestion event occurs.

This feature enables detection of abnormal system flows (total in vs. out packet counters) by showing packet loss

Flow control is a data transmission option that temporarily stops a device from sending data because of a peer data overflow condition. If a device sends data faster than the receiver can accept it, the receiver's buffer can overflow. The receiving device then sends a PAUSE frame, instructing the sending device to halt transmission for a specified period.

Forced periodic ARP refresh adds support for a mechanism that allows forcing ARP/NDP refresh requests to be sent in periodic intervals independently of ARP/NDP entries' confirmed time in the kernel. By default, when a neighbor entry gets confirmed by various processes such as ARP synchronization between MLAG peers, an ARP refresh request is not sent for at least another duration of ARP aging timeout (or ND cache expiry time for the IPv6 case). This feature provides support for a configuration to force sending refresh requests at the configured ARP/ND aging timeout regardless of the last confirmed time.