• Blog

 
 

QoS Basics

ContentsIntroductionQoS FeaturesI. ClassificationII. Marking (Rewrite)III. Transmit Queues & Port Shaping Introduction Quality of service is the ability to provide different priority to different applications, users, or data flows, or to guarantee a certain level of performance to a data flow. QoS processes apply to traffic that flows through Ethernet ports and control planes. These processes can modify data fields (CoS or DSCP) or assign data streams to traffic classes for prioritized handling. QoS Features Consider the following topology for the following examples. I. Classification Quality of Service defines a method of differentiating data streams to provide varying levels of service...
Continue reading →

Mixed ASM and SSM Multicast Client Behaviour

Multicast deployments can be difficult to operate and understand as an environment grows or clients with varying capabilities exist. One example of this is when both Any Source Multicast (ASM) and Specific Source Multicast (SSM) clients are connected to the same device in the same broadcast domain. In such a scenario, it is common that whilst the ASM client will receive all sources, the SSM client will want to receive a specific source. To illustrate this, consider the topology below:     Two sources exist, 101.0.0.1 and 101.0.0.2. Client1 is using IGMPv2 and is therefore only able to request the...
Continue reading →

SNMPd on CVP

ContentsIntroductionGoalInstallation/ConfigurationEdit the snmpd.conf file with vi (make sure to save a copy of the original)Enable and start the snmpd serviceRestart the firewalld serviceSuccess:Supported MIBs: Introduction Many users have SNMP-based monitoring tools deployed to monitor their tools and generate alerts when issues are detected. The goal is enable CVP to be similarly monitored with these tools and provide basic server-level status information. This enables users to detect basic server faults or resource exhaustion. The goal is not to expose any CVP-specific information (for example which services are up or not) via custom MIBs at this point.   Goal SNMPd on CVP...
Continue reading →

Troubleshooting based on Control Plane Policing (CoPP) for Sand Platform

ContentsObjectiveWhen should we perform Copp based troubleshooting ?Command to be used:1. CoppSystemL3DstMiss –2. CoppSystemL2Ucast –3. CoppSystemIpUcast –4. CoppSystemL2Bcast –5. CoppSystemL3LpmOver-If there is Latency (or intermittent drops) in Network Objective This article is going to explain few important Copp queues of Arista Sand platforms- 7280 series, 7020 series, 7500 series that are responsible for handling various control-plane packets and how to troubleshoot based on that.  Copp or Control Plane Policing are various queues or buffers in Arista switches for handling various plane packets. Different Copp handles different kinds of control-plane packets and each has its own significance. The control plane policing...
Continue reading →

LACP Fallback and PXE/Provisioning in Openstack

As shown in the diagram above, we have a Server which needs to PXE to fetch its Kernel/RAM Disk/etc via Provisioning vlan (205). The server is dual homed to LeafA and LeafB which are in MLAG. Use Case: We don’t want to waste a dedicated NIC on the Server, on the Switch and cabling just for PXE boot. We would like to leverage an LACP bond used for other Openstack VLANs (e.g. internal API, Storage, Storage-mgmt, etc) for PXE/Provisioning too. Problem: LACP is configured on the 2 Leaf switches, but the server doesn’t even have an OS to form the...
Continue reading →

Troubleshooting Unknown Unicast Flooding

This article covers some of the scenarios in which the switch will flood Unknown Unicast traffic.  Unknown unicast flooding is observed when traffic received is destined to MAC addresses not learned on the switch. Since MAC addresses are usually learned from the source MAC for the host, switches could either : Never learn the MAC address if we never receive a packet with that source MAC address Time out such MAC addresses if we do not receive traffic with that source MAC for a duration greater than the MAC address timeout interval. Flush MAC addresses on receiving an STP TCN...
Continue reading →

Troubleshooting LACP

ContentsOverviewIntroductionInitial Checks1. Interface status of the bundled interfaces2. LACP valid mode combinations3. Some interfaces not bundled in the port-channel4. The device is not accepting dot1q-tagged LACP PDUsFurther Troubleshooting1. Are we sending/receiving LACPDUs?2. LACP rate-limit exceeded3. Monitoring LACP states4. Identifying the egress port-channel interface Overview This document aims at providing the basic steps to follow for troubleshooting LACP related issues. Introduction The Link Aggregation Control Protocol (LACP), described by IEEE 802.3ad, defines a method for two devices to establish and maintain link aggregation groups (LAGs). When LACP is enabled, our switch can configure LACP-compatible ports into a LAG (also called a...
Continue reading →

Troubleshooting Congestion

ContentsWhat is CongestionExpected Behavior of the switchCauses of Congestion Fan-InMicroburstBuffering capacity of the networking devicesDetecting Congestion on Arista switches using LANZEnabling LANZ and viewing statsMethods to prevent congestionQoSFlow Control / Priority Flow ControlBuffer tuningIncreasing Bandwidth of the interface What is Congestion Congestion happens when a device receives more traffic than it can buffer or send out of an egress port. Whenever the network is congested, we can observe symptoms such as packet loss, slowness, low throughput and high latency.   Expected Behavior of the switch When a switch receives a packet, it takes an action to either forward or drop it....
Continue reading →

Installing/Uninstalling a software patch on an Arista device

ContentsOverviewInstallation StepsPatch installation on single supervisor/fixed system switchesPatch installation on dual supervisor/modular switchesUninstalling a patch from the device Overview This article details the steps needed to install/uninstall patches on Arista devices. Installation Steps Check the hardware platform and the EOS version running on the device using the “show version” command: Switch#show version Arista DCS-7150S-52-CL-F Hardware version: 02.01 Software image version: 4.20.11M Architecture: i386 If specified, ensure the device to which the patch needs to be applied is running on a relevant EOS release and is of the relevant platform. Patch installation on single supervisor/fixed system switches Step1: Download the patch...
Continue reading →

DHCP Server on Arista EOS

Overview In this document, a DHCP server will be installed on the switch, and configured to deliver IP addresses for a subnet/VLAN interface on the switch serving both IPv4 and IPv6 address families. All examples are demonstrated on Arista EOS release 4.20.7M. Install the DHCPD extension An extension is a Red Hat Linux .rpm file type or an archive of .rpm files in a .swi format. For this example the latest .swi package for DHCPD is downloaded from Arista repository to the switch and installed as a recognized extension. Download the extension: Please make sure you have reached out to Arista...
Continue reading →

SNMPv1/v2c Configuration

ContentsObjectiveConfiguration Enable SNMPEnable Trap notificationTroubleshooting   Objective The document is aimed to provide basic configuration needed on Arista switches to be able to monitor the device statistics using an SNMP monitoring tool. Configuration  Enable SNMP SNMPv1: The Simple Network Management Protocol, defined in RFC 1157. This version uses plaintext community name for interacting with the SNMP monitoring server. SNMPv2c: Community-string based Administrative Framework for SNMPv2, defined in RFC 1901, RFC 1905, and RFC 1906. This version also uses plaintext community string. Hence, the security level is the same as SNMPv1.  To query Arista switches using SNMP v1 or v2c, define a community...
Continue reading →

Tap Aggregation PTP Sync & Transparent Clock (TC) PTP Multicast Configuration

Contents DescriptionPlatform Compatibility Topology DiagramsConfigurationNotesNY1NY2NY3NY4  Description This article covers the configuration required to run the PTP TC with multicast transport environment, and Tap Aggregation DUT time sync with PTP GM. Platform Compatibility DCS-7020 DCS-7280SE/ DCS-7500E DCS-7280R/RX DCS-7500R/RX DCS-7150 DCS-7050X/XX DCS-7060X/XX Topology Diagrams     Configuration In this PTP implementation example, the NY1 will act as BC and NY4 as GM. Both are configured as PTP BC clocks but NY4 due to its lower Priority1 value will win the BCMA election and becomes the master. The NY2 and NY3 are configured as E2E TC clocks and will require PIM to forward the...
Continue reading →

Configure Daemon Script to Collect Data

ContentsSummaryCreate Script FileExample bash scriptCreate an EOS daemon processVerify Script is RunningStop the daemonView Log Output Summary The EOS daemon command is typically used to run TerminAttr (Streaming Telemetry Agent) on Arista switches. The daemon command can also be used to create custom daemon processes to do just about anything you wish. It can be useful in troubleshooting and collecting data over a period of time. In the example below, we create a bash script that runs 4 show commands and adds a timestamp to the log and repeats the process every 5 seconds. The output of the script commands...
Continue reading →

Hybrid cloud connectivity with Arista’s Extensible Operating System (EOS) and Amazon Web Services (AWS)

ContentsMotivationIncludes and excludesSolution components overviewPublic cloud detailsVPC detailsSubnet detailsEC2 instance detailsPrivate cloud detailsRouter instances including autonomous system number (ASN)Topology, subnets and interfacesHosts attached to the data centre routerInterconnecting the public and the private cloudSolution overviewAWS configurationDefining the customer gatewaysDefining the VPN gatewayDefining the Site to Site VPN connectionsEnable dynamic routing in the routing tableEOS configurationConfigure data centre router dcInterfaces and BGP routing within the private cloudConfigure customer gateway 1 router cgw1LicensingInterfaces and BGP routing within the private cloudIPSec configurationDynamic routing towards AWSConfigure customer gateway 2 router cgw2LicensingInterfaces and BGP routing within the private cloudIPSec configurationDynamic routing towards AWSSolution verificationAWS VPN...
Continue reading →

Recovering a switch from Aboot

Objective During the upgrade process, there could be multiple reasons for the switch to get stuck in Aboot such as corrupted image, unsupported image, incorrect boot file, or no image to boot from. The aim of this document is to describe the recovery procedure of the switch from Aboot shell.   Procedure Step1: Check if the relevant EOS image is already present in the /mnt/flash Aboot# cd /mnt/flash ← Change the directory to flash Aboot# ls ← list the contents of flash EOS-4.21.6F.swi  boot-extensions   persist scripts FSCK0000.REC   config_match rc.d  startup-config If the relevant EOS image is present, continue from...
Continue reading →

Wireshark LUA dissector for “Mirroring to GRE with Timestamping” feature

Description Time stamping is an important tool for network engineering and performance analysis. EOS-4.21.3F adds support for payload timestamping of all GRE encapsulated mirrored packets at line rate (only supported on the 7500R/7280R/7500R2/7280R2 series). A timestamp is taken on ingress and inserted into the GRE encapsulated mirrored packet payload at egress. More details about this feature can be found at https://eos.arista.com/eos-4-21-3f/mirroring-to-gre-timestamping This feature introduces an enhancement to Arista Timestamp Header Wireshark Dissector described in https://eos.arista.com/analyzing-packet-header-timestamps-in-wireshark/ How to Use All necessary steps on how to use new Wireshark Dissector described at https://github.com/mpergament/wireshark-arista-tagagg-dissector

IGMP Host-proxy

Interfaces on the switch can be configured to serve as IGMP host proxies. An IGMP host proxy exchanges IGMP reports (joins/leaves) between networks whose connection does not support PIM along network boundaries. Let’s take the example shown below: The customer network connects to the sender network through the edge switch’s Ethernet 1 interface, which is configured as an IGMP host proxy. PIM is enabled within the publisher and customer networks but not on the connection between the networks.   When only “ip igmp host-proxy” is configured on the interface, IGMP reports will be sent for any (*,G) or (S,G) entry...
Continue reading →

Configuring the Campus Voice VLAN

ContentsOverviewIntroductionTo provision a VLAN-unaware IP PhoneTo provision a VLAN-aware IP PhoneLLDP’s Role Overview This article is intended to discuss how to configure the Voice VLAN (phone VLAN) on a 720X Arista switch. Introduction In the Campus environment, there are myriad devices which can connect wirelessly or wired. Switches need to be able to classify these devices and forward the traffic appropriately. For IP phones, this is of utmost importance as Voice traffic and Voice signaling need to be treated with priority over generic data traffic to ensure quality of service and minimum to no loss. This article focuses on how...
Continue reading →

Troubleshooting Dequeue deletes on 7280/7500 devices

ContentsOverviewKnown CausesSyslogTroubleshooting DeqDeletePktCnt DropsRecommendations Overview On a 7280 / 7500 devices, any known unicast packet which arrives on an ingress port is classified by the ingress processor based on priority level and egress port, and enqueued in the Virtual Output Queue (VOQ). The same would apply for BUM traffic if ingress only replication is enabled. The packet is enqueued on the ingress chip buffers until the egress packet scheduler issues a credit grant for the given packet. DeqDelete drops indicate stale packets in VOQ, i.e, packets which have been in the VOQ for more than 500ms without getting credits. These...
Continue reading →

ZTP with Arista Switches

ContentsOverviewIntroductionTo provision the switch through Zero Touch ProvisioningActions performed by the switch in ZTP modeCanceling Zero Touch ProvisioningFailure scenario(s)Things to be aware of while using ZTPUse any of the following commands to calculate the MD5 checksum Overview This article is intended to discuss zero to one of ZTP. Introduction Zero Touch Provisioning(ZTP) is a feature that allows users to initially provision Arista network switches without user interaction. The switch enters ZTP mode whenever the device comes up without a start-up configuration in flash. It remains in ZTP mode until a user cancels ZTP mode, or until the switch retrieves a...
Continue reading →

Follow

Get every new post on this blog delivered to your Inbox.

Join other followers: