Introduction and Overview

Open Networking Enterprise Suite (ONES) is a Network Orchestration, Visibility, and Assurance solution for multi-vendor and multi-NOS operated Network Infrastructure. ONES provides a one-stop solution from delivering deep visibility into your datacenter networks to extending 24x7 support functions for SONiC. It also hosts a powerful analytics engine that assists users to identify network issues and troubleshoot their networks, in case of common network anomalies and disruptions.

ONES uses a YAML-based Device Discovery Template for adding switches and continuously collects streaming telemetry data from them to provide insights on;

• Data Center Inventory

• Network State

• Platform and System Health

ONES monitors various control and data plane metrics to provide these insights.

ONES orchestration provide network admins to automate the fabric configuration using configuration templates for provisioning physical interfaces, layer 3 configuration for building IP-CLOS fabric using BGP as a routing protocol including BGP-unnumbered, system services including NTP, SNMP, SYSLOG etc. ONES orchestration not only configures the fabric but also make sure the Fabric is operational by doing verifying the configuration at every stage.

ONES provides north bound API access for configurations originating from external orchestration tools.

Overview

Network Assurance helps the NetOps team validate policy and security compliance checks before making a change in network configuration, an intelligent set of proactive and predictive techniques that validate the Network for readiness without error, conflicts, and disruptions

ONES Assurance module allows you to measure key metrics on-demand so you can deliver SLAs of five-nines (99.999% availability) for your network.

ONES is backed by the Aviz support team located across 4 time zones so you can avail of 24x7 SONiC expertise for any version running on any Switch with any ASIC

• Collaborate with our SONiC experts to expedite your evaluations

• Speed up your SONiC troubleshooting SLAs to as low as 15 minutes regardless of the underlying Switch/ASIC platform

• Minimize operational delays by centralizing issues across multiple platforms

Refer to the "How do I reach Aviz Networks Support?" section of this document for more details

Why do we need Network Orchestration?

Orchestration refers to tasks or actions required to achieve a set of objectives for your Network Infrastructure operations

A centralized application like ONES translates these objectives into a network configuration template, applies and monitors to validate the operational efficiency and functionality

Automated tasks are performed on your Network Fabric in a purposeful order and each step is verified for success before moving to the next

ONES Orchestration - Overview

ONES Orchestration function on the application lets you compose, deploy, and validate network configurations across any SONiC, be it a Community version or a Vendor distro.

As part of the initial release, ONES Orchestration supports to

• Create and configure CLOS topology for ToR, Leaf, Spine, and Super-Spine layers

• Apply and validate configurations pre- and post-deployment

• Compare running configs against applied configs at any point

ONES Orchestration use cases are configured using a set of pre-defined YAML-based templates on ONES Web User Interface

Overview

ONES Telemetry Collector(s) and Analytics bring truly unparalleled visibility across all your switches running SONiC (both community and vendor distros), regardless of the underlying ASIC. ONES front end (UI), will enable network admins to;

• Manage inventory of your network devices running SONiC on Broadcom, Cisco, Marvell, Nvidia ASICs

• View the topology of the entire fabric across multiple hardware platforms, and network operating systems

• Monitor traffic, system health, bandwidth utilization, & more

• Track Switch CPU/memory consumption, bandwidth, link failures, traffic errors, and more in real-time

• Proactively identify and resolve issues that may lead to network downtime

• Instantly connect to individual devices for maintenance and troubleshooting

Supported Matrix

Cumulus support limited to available Open Config models, is considered by ONES as Agent-less

Agent-based vs Agent-less

SONiC-based switches require ONES Agents (Agent-based) to be installed on the switch being monitored, as a pre-requisite for ONES Telemetry and Orchestrator based functions to work.

• ONES Telemetry Agent

• ONES Orchestrator Agent

Proprietary NOS like Arista EOS, Cumulus, and Cisco NX-OS does not require an ONES Agent and instead leverage (Agent-less) feature. OpenConfig extends APIs that provide Network Telemetry information about the resources being monitored via (gRPC Network Management Interface) protocol to the ONES Application

ONES does not support Orchestrator-based functions on Proprietary NOS (non-SONiC).

Agent requirements

• SSH access

• SONiC version beyond 202012 or 202111 are supported

• Only x86 intel-based architectures are supported

This is the first release of the ONES product

Overview

There are three primary components on ONES Application

• ONES Orchestrator for Fabric Configuration

• ONES Telemetry Collector(s) for Network Visibility

• ONES Supportability

Installation Overview

ONES Installation follows the below steps in the order sequence of;

• License Readiness

• Preparing and Installing ONES Application machine

• Installing ONES Agents on SONiC Switches for Orchestrator and Telemetry

• Enabling OpenConfig on non-SONiC Switches for Telemetry

License Readiness

The installer allows a default capability on managing 8 devices without a license. Beyond this, the following license keys are required for proceeding with the Installation;

• ONES Application License (per application)

• ONES Agent License (per switch)

To obtain a license, contact support@aviznetworks.com providing the below details;

• License Type - Evaluation Trial or Full

• Devices Count - 32, 64, 128, 256, 512 or 1024

• Device Serial Numbers

System Hardware Requirements – ONES Application

In the current release, ONES can support managing up to 1024 devices. For ONES Application Installation, the system hardware requirements vary based on the number of devices to manage;

System Software Requirements - ONES Application

Customer Firewall Configuration (Ports to be opened)

NOTE: These port numbers should be available to use and all ports must be allowed in the firewall if the Database server and devices are in the different DMZ zone

Network Switch (Managed Node) Configuration Requirement

• SSH is enabled

• Network Reachability from ONES Application

• OpenConfig feature is enabled (for non-SONiC Switches)

Browser Requirements for ONES Web User Interface access

• Google Chrome version 107 or later

• Mozilla Firefox version 106 or later

ONES provides the following subscriptions to manage and monitor the devices.

• Work with Aviz Sales/Support contact to create an account on Aviz Networks Support Portal

• Login to https://support.aviznetworks.com with your account credentials

• Click on the Downloads section, under ONES, click to download ONES Release 1.0

• Copy ONES Release 1.0 package (tar.gz) to ONES Application server

Overview

ONES requires user to install the below agents on SONiC NOS to allow Network Orchestration and Visibility

• ONES Orchestrator Agent for Network Orchestration

ONES Application system - Installation Steps

• Enable super-user mode

user@ones-application:~$ sudo su

• Extract the contents of the downloaded ONES Installer tar.gz file to this folder

root@ones-application:~$ tar -xvzf ONES-GA.tar.gz

• Go to GA folder on the ONES application server

root@ones-application:~$ cd GA

• Run ones-installer.sh to Install the ONES Application function

root@ones-application:~/GA$ ./ones-installer.sh

NOTE:

1. By default, the installer has license for 8 devices

2. Hitting ENTER without entering a license will activate the ONES application to support a maximum of 8 devices

3. User can provide license key provided by Aviz to activate more device

• Enter ONES application (Host Server) IP for Web GUI Access

NOTE: The system might have multiple IP addresses, so a user needs to enter one of the IP that he wants to use to access the ONES Application dashboard

• Installation begins

• Access ONES Application Web GUI from a supported browser using http://<host-ip-address:3002

• Use Default credentials as below;

  • Username: superadmin

  • Password : Admin@123

• You will see the default Dashboard → Hardware view now

On the ONES application server, go to GA/ones_fm_agent

root@ones-application:~$ cd /GA/ones_fm_agent

Installation (Agent Install on multiple switches at the same time)

• Enter device details (Management IP, Username, Password and License key) on device_info.csv

root@ones-application:~$ vi device_info.csv

• Save the file

• Execute the installation script

root@ones-application:~$ python3 deploy_fmcli.py "install"

• Verify the Installation success by running the fmcli command on individual devices

root@Switch:~$ sudo fmcli

ONES User Interface - Features

• YAML-based Device Discovery

• Deep Telemetry for ASIC and Switch Hardware

• Device Inventory details on

  • Network Operating System (NOS)

  • Firmware versions - ONIE, BIOS, and CPLD

• Inventory Operations

  • Adding/Removing devices using YAML

  • Agent Status Monitoring

• Device Monitoring

  • Device Up/Down State based on Agent and Agent-less

  • Region and Zone Mapping

• Network Compliance with version checks on

  • Telemetry Agent

  • Orchestrator Agent

• Resource Trends

  • CPU and Memory Utilization

  • PSU and Fan Readings

• Topology View

  • Device Connectivity view across Roles and Location

  • Link/Connectivity Status

• Routing Protocol

  • BGP Neighbors

  • Advertised and Received Prefixes

• Orchestrator Use Cases

  • YAML-based Configuration push

  • Image Management via ZTP

• Licensing

  • ONES Application License

  • Telemetry Agent License

• • Add/Edit/Delete User

  • Role Management

• API Access for configurations originating from External Orchestration Tools

ONES allows users to leverage pre-defined templates, and customize them for Ports, IPv4/IPv6 Routes, BGP-Unnumbered, and Switch Services (NTP, SNMP, SYSLOG, ZTP, etc.) functions

On the ONES application server, go to GA/ones_t_agent folder

root@ones-application:~$ cd /GA/ones_t_agent

Installation (Agent Install on multiple switches at the same time)

• Enter device details (Management IP, Username and Password and the license key) on device_info.csv

root@ones-application:~$ vi device_info.csv

• Save the File

• Verify and Clean up old versions of Agents running

root@ones-application:~$ python3 ones_agent_uninstaller.py

• Execute the installation script can be used for installing telemetry agent on one or more devices in the data center.

root@ones-application:~$ python3 ones_agent_installer.py

• Navigate to Inventory

• This page gives the control to onboard the devices with two options

  • Add/Remove Devices using the YAML Editor

  • Upload the YAML file containing the device list

• Click on Add/Remove Devices

• Upload Device Inventory using YAML Editor

• Navigate to Inventory >> Devices >> Add/Remove Devices

• Make sure to use the correct indentation for the YAML files

• Make sure the Switch Name matches to Hostname of the switch

• Click Save & Apply

• ONES Application is now ready to manage the added devices

Overview

The monitor widget in ONES:

• Shows the complete topology view of the fabric

• The Topology view can be filtered by:

  • Region

  • Brick

• This page shows all the links and information connected to each other

• Low and high utilization of all components

• Input / Output errors on links

• ASIC detailed view of all managed devices

• BGP status

  • Neighbour Count

  • Neighbor status (Up/Down)

NOTE: If all the managed devices have the same Region/Brick ID, we do not get any filter ribbon

Topology

• Navigate to Monitor >> Topology

• This shows the complete Topology view, how the devices are connected and to which role they belong to

• We can put the filters to check the customized view of the Diagram by: Region Availability Zone Brick ID

NOTE: we do not get filter ribbon if all devices belong to the same Region/Availability Zone/Brick ID

• We can also check Down Links to check the topology those are having links in the shutdown state as per Role

  • Super Spine

  • Spine

• Users can easily filter the view as per the Region, Availability zone and Brick ID

Links

This page gives a view to the user for all the possible connected links between devices with a few more capabilities

• Navigate to Monitor >> Links

• This page helps a user to get the best view of the number of connections between devices with speed and other manufacturer details

• This page gives the exact view of the interface name, interface speed, transceivers and admin & operator status

Links View with Filter Ribbon

• We can also have a filtered view of links

• Choose a specific Role and Region to filter

• let's choose:

Platform

This page shows the latest utilization of the devices with CPU & Memory utilization, Temperature & Voltage of PSU, and fan speed in RPM

• Navigate to Monitor >> Platform

Platform Status

Platform Status is reported for the following components

• Roles

• SKU/ASIC

• Ports/Max Speed

Customized View

• We can check the health of the device as per some customization

• We can filter the devices by:

  • Roles

Role-based Customization

• We can choose a role using the available Role-based option

• 4 Roles available

  • Super Spine

• After selecting ToR input, here is the new view of only devices that belong to the ToR Role

Region-based Customization

• We can choose a role using the available Region-based option

• Two roles available

  • San Jose

• After selecting Asia Pacific input, here is the new view of only devices that belongs to the Asia Pacific Region

Role / Region based Customization

• We have the liberty to do the filtering by both combination Role-based and Region based at the same time

• As of now, we have two Region available

  • San Jose

• After selecting both Roles and Region-based filter, here is the new view of the device that belongs to the Spine role and parallel belongs to Asia Pacific Region

Per Device Status

• This Platform Widget also gives the option to check the extended capability view of the device

• Apart from this monitoring view, we can also verify/check extended feature sets like:

  • PSU Current (A)

• When we choose a specific device we get an output like this

Device Info Ribbon

CPU Utilization (%)

• Here we get the complete status of CPU utilization with a time range A complete status What was the utilization from starting to end

• To check a specific time detail we can hover the cursor to any level

Memory Utilization (%)

• Now here we get the status of memory utilization of selected device

• To check a specific time detail with memory utilization, we can hover the cursor to any level

CPU Temperature (C)

• This template shows the status of CPU temperature in degree celsius

• Here we get the status of all the CPU and Core running on the device

• To check a specific time detail for all the CPU/Cores, we can hover the cursor on any level

• We can also check Per CPU Core details

• We can see here the customized view per CPU Core

Fan Speed (%)

• Here we get the status of the FAN speed percentage level across all fans available on the device

• To check a specific time detail we can hover the cursor on any level

• We can choose to view the customized view of a few fans at once

• Here we can see the output of fan1 and fan7 in this customized view

• When we move the cursor to check the exact status of fans at a particular time

PSU Temperature (C)

• This widget provides the status of the Power Supply Unit (PSU) temperature (in degree celsius) across all power supplies available on a particular device.

• To check a specific time detail we can hover the cursor on any level

• We can choose to view the customized view of a few PSUs at once

• In our case, we have 2 PSUs, so now let's choose one of them and check the output

• When we move the cursor to check the exact status of the PSU at a particular time

PSU Voltage (V)

• This widget provides the status of the Power Supply Unit (PSU) voltage across all power supplies available on a particular device.

• To check a specific time detail we can hover the cursor on any level

• Same we can check the customized view by choosing one or two PSUs

• Let us choose one of the PSU to check individual Voltage, so here we have taken PSU2 to check individual Voltage

PSU Current (A)

• This widget provides the status of the Power Supply Unit (PSU) current in Amperes, across all PSUs available on a particular device.

• To check a specific time detail, hover the cursor on any level

• Same here as per other power supply customized views we can also check the customized view here by choosing specific PSUs

PSU Power (W)

• Here we get the status of Power supply power in Watt across all power supply present on the device

• To check a specific time detail we can hover the cursor on any level

• We can also check the customized view of this

Services Running

• The best widget here for Services

• We can also check the total number of count of services running on the platform

• This graph shows the red colour bar, red colour show at what time one of the services went down

• To check a specific time detail we can hover the cursor on any level

• Here we get the name and count of all the services running on the platform

Services CPU Consumption (%)

• This widget shows us the CPU consumption percentage level of all services / per service.

• Here we can see we have the option to check the consumption view of CPU

• To check a specific time detail we can hover the cursor on any level

• When we get this output we get a clear view of both CPU Utilization per service based

• We can also customise the view of this widget by choosing some specific application

• We have all the services tab here to choose

• Again we can check the view of all the services we need

• So here is the view of only the BGP service

• We can see the CPU and Memory consumption for these 2 services

Services Memory Consumption (%)

• This widget shows us the Memory consumption percentage level of all services / per service.

• Here we can also check the consumption view of only Memory.

• To check a specific time detail we can hover the cursor on any level

• Here we get a clear view of Memory Utilization per service based

• We can also customise the view of this widget by choosing some specific application

• We have all the services tab here to choose

• So here is the view of only the BGP service

ASIC

This page shows the view of ASIC Capacity and a few more details related to ASIC

• This widget Shows

  • Roles/Region per device

  • SKU and ASIC details per device

This ASIC Widget give us the control to get the output per Role and Region basis also

• Let's choose Leaf Role to get the customized view

• In the same way, we can customize the view by Region

Per Device Status

This widget gives us the capability to check the extended view of the Routes & ACL usage with a range of time

Click on any of the devices to get the extended view

BGP

This BGP Page shares the data of BGP neighbours present across managed devices

Neighbor View

this shows the status of the neighbour's details, the total number of neighbours, received routes, neighbour RID, BGP AS number & much more​​We have the option here to check the neighbour details and status of Routes​​We can click on neighbours to get more details about all neighbours connected

Customized View

• This page gives us the power to check neighbours as per Roles and Region-based

• Let's check how we can filter as per Role-based

• Here is the global view of all the Devices with BGP status

• Choose Leaf Role to check customized view

• After having only Leaf Roles here is the modified output

Traffic

• Using this widget we can check the input and output errors across all the devices

• This widget also shows the input and output packet per device

• Navigate to Monitor >> Traffic

• To check particular device information

  • The orange colour shows: the output packets

  • The grey colour shows: input packets

• This page shows the traffic drop rate per interface based and will be very useful while doing any troubleshooting for a traffic drop

• Using these details a user can check more details inside to fix the issue of dropping/discarding packets

Hardware (Default)

The dashboard provides the NetOps with an overview of the data centre. It contains the entire hardware inventory of the network and shows the status whether these switches are streaming or not streaming.

• After the Installation of ONES Application for the first time, the Dashboard is empty and Devices need to be onboarded for them to reflect

• Dashboard will be used to

  • monitor the status of an agent running on all the devices present

Components

• Navigate to Dashboard >> Components

Software

• Navigate to Dashboard >> Software

Interfaces

• Navigate to Dashboard >> Interfaces

• Using this page a user gets the status of

  • the cables utilized in the network

  • how many pairs of cable can be used for future topology (helps the admins in capacity planning)

This template is built for only IP-CLOS configuration to be done in this topology

• Device used

  • 2- Spine Devices

  • 3- Leaf Devices

  • 3- End Host

Topology

BGP-IP-CLOS Standard Template

• Navigate to Configurations >> Devices

• Allows you to configure new devices

• Supports valid YAML files

• You can download the sample YAML file, edit it, and upload it again with the desired configuration

NOTE: Follow the to know more about configuration

You can onboard all the devices on the application and can get a complete view of all the populated tables

• The Inventory tab has the below mentioned features:

  • Custom OS upgrade: Upgrade the device OS with any customised image. You need to provide the correct path to ensure the OS is updated successfully

  • OS upgrade via ZTP: Upgrade the device OS via Zero touch provisioning

  • Reboot devices: Reboot the device from a single click in the UI

  • Add/Remove devices from the dashboard: You can onboard the devices using the YAML file upload or via in-built editor in the UI

Devices

• Navigate to Inventory >> Devices

• Using this tab, user can:

  • Onboard the device to the application using ADD/Remove Devices

  • Upgrade the device using Custom Upgrade

Add Devices

• Upload Device Inventory using YAML editor

• Navigate to Inventory >> Devices >> Add/Remove Devices

• Use the below format to add devices to the ONES application

• Make sure to use the correct indentation for the YAML files

• Make sure the Switch Name matches to Hostname of the device

• Click Save & Apply

• ONES Application is now ready to manage the added devices

Feature Set

• Role / Region: Shows the device roles and regions

• SKU / ASIC: Shows the device hardware SKU and ASIC vendor

• Port / Max Speed

• PSUs / Fans: Shows the total number of Power supplies and Fans present on a particular device

• NOS Image: Shows the details of the network operating system running on the device and when it was last updated

• ONIE Version: Shows which ONIE version is running on the device and when last reboot time of the device

Remove Devices from the ONES Application

• Navigate to Inventory >> Devices >> Add/Remove Devices

• Remove all the inputs entered by the user

• Click Save and Apply

• Once the ONES Application detects, a user is trying to remove devices it shows one confirmation dialogue box to cross-check the status

• Now all the devices have been removed from the ONES Application

Custom Upgrade

• Select any of the devices to upgrade to the new version

• Click on Custom Upgrade

• put the new Image URL and then Submit

Upgrade via ZTP

• Select any of the devices to upgrade via ZTP

• Click on Upgrade via ZTP

• Click on Yes

Reboot Device

• Here we will see how we can reboot a device using

• We have the option to choose one or multiple devices at a time to reboot

• Choose one of the devices that we want to reboot

Overview

Using this feature setting we can set the acceptable and critical percentage level for the following device components

• Thresholds we can set for Components

• CPU Utilization

• Memory Utilization

• Fan Speed

• PSU Temperature

• PSU Voltage

• As per requirement, we can set different lower and higher threshold values for each component

• User can see the acceptable and critical number of devices in the Monitor Health tab view

• Navigate to Monitor >> Platform when any component breach the higher value

Thresholds

• Navigate to Settings >> Thresholds

• Change the values as per your requirements

• Update it to get these new settings live, after Save Changes, all these metrics will be reflected on devices metric pages under Inventory

Overview

Using this feature to

• Create new users and roles

• Remove and suspend existing user one by one

• Remove and suspend multiple users at the same time

• Password reset of existing users

Only super admin or Enterprise admin can perform these actions

Users

• Navigate to User Management >> Accounts >> Users

• Initially, we get one default Admin User Credential

• In this view, we get the User status and Role given to it and the Last login time by the user.

• On this User tab, we can Add new users and can remove or suspend any existing users

Users - Add New

• Navigate to Accounts >> Users >> Add

• Admin can add

  • Profile picture

  • Username

"on the first login by the user with given details, ONES application prompt with a screen to rest the password

• Now User is ready to login with valid credentials

Users - Reset Password

• Navigate to Accounts >> Users >> Edit User >> Reset Password >> Yes

• Using Admin credentials you can reset the user password

• Click on Reset Password

• Submit & Save

• After this user can try login with the temporary password, on the first login, ONES application actively asks to change the password

Users - Suspend User

• Navigate to Accounts >> Users >> ((Select Users you want to remove)) >> remove >> Yes

• We can remove multiple users at a time

• After this, the user will be removed from the database and cannot use credentials to login again

Users - Suspend User

• Instead of removing any user, we can also suspend the user

• Once we suspend a user it will not be removed from the database but it will be in an Inactive state in the database

• Later if we need we can restore the user to its active state

• Choose users & click on Suspend

Users - Restore User

• Navigate to Accounts >> Users >> ((Select Users you want to restore)) >> Restore >> Yes

Roles

• Navigate to Account >> Roles

• By default, the device comes with these 4 Roles

  • Super Admin

  • Enterprise Admin

Roles - Add User Roles

• Navigate to Accounts >> Roles

• Now let's add few permissions

  • Add/Remove Devices

  • Reboot Device

To access the ONES application, use Server IP with port 3002.

<host-ip>:3002

Use default credentials to login, refer page for default credentials

This template is built for only IP-CLOS configuration to be done in this topology

• Device used

  • 2- Spine Devices

This template is built for only IP-CLOS with L2-VXLAN configuration to be done in this topology

• Device used

  • 2- Spine Devices

Configuring Devices

Most fabric orchestration solutions available today are complex and often difficult to understand. ONES provides simple and effective tools, such as predefined templates (YAML file), to configure data centers at scale. ONES allows a customized way of configuring devices that includes enhancements to the standard configuration.

This template is built for only IP-CLOS with MCLAG configuration to be done in this topology

• Device used

  • 2- Spine Devices