Aviz Networks details ONES 3.1 observability, orchestration and troubleshooting

Aviz Networks released ONES 3.1, adding NVIDIA Spectrum‑X observability, intent-based orchestration for smaller fabrics, and an on-premises conversational troubleshooting assistant in beta, along with new IP change tracking and default monitoring rule templates.

Research Overview

The vendor update describes ONES 3.1 as a new release of the Open Networking Enterprise Suite, with additions across observability, orchestration, and monitoring workflows.

The post highlights compatibility with Cumulus NOS and expanded telemetry for NVIDIA Spectrum‑X platforms, plus capabilities for IP tracking, alerting, and rule management.

Key Findings

ONES 3.1 extends observability to NVIDIA Spectrum‑X NOS versions and adds telemetry for Inventory, environment details, firmware versions, CPU and memory utilization, transceivers, interface counters, LACP, BGP, RoCE metrics, and queue counters.

The release also introduces an “AI assistant: Conversational Troubleshooting (BETA)” and a widget for tracking network device IP changes in real time with alerts when changes are unexpected.

Technical Breakdown

In Spectrum‑X Observability, the post states that ONES 3.1 builds on existing support for Cumulus NOS while extending compatibility to NVIDIA Spectrum‑X platforms running the latest NOS, with a centralized dashboard that includes top GPU utilization for GPU-accelerated servers.

For orchestration, ONES 3.1 adds Fabric Manager capabilities using an intent-based orchestration GUI, including configuration execution and an editor window, configuration comparison, and backups before upgrades or reboots.

Operational Impact

The update adds default rule templates for critical metrics that can be enabled via one click for anomaly detection and alerts, with added monitoring coverage for Docker CPU and memory utilization, Docker down status, and unhealthy device detection.

The post also lists additional enhancements including inbound and outbound traffic comparison, VLAN metrics monitoring, Docker status transition monitoring, a dashboard for top CPU and memory-consuming services, transceiver inventory visualization, unhealthy device notifications, and logical grouping of MCLAG interfaces in topology.

Blog Signals brief: This fact-based summary covers ONES 3.1’s reported additions for observability, small-network orchestration, conversational troubleshooting in beta, IP change tracking, and expanded monitoring and dashboard features for enterprise network operations.