Overview

Omniglass is an opinionated Zabbix distribution built for AV system monitoring. It compiles Zabbix from source with targeted patches, extends it through supported extension points, and packages everything into a single deployable stack with AV-native structure, device drivers, and operator-facing views. It is Zabbix — configured, extended, and ready for AV.

What it does

Omniglass gives AV teams a single platform to monitor devices, control systems, cloud management platforms, and UCaaS services (Zoom, Webex) across every room and site they manage. It adds three things to Zabbix that AV teams need and Zabbix doesn’t provide out of the box:

1. Structure. Organize monitored hosts into systems (conference rooms, video walls, broadcast chains) and locations (campuses, buildings, floors). Omniglass continuously reconciles this structure into Zabbix host groups, tags, and service objects so the data stays consistent without manual upkeep.

2. Collection. Monitor AV equipment that speaks proprietary or non-standard protocols (Crestron CTP, with more drivers planned), poll cloud management APIs (Cisco Webex Control Hub, Crestron XiO Cloud), and ingest data from UCaaS platforms. Drivers run on the proxy close to the devices; API integrations run through the workflow engine.

3. Automation. Run multi-step synthetic tests, ingest data from vendor APIs and webhooks, rotate OAuth tokens, fan out cloud API responses to individual hosts, build custom operator dashboards — all through Node-RED, pre-wired into the stack with Zabbix identity and sender integration.

Everything else — alerting, escalation, SLA reports, dashboards, SNMP, agent-based collection, user management — is Zabbix. Omniglass doesn’t replace any of that; it adds the AV-specific layer on top.

Here be dragons

Monitoring AV is hard. AV devices are agentless, vendor APIs are inconsistent or undocumented, and there is no industry-standard management protocol. Omniglass provides the platform and the extension points, but it is not a turnkey solution — you will need to build templates for your specific devices, write workflows for your vendor integrations, and define the system structure that matches your environment. The payoff is a monitoring foundation that scales, that IT and security teams can audit, and that doesn’t lock you into a single vendor’s tooling.

The problem

Most AV teams do not know precisely how many systems they have, what is in them, or whether they are working at any given moment. The primary reason for this is that AV devices are difficult to monitor: they are agentless, speak non-standard or proprietary protocols, and often lack complete management APIs and documentation. That limits the platforms AV teams can use for data collection, and leaves most teams stuck with manufacturer-specific point tools that only see their slice of the environment.

The result is siloed data, no unified view of system health across rooms or buildings, and no foundation for the kind of reporting that IT and security teams expect. Leaders cannot make data-driven decisions about service quality. Operators cannot produce an auditable record of what is deployed and how it is performing. For anyone who is living with this reality, it is clear that something has to change.

Zabbix is a strong foundation for infrastructure monitoring due to its flexibility in data collection methods, but out of the box it has no concept of AV systems, rooms, or device protocols. While Zabbix is highly extensible and fully open source, most AV teams lack the personnel, expertise, and time required to extend the software to their needs, much less maintain those extensions over time. The barrier to entry is simply too high.

Enter Omniglass: a Zabbix distribution designed for AV, with the necessary extensions built in and pre-configured so that teams can get up and running quickly.

Features

• Everything in Zabbix, plus AV: Alerting, escalation, SLA reporting, dashboards, SNMP, IPMI, agent-based collection, API integrations, and 25 years of production-grade monitoring infrastructure. Omniglass adds to this; it does not replace it.
• Systems & locations: Define AV systems (conference rooms, event spaces, signage clusters) as first-class entities with members, types, and parent relationships. Organize them by physical location with hierarchical location types, geographic coordinates, and custom metadata. The control plane reconciles this structure into Zabbix host groups, tags, and service objects automatically.
• Device protocol drivers: Agentless collection for AV-specific protocols (Crestron CTP, with more planned). Drivers run as Zabbix external checks on the proxy, so collection stays close to the devices.
• AV-native UI: Front-end modules add a systems explorer, system detail views, and location views directly in the Zabbix web interface. Operators work in a single UI.
• Reconciliation engine: The control plane watches for changes in Zabbix (host additions, group modifications, template updates) and continuously reconciles Omniglass-managed state. Drift is repaired automatically.
• Workflow automation: Node-RED is included and pre-wired for synthetic tests, device control sequences, vendor API integrations, and data transforms. Auth plugins bridge Zabbix identity into Node-RED. Every deploy is versioned in the database with diff and restore.
• Smarter SLAs: Minimum severity threshold for SLA downtime calculation, so informational alerts and low-priority warnings do not count against your service level reports. Submitted upstream as ZBXNEXT-10547.
• Removable: If Omniglass is removed, Zabbix continues monitoring. All projected artifacts (host groups, tags, items) remain in Zabbix and are self-explanatory.

Licensing

Omniglass and Zabbix (7.0+) are both licensed under the GNU Affero General Public License v3.0 (AGPLv3). In plain terms:

• You can use Omniglass internally for any purpose — commercial or otherwise. Deploying it to monitor your own infrastructure, your clients’ infrastructure, or anything else you operate does not trigger any source-sharing obligations. Most AV teams fall squarely in this category.
• You can modify it. Fork it, add features, write drivers. Your modifications are yours as long as you are using them internally.
• If you provide a modified version over a network to others — for example, if you host a modified Omniglass as a service for your customers — you must make your source code changes available under the same license. This is the “network use” clause (Section 13) that distinguishes AGPLv3 from standard GPL.
• Monitoring configuration is yours. Zabbix templates (XML/YAML), Node-RED flows (JSON), and .env configuration files are operational data, not modifications to the software. They do not trigger source-sharing obligations.

If you are deploying Omniglass to monitor your own environments or your clients’ environments (the typical AV integrator use case), the AGPLv3 license imposes no source-sharing obligations. The network-use clause only applies if you modify the software itself and make that modified version available to others over a network.

This is not legal advice. If your organization has questions about AGPLv3 compliance, consult your legal team.

Enterprise features

While Omniglass is fully open source under AGPLv3, we reserve the right to offer enterprise features and support under a commercial license in the future. No future enterprise offerings will impact, limit, or impose upon the functionality of Zabbix or the core Omniglass features that are available under AGPLv3.

Where to start

• Quick Start — Get a stack running for evaluation
• Architecture — System design, extension points, and patching model
• Systems & Locations — How Omniglass structures AV environments