Zabbix 8.0 LTS: what's new, features, and what changes

On November 5, 2026, Zabbix officially launched Zabbix 8.0 LTS. Readers who follow the Target blog already know part of what was planned: in September, we published a series on the version roadmap, from observability and AI novelties to the performance core, the new interface, and what comes after 8.0. Now the version is available.

The vendor message is direct: 8.0 places Zabbix firmly in the observability market, for both IT and operational technology (OT) environments. The change rests on four pillars that complement one another: APM, CEP, AI, and OT.

More data, more tools, more noise: the scenario 8.0 faces

Zabbix starts from a diagnosis any infrastructure team recognizes. Four problems often appear together:

• Too many tools. Metrics live on one platform, traces on another, and logs on a third. The sum raises operational cost and scatters incident investigation.
• Unbounded data volume. IoT devices, dependence on third-party APIs, and growing use of AI multiply what must be collected and stored.
• Noise for decision-makers. Uncorrelated events reach operations, management, and leadership as a stream of alerts, with no clear indication of what matters most.
• IT and OT in silos. Each side has its own leadership, budget, and tools, although the operation of a factory or building depends on networks and servers. That intersection rarely has full coverage.

The four pillars of 8.0 were designed to address these points.

The four pillars of Zabbix 8.0 LTS

APM: metrics, traces, and logs in one place

Zabbix consolidated itself in metrics monitoring, one of the main observability data sources. Application traces and logs were left to other tools or to modules and integrations built by partners. In 8.0, those data become native, based on OpenTelemetry, the open instrumentation standard.

The gain appears when the failure is not in the infrastructure. Because collection follows OpenTelemetry, instrumentation the company already has can be reused, without vendor lock-in.

The caveat is that the application must be instrumented, which usually involves the development team. Zabbix sums up the pillar as more functions with fewer tools.

CEP: fewer alerts, faster root cause

Complex Event Processing is Zabbix’s new event correlation engine and replaces the previous, simpler model that depended on manual configuration. For each event, it seeks to answer the question that consumes the most operations time: is this a cause or a symptom?

The engine groups related events, eliminates duplicates, suppresses what does not matter, and enriches events with context, including custom logic in JavaScript. Through 7.x, indicating which problems were symptoms of another was manual work on the Problems screen.

CEP arrives together with APM for a simple reason: traces and logs greatly increase data volume, and more data without correlation becomes only more noise. Because most incident response time goes to investigation, not correction, reaching root cause sooner shortens the incident and opens room for preventive action.

AI: talk to Zabbix, not only operate it

8.0 brings an official MCP (Model Context Protocol) server that connects to Zabbix the AI agents each company prefers to use, from different providers and even several at once. In the launch webinar, Zabbix cited Claude, ChatGPT, and Gemini as examples.

These agents can read metrics, traces, and logs, change Zabbix data, and respond in natural language, by text or voice. A manager can ask how the e-commerce environment is doing at that moment. A managed services provider can request a weekly report with customer service SLAs every Monday at 8 a.m.

For those who view AI cautiously, some details weigh heavily: the MCP server uses Zabbix API tokens, and what each agent can do depends on the permissions of the user associated with that token. Configuration should be done carefully, because MCP can also execute actions on the instance—not only query data. AI acts as operational support, with data from the company’s own environment.

OT: scale to unite IT and OT

The fourth pillar answers an old question about Zabbix: how far it scales. 8.0 accepts ClickHouse, a non-relational database, as an optional history back end, and MySQL, PostgreSQL, and Elasticsearch remain supported.

Zabbix reported that in tests the new architecture delivered up to ten times more performance, with less CPU and disk use, and passed one million values per second—results obtained in a test environment, which do not guarantee production performance. As a reference, the largest Zabbix environments in the world today operate around 200 thousand values per second, and large Brazilian environments sit between 15 and 20 thousand. We detail ClickHouse in our article on the 8.0 core.

That scale makes it possible to treat IT and OT on the same platform. OT is in access control, turnstiles, climate control, and elevators in a building, and in machines on a factory floor or in agribusiness operations. Zabbix has monitored these environments for a long time. The novelty is bringing both worlds together on unified dashboards and correlating, for example, a network event with a production stoppage or lost sales.

Pillars that reinforce one another

The four pillars make more sense together. APM increases the volume of data collected, CEP keeps that volume from becoming noise, AI changes how everything is queried and acted on, and the new architecture provides the scale needed to unite IT and OT.

With this integration, Zabbix supports its new positioning: a universal open source observability platform for IT and OT.

Beyond the pillars: interface, dashboards, and cloud costs

8.0 also brings changes operations will feel day to day, such as tables with customizable columns, dashboard export and import, new widgets, and the official iOS and Android app, announced as “coming soon.”

We cover these novelties in the article on the new interface. There is also a cost panel for AWS, Azure, Google Cloud, and Oracle Cloud and new templates such as Kubernetes Cluster, Hyper-V, and Cisco Secure Firewall, all listed on the official what’s new page.

When to upgrade to Zabbix 8.0 LTS

The lifecycle already gives concrete reasons to plan the move. Full support for Zabbix 7.4 ended on 11/12/2026, and for 7.0 LTS on 06/30/2027. Even so, the upgrade requires preparation. Zabbix assesses that it is not complex, but has steps that demand attention, such as new minimum database versions and API access, disabled by default for all functions on new installations; on existing installations, this rule applies to newly created functions.

Target Solutions is a certified Zabbix partner and supports companies from assessment of the current environment through adoption of the new features.

Talk to our specialists.

About Target Solutions

Target Solutions has spent more than 15 years helping organizations evolve their infrastructure, monitoring, observability, automation, and operational intelligence operations. Across that journey, it has participated in projects in complex environments such as telecom, technology, government, energy, and large enterprises, helping increase operational efficiency and service reliability.

As a Zabbix Certified Partner in Brazil, Target Solutions supports the deployment, evolution, sustainment, and optimization of enterprise Zabbix environments, integrating monitoring with ITSM, observability, and operational processes at scale.

About the author

Larissa Perestrêlo, telecommunications engineer on the Target Solutions team.