HiveMQ Edge 2026.10 is Released
Overview
HiveMQ Edge 2026.10 improves the OPC-UA adapter following input from a customer POC and removes MQTT-SN support from the runtime. This release fixes three OPC-UA bugs that affected tag schema generation, struct-array fields, and JSON-to-Int64 writes. 2026.10 also resolves a long-standing bug that caused the adapter to ignore the configured publishingInterval, and closes two CVEs in the Netty stack through routine dependency maintenance.
OPC-UA: configured publishing interval is now honored
What It Is
The OPC-UA adapter now uses the configured publishingInterval from the moment a subscription is created on the server. Previously, the value from OpcUaToMqttConfig was silently ignored and every subscription ran at Eclipse Milo's 1000 ms default regardless of what the user configured.
How It Works
The fix applies the configured value in the initial CreateSubscription request on both the first-connect path and the transfer-recovery path. The adapter also logs the server's revised publishingInterval. When the server clamps or rounds the requested value, the actual cadence is visible in the log without inspecting the source code.
How It Helps
Some OPC-UA deployments need a faster cycle than the 1000 ms default. Examples include fast-changing process signals and sub-second control loops. For these cases, setting publishingInterval in the adapter configuration now applies as documented.
OPC-UA: tag-schema and write-path fixes
What It Is
Three OPC-UA bugs surfaced during a customer POC are fixed in 2026.10:
- Custom-namespace struct fields no longer fail schema generation. Tags on structured nodes can use a custom non-structure DataType, such as a String subtype or an enum in the server's namespace. These tags previously caused schema generation to return an HTTP 500 — Problem accessing node error. Schema generation now resolves field types through the DataType tree directly, so custom-namespace fields render correctly: scalar where the builtin is scalar-mappable,
anyfor opaque types likeVariant/DataValue/DiagnosticInfo. - Struct fields that are arrays render as arrays. A struct field with
ValueRank ≥ 1(for exampleInt16[10]) was previously flattened to a scalar in the generated schema. Array fields now keep their dimensions and render as arrays of the element type. Array-of-struct fields remain a known limitation; the schema builder still emits scalar items for them. - JSON integers can be written to OPC-UA
Int64/UInt64nodes. Writing a small JSON integer (anything that fits in a Javaint) to anInt64/UInt64node used to throw "Can not convert 'N' of class 'IntNode' to 'Int64'" and silently drop the message, contradicting the documented JSON-integer →Int64mapping. The write path now accepts bothIntNodeandLongNode, matching the behavior already in place forUInteger.
How It Helps
Customers integrating against real-world OPC-UA servers with devices and PLCs where structured nodes, custom namespaces, and mixed-width JSON payloads are common. These servers now return correct schemas and successful writes out of the box, instead of opaque HTTP 500 errors or silently dropped messages.
Breaking Change: MQTT-SN support removed
HiveMQ Edge no longer ships MQTT-SN support. The codec, UDP listener, and supporting runtime are removed. Existing configurations still parse, and the <mqtt-sn> and <mqtt-sn-listeners> blocks are still accepted. However, the runtime response changes:
- A configuration with an active
<mqtt-sn-listeners><udp-listener>now logs ERROR and does not start the listener. - A configuration that only carries the deprecated
<mqtt-sn>block now logs WARN.
The MQTT-SN spec is not finalized and the upstream library we forked has stalled. With no production users on MQTT-SN, the carrying cost was no longer justified.
Action required for operators: remove <mqtt-sn-listeners> from your config and migrate affected devices to MQTT over TCP.
Dependency and Security Updates
Routine dependency maintenance during the 2026.10 cycle included patched releases of Netty that close two CVEs flagged by Snyk against the codebase. netty-codec-http and netty-codec-compression are now at 4.2.15.Final (up from 4.2.12), past the fixed versions for CVE-2026-41417 (HTTP request smuggling via CRLF in setUri()) and CVE-2026-42583 (resource exhaustion in the compression codec). No application code change was needed — the version catalog bumps pulled the fixes in through the existing Renovate pipeline.
The same cycle also modernized the frontend toolchain: a broad set of minor npm dependencies across the HiveMQ Edge frontend and OpenAPI tooling were updated, and the build moved from pnpm 10 to pnpm 11. Together these reduce supply-chain risk and keep the developer-facing build stack current.
Get Started Today
Use the download link to get HiveMQ Edge 2026.10, or find us on GitHub and Docker:
Get started by running
docker run --name hivemq-edge --pull=always -d -p 1883:1883 -p 8080:8080 hivemq/hivemq-edge
Or clone our repository
git clone git@github.com:hivemq/hivemq-edge.git
You may also try out our Helm Chart
helm repo add hivemq https://hivemq.github.io/helm-charts && helm repo update
HiveMQ Team
Team HiveMQ brings together deep expertise in MQTT, Industrial AI, IoT data streaming, UNS, and Industrial IoT protocols. Follow us for practical deployment guidance, best practices for building a secure, reliable data backbone, and insights into how we are shaping the future of connected industries.
Our mission is to transform industrial data into real-time intelligence, actionable insights, and measurable business outcomes.
Have questions or need support? Contact us. Our experts are ready to help.
