UNS OpenHub 2.1.143: from device connectivity to analysis
The public Runtime 2.1.143 release brings together a more inspectable path from device connection and UNS mapping to retained history, analysis, and event capture. The OPC UA and MQTT bridges are public; the example recipe kit remains a preview.
A value from a device is only the start. An operator still needs to know which machine it belongs to, whether a status is text or a measurement, where its history is stored, and how to examine it later. UNS OpenHub Runtime 2.1.143 brings several recent updates into a clearer working path, from a reviewed device connection to repeatable analysis and service operations.
Runtime 2.1.143 and Bootstrap 1.0.128 are publicly released. The release is available to install; publishing it does not upgrade an existing Runtime automatically.
Connect a device without guessing its identity
The OPC UA Bridge 2.0.1 and MQTT Bridge 2.0.1 are public releases under the MIT licence and are available through the official signed add-on catalog. An operator can review a source endpoint, OPC UA nodes or MQTT topics and payload fields, then map selected signals to an existing UNS Object ID.
Creation and ingestion are deliberately separate: review the mapping, create a stopped connection, and start it explicitly. For example, a device might publish a numeric temperature and a text status such as IDLE. Keeping those types distinct makes the first values easier to check in the namespace and later in Explore.
A small portable device-recipe kit can reduce repeated setup work, but it is still marked as a preview. It carries example mappings and checksums, not a signed package or a substitute for choosing the local endpoint, target identity, and credentials.
Carry the signal into history and analysis
The release strengthens the path from a live value to a useful historical view. Open a temperature and status together, choose the time window and the correct retained history source, then save and reopen the analysis. The same saved context can be placed on a dashboard or examined through the historical query and Excel export.
Explore, dashboards, SQL, and Excel are existing Runtime capabilities; this update focuses on keeping their selected sources, time range, data shapes, and provenance connected through the workflow. A number and a state transition are not forced into one misleading series. Operators can also distinguish a current publisher mapping from retained history, so removing a connection does not by itself mean its archived observations should disappear.
Historical analysis requires the relevant storage and query services to be available, including UNS Archiver and API Global. Their current public releases are Archiver 5.2.20 and API Global 4.1.12.
Prepare named services, then choose where to run them
The shared Micro services view makes it easier to inspect a service across a selected controller scope. A named instance represents one logical service, so the same application package can support separate workers with independent names and configurations. Installing a version, preparing a controller to run it, and starting the process remain distinct operator actions.
Reviewed runtime configuration can be prepared on eligible controllers, while stopped copies can be brought up to date before an operator selects where to start the service. A running process is not silently restarted just because its prepared configuration changed; pending changes and conflicts remain visible for review. Process status and resource signals belong to the controller where that copy actually runs, rather than being copied from a stale or stopped instance.
This is more deliberate service placement, not an unrestricted fleet scheduler or a claim of physical multi-host high availability. A move still depends on a compatible, prepared target and an explicit operator action. Secret values remain local to their controller; shared configuration can refer to them without distributing the values themselves.
Go from an event to the exact values captured
Triggers and Captures are not a new rule engine. The update improves how operators create and inspect them: a draft stays disabled until it is reviewed and explicitly enabled, and a saved definition is shown separately from runtime confirmation and actual events or sessions.
A closed Capture session can be opened in Explore as that specific session, carrying its time bounds and session filter into the query and export. Two captures that share an output therefore need not be blended into one broad time-range result. Clients that update or delete automation definitions must read the definition revision and send expectedRevision, so a concurrent edit cannot quietly overwrite another operator's change. These controls support reviewable event-driven work; they do not claim exactly-once archival or replace deterministic control.
See the path in a concrete production example
The updated fictional hot-rolling example makes the data path tangible. An operator can prepare a furnace charge, follow material through process stages and through merge or split events, then compare batches using their lineage and recorded measurements. The simulator release 6.1.16 provides the example; it is not a universal model for every plant or production process.
For the full device-to-analysis procedure, see the public connectivity guide. The integrations page shows which building blocks are available and which, including the Ignition Edge path, remain in pilot preparation.