Infohub (Flow)Reference

Consumer Capability Matrix

Which of the 13 consumers support which capabilities, read from the connector source.

Updated 15 Oct 2026Infohub Docs

Not every consumer can do everything. A connector declares the capabilities it implements, and Infohub only offers you the features its declared capabilities support — which is why an option that exists for one consumer is simply absent for another.

This table is read from the connector source at v7.2.0, so it says what the shipped code declares rather than what the sales sheet claims.

What the capabilities mean#

Capability	What it gives you
Event period comments	Publishes comments on event-sliced values.
Event period exceptions	Publishes exceptions on event-sliced values.
Event period values	Publishes measure values sliced by event period.
Value exceptions	Publishes exceptions raised against measure values.
Measure values	Publishes interval measure values.
Value comments	Publishes comments attached to measure values.
Event schemes	Publishes the event scheme definition and its periods.
Test button	The connection dialog can test the connection before you save it.
Configuration migration	Existing connection configuration is upgraded automatically when the connector version changes.
Event attributes	Publishes event attribute values.
Catalogues	Lists the databases or catalogues on the server so you can choose one when configuring the connection.
Namespace browsing	The source exposes its tag tree, so you pick tags in Flow Config rather than typing paths.
Event period comments	Publishes comments recorded against an event period.
Raw values	Publishes the raw value stream rather than interval summaries.
Working directory	Keeps state on disk between runs, such as a file sequence number or a resume token.
Lazy namespace browsing	Loads the tree branch by branch as you expand it, and supports find-next searching — how a namespace with hundreds of thousands of tags stays usable.
Measure limits	Publishes limit definitions and their values.
Infohub context	Receives Infohub's own connection details, so the consumer can read back from the Infohub database.
Persistent connection	Holds the connection open between retrievals instead of reconnecting each time.

Which consumers support what#

Event period comments — 12 of 13

AWS Kinesis Data Streams · Azure Event Hub · CSV · Canary · Endpoint Consumer · Flow Consumer · MQTT Broker · Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake · SparkplugB Broker

Event period exceptions — 12 of 13

Event period values — 12 of 13

Value exceptions — 12 of 13

Measure values — 12 of 13

Value comments — 11 of 13

AWS Kinesis Data Streams · Azure Event Hub · CSV · Canary · Flow Consumer · MQTT Broker · Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake · SparkplugB Broker

Event schemes — 10 of 13

AWS Kinesis Data Streams · Apache Kafka · Azure Event Hub · CSV · Canary · MQTT Broker · Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake

Test button — 10 of 13

Apache Kafka · Canary · Endpoint Consumer · Flow Consumer · MQTT Broker · Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake · SparkplugB Broker

Configuration migration — 5 of 13

MQTT Broker · Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake

Event attributes — 5 of 13

MQTT Broker · Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake

Catalogues — 4 of 13

Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake

Namespace browsing — 4 of 13

Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake

Event period comments — 4 of 13

Microsoft SQL Database · MySQL Database · PostgreSQL Database · Snowflake

Raw values — 2 of 13

Apache Kafka · SparkplugB Broker

Working directory — 1 of 13

SparkplugB Broker

Lazy namespace browsing — 1 of 13

Flow Consumer

Measure limits — 1 of 13

Flow Consumer

Infohub context — 1 of 13

SparkplugB Broker

Persistent connection — 1 of 13

SparkplugB Broker

Important

Only a raw-values consumer can be attached to a tag. A Tag interval measure holds raw timestamped values, so publishing one requires the Raw values capability. Just 2 of the 13 consumers have it:

Apache Kafka · SparkplugB Broker

Attaching any other consumer to a tag produces a binding that cannot deploy — the Consumer Host rejects it with "does not implement the ICanPushRawValues interface". Infohub now refuses the attachment up front, in the Integration panel and from --attach-measure-consumer alike. Earlier builds accepted it and failed later, so a model carried over from one may contain bindings that have never worked.

To publish data from a tag through one of the other eleven, aggregate it to an interval measure first and publish that.

Reading this table#

A capability a connector does not declare is not a defect. Most sources genuinely cannot answer the question — a flat file has no namespace to browse, and a broker has no history to replay.
Declaring a capability means the connector implements the interface. It does not promise the server on the other end is configured to allow it; permissions and licensing still apply.
If a feature you expect is missing in Flow Config, check this table before raising it: the field is usually hidden because the connector never declared the capability behind it.

Consumer Capability Matrix

What the capabilities mean#

Which consumers support what#

Reading this table#

See also#