Infohub (Flow)How-to

T01: Your First Connection

Connect to the built-in simulator, create your first retrieved measure, deploy it, and see values arrive.

Updated 15 Oct 2026Infohub Docs

What you will build: one connection, one folder, one metric, one retrieved measure, deployed and producing values.

Time: about 20 minutes.

You need: a Flow 7.2.0 instance you can edit, Flow Config connected to it, and the Flow Bootstrap service running. Nothing else — this tutorial uses the simulator that ships with Infohub.

Step 1 — Create the connection#

In the Data Sources view, create a new connection of type Flow Historian Simulator.

Name it Simulator. The simulator reads its tag definitions from a file installed with the product, so the defaults are correct.

Save it.

Step 2 — Browse the tags#

Browse the connection's tag namespace.

You should see a juice factory — boilers, mixers, fillers, water, and power. Look for 010-FT-001.PV. That is the boiler steam flow, and it moves in the region of 3.2 to 4.5 kg/s.

Tip

Browsing is the best possible first test. If tags appear, three things are proven at once: the plugin framework is working, the connection is configured correctly, and the credentials are valid. Always browse rather than typing tag names — it also guarantees you have the tag spelled exactly right.

Step 3 — Create the folder structure#

In the Model view you already have a root folder named after your instance — --create seeds it when the database is built.

Under it, create:

<Your Instance>
+-- Utilities
    +-- Boiler 1

Folders hold no data. They exist so people can navigate, and so templates and model attributes have something to attach to.

Step 4 — Create a metric#

Under Boiler 1, create a metric called Steam.

A metric groups the measures that describe one thing. It holds no values itself — the numbers live in its measures. By the end of the next tutorial, Steam will contain three measures.

Step 5 — Create the retrieved measure#

Inside Steam, create a measure:

Field	Value
Name	`Steam Flow`
Retrieval type	Retrieved
Interval	Tag
Connection	`Simulator`
Tag	`010-FT-001.PV`
Unit	`kg/s`
Format	`0.0`

Note two choices.

Interval is Tag. This measure stores raw values at whatever times the simulator produced them. It is not on the interval grid at all — it has no interval. That is deliberate, and the next tutorial builds the interval chain on top of it.

Unit and format are set now. Not later. The unit is free text that Infohub never validates, and the format controls how the value appears in every chart and report from here on. Setting them at creation is a habit worth forming.

Step 6 — Deploy#

Nothing you have built is running yet. In Infohub, configuring and activating are separate.

Open the Deployment view.

First, check the Platform's host name. It must match the machine running the Flow Bootstrap service. Flow Bootstrap claims a Platform by matching host name, so a mismatch means nothing will ever start — no errors, no data, everything looking correctly configured.

Then deploy:

The connection
The engine

Important

"Different" means your change is not live. Once something is deployed, editing it puts it in the Different state — it is running the old configuration. Nothing warns you. If a change appears to have no effect, this is the first thing to check.

Step 7 — Watch values arrive#

Within a few minutes, Steam Flow should start receiving values.

Look at the measure's data in Flow Config. You should see irregularly spaced values in the region of 3.2 to 4.5 — irregular because Tag-interval measures store values at the source's own timestamps, not on a grid.

If nothing appears#

Work down this list in order. Each step eliminates a class of problem.

Does the Platform's host name match the machine running Bootstrap?
Is Flow Bootstrap running? Check the Services list.
Is the connection deployed, and showing Deployed rather than Not Deployed, Different, or Faulted?
Is the engine deployed?
Can you still browse the namespace?
Does the tag match exactly — including case and no trailing spaces?
What does the Flow event log say? It is a custom log named Flow, not the Application log.

Nine times in ten it is item 1 or item 3.

What you have learned#

A connection is configured once and shared. It carries how to reach a system — address, credentials. The measure carries only what to read. One connection serves thousands of measures.

The model is a tree. Folders organise, metrics group, measures hold values.

Tag interval is not an interval. A Tag measure holds raw source values and sits outside the interval grid. You cannot chart it on an interval chart, and you cannot aggregate straight from it to Daily.

Configuration is separate from activation. Save, then deploy. This catches everybody once.

Browse, don't type. It proves the connection and gets the tag exactly right.

Try this before moving on#

Add a second retrieved measure to the same metric — Steam Pressure, from tag 010-PT-001.PV, unit MPa, format 0.00. It should read between about 1.2 and 1.5.

You should not need to deploy the connection again; it is already deployed. The new measure will be picked up.

That gives you a metric with two measures describing one piece of equipment, which is exactly the shape a real model takes.

Next: T02: Building an Aggregation Chain — putting those raw values onto the interval grid.