Monitoring

Monitoring is a feature which lets you perform arbitrary actions based on the state of channels and devices in your system.

For example, you could generate a ticket for support when a channel with the Production tag is stopped for more than 1 hour. Or, scrape message counts and performance data for specific channels and forward that to a metrics stack.

As this feature is based on the Custom Code control, there is a lot of scope for including other functionality - whether from .NET, Nuget, or your own assemblies.

Add a new Monitoring group

Add a name and then click the Save button (ctrl + S)

Next, switch to the Monitoring tab and review the API. In the image below is an example of looping through each statistic exposed. Note that statistics are nested by Group, Tab, Channel, and then device - mirroring the layout of the management user interface. Note that a number of the exposed objects are dictionaries, so you can access directly by key, if desired.

A contrived sample showing looping through all statistics.

By default, the OnStateChangedAsync callback is only invoked when state or statistic information changes, and only contains the changed objects. This means when the Connexion service first starts, your code will receive all statistics once, and subsequently only receive deltas.

If you want all statistics included with every invocation of OnStateChangedAsync, set the Reset property to true (per line 46 in the above image).

You can use the Tabs & Channels tab to restrict the objects included in the OnStateChangedAsynccallback.