A thought on Monitoring ServiceNow Integrations (part 1): Intro

[Richard Smith]

I was asked recently by a customer how they might bring together all the integrations relevant to their ServiceNow platform together in one place. So they could have a single pane of glass through which they could see all integrations and their current status. Imagine a RAG status against everything. That would be cool.

There are a number of solutions on the platform that help us understand what the state of integrations might be. The Service Graph Connection Center, The integration hub dashboard to name two.

These solutions show a subset of the integrations that exist though, not all of them. They are great at what they do but they do not cover "all" integrations. They also don't really fully take into account what might constitute an "error" might be. There's a good reason for this though; not all integrations are equal.

For example, consider an ad-hoc email inbound integration. The fact that it does not get triggered might not be an issue at all. What about two inbound table API integrations? One might be expected to be hit daily but another monthly. It might be fine for one to include zero records to update, but for another, we always expect many records.

The point is: what constitutes Green, Amber or Red is dependent on the integration itself. How often should, or must it run? Are there any thresholds for the data it should contain?

I think there's actually then pretty straightforward technical solution to bringing all the integrations monitoring together in one place, but way before we define what that technical solution looks like you must first go through a process of gathering all the right info together and documenting it. You must make sure it is sustainable and maintained because your integrations will evolve over time and you will add new ones. You must consider the building blocks carefully.

1. Governance & Preparation
2. The building blocks
3. Bringing it all together