topic Re: IBM datapower monitoring in Extensions

IBM datapower monitoring

bernardifa — Mon, 23 Feb 2026 09:15:03 GMT

We installed IBM datapower on an EKS cluster; we are currently monitoring infrastructure metrics (cluster, node, namespaces, workload) using dynatrace operator
However services data are available only for few services and traces are not showing data even if I can see link between processes that calls and are called by IBM datapower
Is the IBM Datapower extension needed in order to have a complete visibility on IBM datapower?
Is possible with IBM Datapower extension installed to have performance data on missing services?

Re: IBM datapower monitoring

Esam_Eid — Thu, 25 Jul 2024 13:05:28 GMT

i have the IBM DataPower installed, its very useful for the below metrics

also, you can enable the below out of the box alerts from the anomaly detection settings

Regards,

Re: IBM datapower monitoring

bernardifa — Thu, 25 Jul 2024 14:42:45 GMT

And what about services data?
Are you able to track real requests? We only have a few services and some of them are opaque

Re: IBM datapower monitoring

AntonPineiro — Thu, 25 Jul 2024 17:02:59 GMT

Hi,

Dynatrace pop up message is pointing you to OneAgent SDK.

Best regards

Re: IBM datapower monitoring

36Krazyfists — Mon, 09 Feb 2026 22:57:39 GMT

DataPower does seem to have a built-in OpenTelemetry integration:
OpenTelemetry integration - IBM Documentation

I just found this myself, so I don't know how well it works, but I don't see why this wouldn't work with Dynatrace, since Dynatrace fully supports OpenTelemetry and the traces/spans should be distributed with other Up/Down stream traces, even if they are generated by OneAgent.

Re: IBM datapower monitoring

36Krazyfists — Fri, 20 Feb 2026 15:24:09 GMT

As I stated in my other reply in this thread, DataPower does have a built-in OpenTelemetry exporter and sampler that you can activate on the appliances.

One problem we've discovered with it though is that it doesn't seem to propagate context if it receives a request that already has OTEL trace data in it.

A lot of our API requests go through two DataPowers. An external one that sits in our DMZ and an internal one that does some transformations. We've insured that the traceparent and tracestate headers are getting propogated through them and, when we see the OTEL traces from the internal appliance, we see the traceparent header with the correct trace id from the external appliance where the request originally came in, but for some reason DataPower still creates a whole new trace id on the internal one. It makes no sense. We put a support case in with DataPower. We'll see how it goes.

Re: IBM datapower monitoring

Dant3 — Tue, 05 May 2026 02:28:23 GMT

Hey @36Krazyfists did you ever got any info back from IBM? We notice the same, setting all transactions as root_span: true.

Could you fix it?