# Dynatrace Otel collector not connecting to the containerized ActiveGate otlp endpoint

**URL:** <https://community.dynatrace.com/t/dynatrace-otel-collector-not-connecting-to-the-containerized-activegate-otlp-endpoint/53779>\
**Category:** OpenTelemetry\
**Tags:** opentelemetry, kubernetes\
**Created:** [October 20, 2025, 5:32am UTC](https://community.dynatrace.com/t/dynatrace-otel-collector-not-connecting-to-the-containerized-activegate-otlp-endpoint/53779 "2025-10-20T05:32:29Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![vinsontan\_anz](https://avatars.discourse-cdn.com/v4/letter/v/48db29/32.png) [@vinsontan\_anz](https://community.dynatrace.com/u/vinsontan_anz)\
**Post date:** [October 20, 2025, 5:32am UTC](https://community.dynatrace.com/t/dynatrace-otel-collector-not-connecting-to-the-containerized-activegate-otlp-endpoint/53779/1 "2025-10-20T05:32:29Z")

</div>

I have setup dynatrace components in a GKe cluster with telemetryingest capability which deployed dynatrace otel collector as part of the dynakube deployment. I have setup app pod to export otel traces to dynatrace otel collector but from the otel collector log, it shows bad gateway on connecting to dynatrace activegate otlp endpoint. I have setup custom properties on the spec.activegate to enable otlp.

The following is the log from the dynatrace otel collector:

```markup
2025-10-20T04:37:56.170Z  info  internal/retry_sender.go:133  Exporting failed. Will retry the request after interval.  {"resource": {"service.instance.id": "e0b013ae-1f92-4803-bd84-6d08e2b2a5e6", "service.name": "dynatrace-otel-collector", "service.version": "0.38.0"}, "otelcol.component.id": "otlphttp", "otelcol.component.kind": "exporter", "otelcol.signal": "traces", "error": "failed to make an HTTP request: Post \"https://dynakube-activegate.dynatrace.svc/e/pyx45104/api/v2/otlp/v1/traces\": Bad Gateway", "interval": "13.351872759s"}

```

In the otel collector otlp endpoint configmap:

```markup
apiVersion: v1
data:
  DT_ENDPOINT: https://dynakube-activegate.dynatrace.svc/e/pyx45104/api/v2/otlp
kind: ConfigMap
metadata:
  annotations:
    internal.operator.dynatrace.com/template-hash: "3896151151"
  creationTimestamp: "2025-10-17T07:45:16Z"
  labels:
    app.kubernetes.io/component: dynatrace-opentelemetry-collector
    app.kubernetes.io/created-by: dynakube
    app.kubernetes.io/name: dynatrace-operator
    app.kubernetes.io/version: v1.7.1
  name: dynatrace-otlp-api-endpoint
  namespace: dynatrace
  ownerReferences:
  - apiVersion: dynatrace.com/v1beta5
    blockOwnerDeletion: true
    controller: true
    kind: DynaKube
    name: dynakube
    uid: a219c91a-cd92-42f7-bbf5-a99e1dab9abe
  resourceVersion: "1760687116331343019"
  uid: 785693bf-0499-4ed7-aad3-679992f2de27

```

So what could be the reason for the bad gateway ?

---

<div class="post-metadata">

**Author:** ![MichalGebacki](https://dub1.discourse-cdn.com/dynatrace/user_avatar/community.dynatrace.com/michalgebacki/32/41585_2.png) [@MichalGebacki](https://community.dynatrace.com/u/MichalGebacki)\
**Post date:** [December 16, 2025, 3:59pm UTC](https://community.dynatrace.com/t/dynatrace-otel-collector-not-connecting-to-the-containerized-activegate-otlp-endpoint/53779/2 "2025-12-16T15:59:08Z")

</div>

Hello,&nbsp;[@vinsontan\_anz](https://community.dynatrace.com/user/96360)!

Were you able to find a solution or workaround for this issue? Is it still affecting your environment?

Thank you for any feedback!

---

<div class="post-metadata">

**Author:** ![vinsontan\_anz](https://avatars.discourse-cdn.com/v4/letter/v/48db29/32.png) [@vinsontan\_anz](https://community.dynatrace.com/u/vinsontan_anz)\
**Post date:** [December 16, 2025, 11:02pm UTC](https://community.dynatrace.com/t/dynatrace-otel-collector-not-connecting-to-the-containerized-activegate-otlp-endpoint/53779/3 "2025-12-16T23:02:02Z")

</div>

Yes, it was a bug on the otel collector config when Proxy is used in dynakube. The fix was released in dynatrace operator 1.7.2. The issue was basically the NO\_PROXY env variable defined in the Dynatrace otel collector was not defining the AG endpoint with .svc.
