k8sattributes not functioning correctly on Azure AKS

[kadhamecha-conga]

Component(s)

processor/k8sattributes

What happened?

hi,

The k8sattributes is not behaving as expected when deploying resources on Azure Kubernetes Service (AKS).
The same k8sattributes and resources deploy and function correctly on Amazon EKS.

debug logs:

thanks,

Collector version

0.73

Environment information

Environment

AKS
1.30

OpenTelemetry Collector configuration

      k8sattributes:
        extract:
          labels:
          - key: app
          - key: release
          - from: pod
            key: service.namespace
            tag_name: service.namespace
          - from: pod
            key: service.name
            tag_name: service.name
          metadata:
          - k8s.pod.name
          - k8s.namespace.name
          - k8s.deployment.name
        filter:
          node_from_env_var: KUBE_NODE_NAME

Log output

`AWS
Resource attributes:
     -> service.image.version: Str(202501.2.4)
     -> service.name: Str(test-api)
     -> service.namespace: Str(testing)
     -> service.version: Str(1.0)
     -> service.instance.id: Str(9d1ae59c-6963-4bb5-9fa5-719570f952f2)
     -> k8s.cluster.name: Str(utest1)
     -> deployment.environment: Str(testenv)
     -> k8s.pod.ip: Str(10.12.116.56)
     -> k8s.pod.name: Str(test-api-7546bc9865-sn9c2)
     -> k8s.namespace.name: Str(test-app)
     -> k8s.deployment.name: Str(test-api)

AZURE
Resource attributes:
     -> service.image.version: Str(202501.1.38)
     -> service.name: Str(test-api)
     -> service.namespace: Str(testing)
     -> service.version: Str(v1)
     -> service.instance.id: Str(19ebb4c4-208b-4085-92c1-e2d9af19642f)
     -> k8s.cluster.name: Str(utest1)
     -> deployment.environment: Str(testenv1)
     -> k8s.pod.ip: Str(10.48.2.44)`

Additional context

No response

[github-actions]

Pinging code owners:

• processor/k8sattributes: @dmitryax @fatsheep9146 @TylerHelmuth @ChrsMark

See Adding Labels via Comments if you do not have permissions to add labels yourself.

[bacherfl]

Hi @kadhamecha-conga! Is this issue occurring consistently, or from time to time? This could be related to a similar issue where e.g. the deployment name of a pod was missing in the attributes: #37056 - The fix for that will be included in the next release (v0.119.0) if I'm not mistaken, so it might be worth to try again with that version as soon as it becomes available

[kadhamecha-conga]

@bacherfl , k8s metadata is not at all coming.

[bacherfl]

Ok then it is problably a different issue than the one i linked. Can you try to set the log verbosity to debug and see if there are any error messages logged by the collector?

[atoulme]

Please also try to reproduce with a recent collector build, as 0.73.0 is old enough it won't support k8s 1.30.