Bad value for Kubernetes base pod name for all batchs from GKE Composer service

[AurelienGravier]

Hello,

The base pod name value detected by dynatrace for each process group of my batchs generated by GKE Composer service (Apache Airflow 2.6.3 and Cloud Composer 2.5.1) are not working as expected.

For example :

If the the full pod name is : rtl-est-okko-anr-anr-nw5h5oji

The Kubernetes base pod name value will be : rtl-est-okko-anr-anr-nw*h*oji

On another batch (using cron job) the Kubernetes base pod name value would be : rtl-est-okko-anr-anr-*

It impacts the split of process groups related to my Airflow batches because I use the namespace and kubernetes base pod name as container detection rule : 

So each batch execution creates a new process group.

Does anyone have an idea of how to solve this issue?

Thank you.

Regards Aurélien.

[ChadTurner]

Are you still having issues with this?