Agent Environment
gcr.io/datadoghq/agent:7.54.0

Describe what happened:
When the pods start with single step instrumentation (probably on a new node but not sure), we get this error about the datadog agent jar and the pod goes into crashloopbackoff. Deleting the problematic pod, we get a new pod started correctly without any error.

Describe what you expected:

Pods should start with datadog agent instrumented everytime.

Steps to reproduce the issue:

I cannot be sure about this but better to test on a new node and first time pod start on a new node.

Additional environment details (Operating System, Cloud provider, etc):

AWS EKS 1.29, Self managed node groups. m5.2xlarge nodes. Spring grpc, Datadog agent with single step instrumentation enabled.