Getting UNAVAILABLE: Socket closed while connecting to Elastic APM Server

We are trying to perform a POC to get nginx traces to Elastic APM Server using the grpc protocol.

otel agent -- > APM server is not working and giving the error UNAVAILABLE: Socket closed

although the flow otel agent --> otel collector --> APM server is working

Both the agent and otel collector are deployed on same K8's env.

Otel collector configuration looks like based on example https://github.com/esigo/nginx-example

collector.yaml

    exporters:
      otlp:
        endpoint: "<apm server url>:8200"
        headers:
          # Elastic APM Server API key
          Authorization: "Bearer <token>"
        tls:
          insecure: false
          insecure_skip_verify: true

Steps to reproduce
Try to install ingress-nginx and point the endpoint to APM server

helm upgrade --install ingress-nginx ingress-nginx/ingress-nginx
--namespace ingress-nginx --create-namespace --set controller.opentelemetry.enabled=true
--set controller.service.type=NodePort -f ingress/controller-config.yml

controller-config.yaml

controller:
  opentelemetry:
    enabled: "true"
  config:
    log-level: "debug"
    enable-opentelemetry: "true"
    opentelemetry-operation-name: "HTTP $request_method $service_name $uri"
    opentelemetry-trust-incoming-span: "true"
    otlp-collector-host: "<apm server url>"
    otlp-collector-port: "8200"
    otel-exporter-otlp-headers: "Authorization=Bearer <token>"
    otel-max-queuesize: "2048"
    otel-schedule-delay-millis: "5000"
    otel-max-export-batch-size: "512"
    otel-service-name: "nginx-proxy" # Opentelemetry resource name
    otel-sampler: "AlwaysOn" # Also: AlwaysOff, TraceIdRatioBased
    otel-sampler-ratio: "1.0"
    otel-sampler-parent-based: "true"

What is the expected behavior?
The Agent should be able to send the traces to APM Server as APM Server also support the grpc protocol

What is the actual behavior?
We are getting this error

[Error] File: /tmp/build/opentelemetry-cpp-v1.8.1/exporters/otlp/src/otlp_grpc_exporter.cc:67[OTLP TRACE GRPC Exporter] Export() failed with status_code: "UNAVAILABLE" error_message: "failed to connect to all addresses; last error: UNAVAILABLE: Socket closed"

Additional context
Add any other context about the problem here.

Start with the ingress-nginx controller configuration and collector.yaml, then compare the direct OTLP gRPC path to the working collector path. Review the reported otlp_grpc_exporter.cc error and the APM Server endpoint, TLS, and authorization settings. Done means identifying why the direct connection closes and documenting a configuration that sends traces successfully.