Error apigateway pod cannot resolve NIM instance

[Alarnos]

Issue Summary

When using NGINX Plus R33 and NIM 2.18 with the standard /config declaration version 5.2, the following error occurs:

requests.exceptions.ConnectionError: HTTPSConnectionPool(host='nms.example.com', port=443): Max retries exceeded with url: /api/platform/v1/instance-groups (Caused by NameResolutionError("<urllib3.connection.HTTPSConnection object at 0x7fa7543572f0>: Failed to resolve 'nms.example.com' ([Errno -2] Name does not resolve)"))

Environment

NGINX Plus Version: R33
NIM Version: 2.18
Config Declaration Version: 5.2
Deployment: Kubernetes

Steps to Reproduce

Deploy NGINX Plus R33 on Kubernetes.
Use a DNS server on the same network as the Kubernetes cluster, configured for node resolution.

Observed Behavior

When the NIM instance DNS name is used in the nim_host Postman variable to publish the example API (v5.2), the following error occurs:

requests.exceptions.ConnectionError: HTTPSConnectionPool(host='nms.example.com', port=443): Max retries exceeded with url: /api/platform/v1/instance-groups (Caused by NameResolutionError("<urllib3.connection.HTTPSConnection object at 0x7fa7543572f0>: Failed to resolve 'nms.example.com' ([Errno -2] Name does not resolve)"))

However, the NIM instance resolves correctly when using the nslookup command inside the pod.

Expected Behavior

The NIM host should be correctly resolved by the Declarative API instance.

If additional logs or screenshots are required to clarify the issue, I can provide them upon request.

[fabriziofiorucci]

Thank you @Alarnos could you please share the output of the following commands, run inside the declarative API pod:

nslookup -type=a nms.example.com

ping nms.example.com

cat /etc/resolv.conf