During SO Macro Instantiation workflow request, First time SDNC is called (as far as I understand, even before ServiceAssignBB), SDNC queries AAI about service information. This request fails due to SDNC trying to access AAI via url "aai.api.simpledemo.onap.org" instead of service name ("aai" or "aai.$NAMESPACE").

This results with following example error in karaf logs (see attachment) and that leads to failing DG and sending error message to SO:

  "action": "createInstance",
  "requestStatus": "IN_PROGRESS",
  "statusMessage": "Error from SDNC: Failure calling AAI to update the service instance",
  "rollbackStatusMessage": null,
  "flowStatus": null,

My first guess was that it's caused by configfile "hardcoded" in docker image - `/opt/onap/ccsdk/data/properties/aaiclient.properties` that contained following entry:

org.onap.ccsdk.sli.adaptors.aai.uri=https://aai.api.simpledemo.openecomp.org:8443

as opposed to helm-provisioned `/opt/onap/sdnc/data/properties/aaiclient.properties` with correct URI (https://aai.onap:8443).

But after some hacking and provisioning this configfile to pod with this entry corrected, SDNC still attempted to contact AAI via wrong URL.

So to sum up, in order to fix this issue, SDNC should contact SDNC with it's inter-cluster DNS name, that currently is not used. How to achieve this is, however, not obvious for me as of now.

I've been working on master oom branch with SDNC images on version 1.7.6 as 1.8.0 is not yet released, however I think this issue should be solved on oom level.
If you need any more logs/info from my environment that I've been using, please ask.