<StarlingX 2.0.1> Unable to Integrate with Open Source Mano

Bug #1850745 reported by ANIRUDH GUPTA
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
StarlingX
Won't Fix
Low
Unassigned

Bug Description

Brief Descriptor
--------------------------
I am trying to integrate StarlingX 2.0 with Open Source Mano 6.0, but facing errors as the public endpoints for every services are not exposed.
I have also configured helm endpoint domain from the below link and updated the wild card entry in dnsmasq.conf file on StarlingX machine

https://docs.starlingx.io/deploy_install_guides/r2_release/openstack/access.html#configure-helm-endpoint-domain

Issue
---------------------------------

I am getting error 503
code: 503\\n description: 'VIM Exception vimconnUnexpectedResponse EndpointNotFound: public\\n endpoint for image service not found

However this was not the case with StarlingX previous release.
I had successfully integrated it.
This is because now the openstack services are now containerized and URL is not publicly exposed.

Can someone please update the steps in order to expose all services public endpoints outside StarlingX

Revision history for this message
Ghada Khalil (gkhalil) wrote :

@Anirudh, Suggest you send an email to the stx-discuss mailing list for input on this.

Revision history for this message
ANIRUDH GUPTA (anyrude10) wrote :

I think this is an issue as I am not able to use Openstack Services Rest API's from a remote server without passing Host.

The Hostname varies with each service and also it is not publicly accessible.

Steps to Reproduce
--------------------------------
The command to get tokens mentioned in the below link works locally on StarlingX setup.

https://docs.starlingx.io/deploy_install_guides/r2_release/openstack/access.html#rest-apis

In order to get a token from a remote server, we need to add Host corresponding to each service in curl command

curl -i -H "Content-Type: application/json" -H "Host: keystone.hsc-starlingx-company.com" -d '{ "auth": {
    "identity": {
      "methods": ["password"],
      "password": {
        "user": {
          "name": "admin",
          "domain": { "id": "default" },
          "password": "Centos@123"
        }
      }
    },
    "scope": {
      "project": {
        "name": "admin",
        "domain": { "id": "default" }
      }
    }
  }
}' http://192.168.121.141:80/v3/auth/tokens

TOKEN=<Set token retrieved from above request>

curl -i http://192.168.121.141:80/v2.1/flavors -X GET -H "Content-Type: application/json" -H "Accept: application/json" -H "X-Auth-Token:${TOKEN}" -H "Host: nova.hsc-starlingx-company.com" | tail -1 | python -m json.tool

where 192.168.121.141 - OAM IP which is publicly accessible.

keystone.hsc-starlingx-company.com - Keystone endpoint
nova.hsc-starlingx-company.com - Nova Endpoint

Issue
-------------------------

It is not practically possible to pass FQDN hostname for every service request in the URL.
So, is there any way StarlingX automatically routes the ingress request internally ?
Or
Can we install all the Openstack Helm Chart services on Nodeport?

Please suggest the best way to resolve the issue?

Ghada Khalil (gkhalil)
Changed in starlingx:
importance: Undecided → Low
status: New → Triaged
Revision history for this message
Ramaswamy Subramanian (rsubrama) wrote :

No progress on this bug for more than 2 years. Candidate for closure.

If there is no update, this issue is targeted to be closed as 'Won't Fix' in 2 weeks.

Revision history for this message
Ramaswamy Subramanian (rsubrama) wrote :

Changing the status to 'Won't Fix' as there is no activity.

Changed in starlingx:
status: Triaged → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.