ISC | Trying to delete an empty VA cluster throws an error

Feedback Bugs
,
1

What problem are you observing?

Trying to delete an empty VA cluster in ISC throws an error. There are basically 3 issues here:

  • error message is broken and shows its own error
  • moving sources from one cluster to another, shows them in connections for both the old and the new cluster
  • trying to delete the empty cluster fails even though its marked as empty

What is the correct behavior?

Successful deletion

What product feature is this related to?

ISC - Virtual Appliances

What are the steps to reproduce the issue?

Not sure but for us it was to create a standard VA cluster, add 2 VAs and then sources. In our case it was 5 sources. Then move sources to another cluster. Delete VAs. Try to delete cluster. The cluster was configured only for Cloud Connector Gateway. Turning all components off didnt help.

Do you have any other information about your environment that may help?

image
image1774Ă—190 6.86 KB

image
image606Ă—186 4.3 KB

Error message isnt very helpful. Done through UI on MS Edge Version 143.0.3650.96 (Official build) (64-bit). In the console I can see a failed DELETE request.

Trying to delete it through API results in:

{
  "detailCode": "400.2.1.0 Object in use by another",
  "trackingId": "20d945dfc931447192e456b950469765",
  "messages": [
    {
      "locale": "en-US",
      "localeOrigin": "DEFAULT",
      "text": "The request was syntactically correct but its content is semantically invalid."
    },
    {
      "locale": "en-GB",
      "localeOrigin": "REQUEST",
      "text": "The request was syntactically correct but its content is semantically invalid."
    }
  ],
  "causes": [
    {
      "locale": "en-US",
      "localeOrigin": "DEFAULT",
      "text": "This cluster cannot be deleted as it is connected to the services"
    }
  ]
}
2

Hi @lukas_ceremeta

Usually means the cluster still has a backend dependency, even if the UI looks empty. In ISC, the supported delete flow is:

  1. Prove the old cluster has zero connected sources
    Go to Admin → Connections → Virtual Appliances → (old cluster) → Details → Connections and verify nothing is listed. If anything is listed, edit each source and switch its Virtual Appliance Cluster to the new cluster and Save (that’s the supported way to detach).

  2. Make sure there are no “orphaned” managed clients still tied to the cluster
    Even after deleting VAs in the UI, a managed-client record can remain and keep the cluster “connected to services.” Validate via API:

  • GET /v3/managed-clients (find any client still associated to the old cluster)

  • DELETE /managed-clients/{id} for each leftover client

  1. Delete the cluster
    After (1) and (2), delete via Actions → Delete Cluster (UI) per docs or via API DELETE /v3/managed-clusters/{id}.
3

Hi,

as I stated in the first post. The connection section contains the old sources even though I moved them one by one to another cluster. So now I have 2 clusters with these sources listed. When you go to the sources themselves, they show the new cluster.

I just tried using /managed-clients but no clients are associated with the old cluster.

4

The error went away after waiting for 1-2 days.

2 Likes
5

Hi Lukas,

Thanks for the update! I’m glad to hear the error has been resolved.

If you encounter the issue again or run into any further issues, please create a SailPoint support ticket with our team here, and they’ll be happy to assist you further.

1 Like