Use Rancher API v3 to update Cluster Workspace

SUSE Rancher Prime
1

I’m new with Rancher and I’m trying to automate the importing of an existing k3s (v1.25.10+k3s1) single-node cluster in my Rancher instance on Azure AKS (k8s v1.30.4, Rancher v2.9.1). What i’m trying to do is to move thel cluster (once registered) from the fleet-default workspace, to my custom “dev” workspace.

To do so, i’m using a bash script the following API call:

  1. Get LoginToken:

      LOGINTOKEN=$(curl -s \
  "https://${SERVER_URL}/v3-public/localProviders/local?action=login" \
  -H 'content-type: application/json' \
  --data-binary "${AUTH_DATA}" | jq -r '.token')

  2. Get API Token:

     APITOKEN=$(curl -s \
   "https://${SERVER_URL}/v3/token" \
   -H 'content-type: application/json' \
   -H "Authorization: Bearer $LOGINTOKEN" \
   --data-binary "${JSON_DATA}" | jq -r '.token')

  3. Cluster Definition: Here i try to set the payload of the API call in order to put the cluster in the “dev” workspace (and cluster-group, already present).

    JSON_DATA=$(jq -n --arg orderId "$ORDER_ID" \
    '{
        "type": "cluster",
        "name": $orderId,
        "import": true,
        "metadata": {
            "labels": {
                "environment": "dev",
                "fleet.cattle.io/workspace": "dev",
                "fleet.cattle.io/cluster-group": "dev"
            },
            "fleetWorkspaceName": "dev"
        },
        "annotations": {
            "fleet.cattle.io/workspace": "dev",
            "fleet.cattle.io/cluster-group": "dev"
        },
        "spec": {
            "fleetWorkspaceName": "dev"
        }
    }')
 
CLUSTERID=$(curl -s "https://${SERVER_URL}/v3/cluster" \
                  -H 'content-type: application/json' \
                  -H "Authorization: Bearer $APITOKEN" \
                  --data-binary "${JSON_DATA}" | jq -r '.id') ```

  4. Cluster Registration:

     JSON_DATA=$(jq -n --arg clusterId ${CLUSTERID} \
    '{
        "type": "clusterRegistrationToken",
        "clusterId": $clusterId
    }'

  ID=$(curl -s "https://${SERVER_URL}/v3/clusters/${CLUSTERID}/clusterregistrationtoken" \ 
  -H 'content-type: application/json' \
  -H "Authorization: Bearer $APITOKEN" \
  --data-binary "${JSON_DATA}" | jq -r '.id')

So, at this point, the cluster is registered on Rancher but it is inside the fleet-default workspace, even if i tried to set fleetWorkspace=“dev”.
What i thought is to try to update the cluster with a PUT API call, once the cluster in “Active” state.
To do so:

1 - Get the Cluster config:

  current_config=$(curl -s \
     "https://${SERVER_URL}/v3/clusters/${CLUSTERID}" \
     -H  "Authorization: Bearer ${APITOKEN}")`

2 - Update Cluster Config:
local PUT_JSON=$(echo "$current_config" | jq --arg ws "dev" '. | .fleetWorkspaceName = $ws' )

3 - POST API Call to push new cluster config:

curl -s \
    -X PUT "https://${SERVER_URL}/v3/clusters/${CLUSTERID}" \
    -H "Authorization: Bearer ${APITOKEN}" \
    -H "Content-Type: application/json" \
    --data-binary "${PUT_JSON}"

Once all those commands are executed, the Cluster still remain in fleet-default workspace. Now, if i retrieve once again the cluster config, inside the “conditions” field, i can notice the following element:

{
            "lastUpdateTime": "2024-10-29T10:03:41Z",
            "status": "True",
            "type": "Updated"
 }

I don’t know if this is about my last PUT api call, but still it didn’t work.

I have also enabled the “provisioningv2-fleet-workspace-back-population” Feature Flag from the Rancher Global Settings in the Rancher UI

At this point, i don’t know if Rancher API supports this kind of operation or if i’m missing something?