search

nutanix / terraform-provider-nutanix

Terraform Nutanix Provider
https://www.terraform.io/docs/providers/nutanix/
Mozilla Public License 2.0
95 stars 111 forks source link

Data resource "nutanix_cluster" fails when given "name" argument #542

Open kingsleyck opened 1 year ago

kingsleyck commented 1 year ago

Nutanix Cluster Information

abhimutant commented 1 year ago

Hi, it seems you have more than one cluster with same name. You can fetch the cluster details with cluster_uuid as well . 

data "nutanix_cluster" "cluster" {
     cluster_id = "{{ id }} "
}
kingsleyck commented 1 year ago

In this case Prism Central does not have two clusters with the same name. The issue is not limited to one cluster's name, either. Any cluster name tested produces the same result. The behavior of this resource is different between pc.2022.1.0.2 and pc.2022.6.0.1. It works as expected on pc.2022.1.0.2 (name argument returns a single cluster).

kingsleyck commented 1 year ago

To collect another data point, I used the nutanix_clusters data source with a for loop:

variable "cluster_name" {
  default = CLUSTERNAME
}

data "nutanix_clusters" "clusters" {}

output "clusters" {
  value = [
    for entity in data.nutanix_clusters.clusters.entities : entity.name if entity.name == var.cluster_name
  ]
}

Which returns the following:

Changes to Outputs:
  + clusters = [
      + "CLUSTERNAME",
      + "CLUSTERNAME",
    ]

There are two cluster entities returned whose properties are all identical (uuid, name, etc). This result is unexpected given neither the GUI nor REST API v3/clusters/list display or return 2 cluster entities with the same name.

abhimutant commented 1 year ago

Can you please collect the Trace log and share it ? Have to check the response from /clusters/list API.
export TF_LOG=TRACE Also, you can confirm whether the API returns with 2 cluster entities. Trace log will help how API is returning the response vs how we are handling it.

kingsleyck commented 1 year ago 
POST /api/nutanix/v3/clusters/list HTTP/1.1
Host: prism.fqdn:9440
User-Agent: nutanix/v3
Content-Length: 44
Accept: application/json
Authorization: Basic <removed>
Content-Type: application/json
Accept-Encoding: gzip

{
 "filter": "",
 "kind": "cluster",
 "length": 100
}

-----------------------------------------------------: timestamp=2023-02-03T11:47:19.615-0700
2023-02-03T11:47:21.539-0700 [DEBUG] provider.terraform-provider-nutanix_v1.7.1.exe: {"@level":"info","@message":"2023/02/03 11:47:21 [DEBUG] Nutanix API 
Response Details:\n---[ RESPONSE ]--------------------------------------\nHTTP/1.1 200 OK\r\nTransfer-Encoding: chunked\r\nContent-Security-Policy: 
default-src 'self' https://*.nutanix.com; script-src 'self' 'unsafe-inline' 'unsafe-eval'; style-src 'self' 'unsafe-inline'; connect-src 'self' wss: 
https://downloads.frame.nutanix.com https://downloads.frame.nutanix.us; img-src 'self' blob: data:; frame-src 'self' https://*.nutanix.com blob: 
data:\r\nContent-Type: application/json\r\nDate: Fri, 03 Feb 2023 18:47:20 GMT\r\nServer: envoy\r\nSet-Cookie: <ommitted>; Expires=Fri, 03-Feb-2023 19:02:21 GMT; Secure; HttpOnly; 
Path=/; SameSite=Lax\r\nStrict-Transport-Security: max-age=8640000:includeSubdomains\r\nVary: Accept-Encoding\r\nX-Content-Type-Options: 
nosniff\r\nX-Dns-Prefetch-Control: off\r\nX-Envoy-Upstream-Service-Time: 1414\r\nX-Frame-Options: SAMEORIGIN\r\nX-Ntnx-Env: pc\r\nX-Ntnx-Product: 
pc.2022.6.0.1\r\nX-Permitted-Cross-Domain-Policies: master-only\r\nX-Xss-Protection: 1; mode=block\r\n\r\n8b2d6\r\n{\n \"api_version\": \"3.1\",\n 
\"metadata\": {\n  \"total_matches\": 122,\n  \"kind\": \"cluster\"\n },\n \"entities\": [\n  {\n   \"status\": {\n    \"state\": \"COMPLETE\",\n    
\"name\": \"CLUSTERNAME\",\n    \"resources\": {\n     \"nodes\": {\n      \"hypervisor_server_list\": [\n       {\n        \"ip\": \"<removed>\",\n (removed remainder)
...
<repeats same API call and output three times>
...
2023-02-03T11:47:24.427-0700 [INFO]  provider.terraform-provider-nutanix_v1.7.1.exe: 2023/02/03 11:47:24 [Debug] total=122, remaining=-78, offset=200 
len(entities)=244: timestamp=2023-02-03T11:47:24.427-0700
data.nutanix_clusters.clusters: Still reading... [10s elapsed]
data.nutanix_clusters.clusters: Read complete after 11s [id=terraform-20230203184726004500000001]
2023-02-03T11:47:30.612-0700 [DEBUG] provider.stdio: received EOF, stopping recv loop: err="rpc error: code = Unavailable desc = error reading from 
server: EOF"
2023-02-03T11:47:30.623-0700 [DEBUG] provider: plugin process exited: 
path=.terraform/providers/registry.terraform.io/nutanix/nutanix/1.7.1/windows_amd64/terraform-provider-nutanix_v1.7.1.exe pid=7268
2023-02-03T11:47:30.623-0700 [DEBUG] provider: plugin exited
2023-02-03T11:47:30.684-0700 [DEBUG] building apply graph to check for errors
2023-02-03T11:47:30.684-0700 [DEBUG] ProviderTransformer: "data.nutanix_clusters.clusters (expand)" (*terraform.nodeExpandApplyableResource) needs 
provider["registry.terraform.io/nutanix/nutanix"]
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "data.nutanix_clusters.clusters (expand)" references: []
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "var.nutanix_username" references: []
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "var.nutanix_password" references: []
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "var.nutanix_endpoint" references: []
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "var.nutanix_port" references: []
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "var.cluster_name" references: []
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "output.clusters (expand)" references: [data.nutanix_clusters.clusters (expand) 
var.cluster_name]
2023-02-03T11:47:30.684-0700 [DEBUG] ReferenceTransformer: "provider[\"registry.terraform.io/nutanix/nutanix\"]" references: [var.nutanix_port 
var.nutanix_username var.nutanix_endpoint var.nutanix_password]
2023-02-03T11:47:30.684-0700 [DEBUG] pruneUnusedNodes: data.nutanix_clusters.clusters (expand) is no longer needed, removing
2023-02-03T11:47:30.684-0700 [DEBUG] pruneUnusedNodes: provider["registry.terraform.io/nutanix/nutanix"] is no longer needed, removing
2023-02-03T11:47:30.684-0700 [INFO]  backend/local: plan operation completed

Changes to Outputs:
  + clusters = [
      + "CLUSTERNAME",
      + "CLUSTERNAME",
    ]

This Prism Central has 122 clusters attached. I sanitized and removed most of the API return for legibility. Based on the API call above to clusters/list I would expect to see an offset added to a subsequent call, with only two API calls required to retrieve all 122 entities. Instead it is sending 3 separate calls with length 100 and no offset specified. It's not clear to me why it is making 3 API calls when only 2 would be required.