"kube_config_hash":cty.UnknownVal(cty.String)}) does not correlate with any element in actual

[korenlev]

error from 0.1.6 :

\"kubeconfig-user-5zvlgzwq8s:sdwgk7n6bdhd5gxs6jfk4lf7t942kcdssgb9mgtvmm9q5n9s6gmhlz\"\n\n\ncontexts:\n-
name: \"development\"\n  context:\n    user: \"development\"\n    cluster:
\"development\"\n- name: \"development-masters-2\"\n  context:\n    user:
\"development\"\n    cluster: \"development-masters-2\"\n- name:
\"development-masters-1\"\n  context:\n    user: \"development\"\n    cluster:
\"development-masters-1\"\n- name: \"development-masters-3\"\n  context:\n
user: \"development\"\n    cluster:
\"development-masters-3\"\n\ncurrent-context: \"development\"\n"),
"kube_config_hash":cty.UnknownVal(cty.String)}) does not correlate with any
element in actual.

This is a bug in the provider, which should be reported in the provider's own issue tracker.

[xuzhang3]

@korenlev Can you share your TF script and the reproduce steps, thanks

[korenlev]

Create new managed k8s cluster using rancher 2.6.2 :

resource "rancher2_cluster" "rancher_cluster" {
  name = var.rancher_managed_cluster_name
  windows_prefered_cluster = (var.windows == "yes" ? true : false)
  cluster_auth_endpoint {
    enabled = true # Allow communicate with the cluster without main rancher management
  }
  rke_config {
   etc .........

TFS service connection creation with kubeconfig :

resource "azuredevops_serviceendpoint_kubernetes" "service_connection" {
  project_id            = var.Project_ID
  service_endpoint_name = "rancher-azure-${var.rancher_managed_cluster_name}"
  apiserver_url         = "https://${var.rancher_url}"
  authorization_type    = "Kubeconfig"

  kubeconfig {
    kube_config            = rancher2_cluster.rancher_cluster.kube_config
    accept_untrusted_certs = true
  }
}

resource "azuredevops_resource_authorization" "auth" {
  project_id  = var.Project_ID
  resource_id = azuredevops_serviceendpoint_kubernetes.service_connection.id
  authorized  = true
}

error (after dump of the actual kubeconfig text

"kube_config_hash":cty.UnknownVal(cty.String)}) does not correlate with any element in actual.

This is a bug in the provider, which should be reported in the provider's own issue tracker.

##[error]Bash exited with code '1'. Finishing: Terraform_run

[korenlev]

masked kcfg string:

2021-12-15T15:31:37.5830571Z module.managed_cluster.azuredevops_serviceendpoint_kubernetes.service_connection 
2021-12-15T15:31:37.5831779Z to include new values learned so far during apply, provider 
2021-12-15T15:31:37.5835148Z "registry.terraform.io/microsoft/azuredevops" produced an invalid new value 
2021-12-15T15:31:37.5836482Z for .kubeconfig: planned set element 
2021-12-15T15:31:37.5837864Z cty.ObjectVal(map[string]cty.Value{"accept_untrusted_certs":cty.True, 
2021-12-15T15:31:37.5839329Z "cluster_context":cty.StringVal(""), "kube_config":cty.StringVal("apiVersion: 
2021-12-15T15:31:37.5841171Z v1\nkind: Config\nclusters:\n- name: \"prod-tools\"\n cluster:\n server: 
2021-12-15T15:31:37.5842808Z \"https://removed.com/k8s/clusters/c-wrwdh\"\n- name: 
2021-12-15T15:31:37.5844340Z \"prod-tools-masters-2\"\n cluster:\n server: 
2021-12-15T15:31:37.5845954Z \"https://1.1.1.1:6443\"\n certificate-authority-data: 
2021-12-15T15:31:37.5847187Z \"0000000000000000\\\n 
2021-12-15T15:31:37.5869840Z klDQVRFLS0tLS0K\"\n- name: \"prod-tools-masters-3\"\n cluster:\n server: 
2021-12-15T15:31:37.5871865Z \"https://1.1.1.2:6443\"\n certificate-authority-data: 
2021-12-15T15:31:37.5873036Z \"0000000000000000\\\n 
2021-12-15T15:31:37.5918623Z klDQVRFLS0tLS0K\"\n- name: \"prod-tools-masters-1\"\n cluster:\n server: 
2021-12-15T15:31:37.5920515Z \"https://1.1.1.3:6443\"\n certificate-authority-data: 
2021-12-15T15:31:37.5921698Z \"0000000000000000\\\n 
2021-12-15T15:31:37.5956118Z klDQVRFLS0tLS0K\"\n\nusers:\n- name: \"prod-tools\"\n user:\n token: 
2021-12-15T15:31:37.5959002Z \"kubeconfig-user-th8n7ww6j6:removed\"\n\n\ncontexts:\n- 
2021-12-15T15:31:37.5960353Z name: \"prod-tools\"\n context:\n user: \"prod-tools\"\n cluster: 
2021-12-15T15:31:37.5961528Z \"prod-tools\"\n- name: \"prod-tools-masters-2\"\n context:\n user: 
2021-12-15T15:31:37.5962602Z \"prod-tools\"\n cluster: \"prod-tools-masters-2\"\n- name: 
2021-12-15T15:31:37.5963701Z \"prod-tools-masters-3\"\n context:\n user: \"prod-tools\"\n cluster: 
2021-12-15T15:31:37.5964822Z \"prod-tools-masters-3\"\n- name: \"prod-tools-masters-1\"\n context:\n 
2021-12-15T15:31:37.5965754Z user: \"prod-tools\"\n cluster: 
2021-12-15T15:31:37.5966681Z \"prod-tools-masters-1\"\n\ncurrent-context: \"prod-tools\"\n"), 
2021-12-15T15:31:37.5967509Z "kube_config_hash":cty.UnknownVal(cty.String)}) does not correlate with any 
2021-12-15T15:31:37.5968163Z element in actual. 
2021-12-15T15:31:37.5968383Z 
2021-12-15T15:31:37.5969224Z This is a bug in the provider, which should be reported in the provider's own 
2021-12-15T15:31:37.5969873Z issue tracker. 
2021-12-15T15:31:37.5970732Z �[0m�[0m 
2021-12-15T15:31:37.8959238Z ##[error]Bash exited with code '1'. 
2021-12-15T15:31:37.8993686Z ##[section]Finishing: Terraform_run

[xuzhang3]

@korenlev I cannot reproduce your.
kube_config_hash":cty.UnknownVal(cty.String)}) does not correlate with any is not a bug log. From the log, I cannot find any provider crash log. When is the provider crashed, can you paste the errors here? There should be text like nil pointer, invalid address, error ... etc.

[korenlev]

adding again here , it constantly fails to add a rancher 2.6.2 cluster kubeconfig ( just create a cluster and try, was working before that version) : terraform resources:

resource "azuredevops_serviceendpoint_kubernetes" "service_connection" {
  project_id            = var.Project_ID
  service_endpoint_name = var.rancher_managed_cluster_name
  apiserver_url         = "https://${var.rancher_url}"
  authorization_type    = "Kubeconfig"

  kubeconfig {
    kube_config            = rancher2_cluster.rancher_cluster.kube_config
    accept_untrusted_certs = true
  }
}

resource "azuredevops_resource_authorization" "auth" {
  project_id  = var.Project_ID
  resource_id = azuredevops_serviceendpoint_kubernetes.service_connection.id
  authorized  = true
}

================== error log

╷
│ Error: Plugin did not respond
│ 
│   with module.managed_cluster.azuredevops_serviceendpoint_kubernetes.service_connection,
│   on managed/managed.tf line 152, in resource "azuredevops_serviceendpoint_kubernetes" "service_connection":
│  152: resource "azuredevops_serviceendpoint_kubernetes" "service_connection" {
│ 
│ The plugin encountered an error, and failed to respond to the
│ plugin.(*GRPCProvider).ApplyResourceChange call. The plugin logs may
│ contain more details.
╵

Stack trace from the terraform-provider-azuredevops_v0.1.8 plugin:

panic: interface conversion: interface {} is nil, not []interface {}

goroutine 57 [running]:
github.com/microsoft/terraform-provider-azuredevops/azuredevops/internal/service/serviceendpoint.expandServiceEndpointKubernetes(0xc000461e30, 0xc0006aac90, 0xc000195ba0, 0xc000195c70, 0x0)
	github.com/microsoft/terraform-provider-azuredevops/azuredevops/internal/service/serviceendpoint/resource_serviceendpoint_kubernetes.go:211 +0x197f
github.com/microsoft/terraform-provider-azuredevops/azuredevops/internal/service/serviceendpoint.genServiceEndpointCreateFunc.func1(0xc000461e30, 0xfdf180, 0xc00041c200, 0x2, 0x1a0d600)
	github.com/microsoft/terraform-provider-azuredevops/azuredevops/internal/service/serviceendpoint/commons.go:147 +0x65
github.com/hashicorp/terraform-plugin-sdk/helper/schema.(*Resource).Apply(0xc000334960, 0xc0001a8eb0, 0xc000791280, 0xfdf180, 0xc00041c200, 0x106c601, 0xc000195bf8, 0xc0006aac90)
	github.com/hashicorp/[email protected]/helper/schema/resource.go:320 +0x375
github.com/hashicorp/terraform-plugin-sdk/helper/schema.(*Provider).Apply(0xc00004ad80, 0xc0003d9a38, 0xc0001a8eb0, 0xc000791280, 0xc0006a9168, 0xc00065b860, 0x106eca0)
	github.com/hashicorp/[email protected]/helper/schema/provider.go:294 +0x99
github.com/hashicorp/terraform-plugin-sdk/internal/helper/plugin.(*GRPCProviderServer).ApplyResourceChange(0xc00011cc18, 0x13d6270, 0xc0006686f0, 0xc000461650, 0xc00011cc18, 0xc0006686f0, 0xc0003abba0)
	github.com/hashicorp/[email protected]/internal/helper/plugin/grpc_provider.go:895 +0x8a5
github.com/hashicorp/terraform-plugin-sdk/internal/tfplugin5._Provider_ApplyResourceChange_Handler(0x118dc20, 0xc00011cc18, 0x13d6270, 0xc0006686f0, 0xc000115440, 0x0, 0x13d6270, 0xc0006686f0, 0xc00024e800, 0x38c)
	github.com/hashicorp/[email protected]/internal/tfplugin5/tfplugin5.pb.go:3305 +0x214
google.golang.org/grpc.(*Server).processUnaryRPC(0xc0003576c0, 0x13e4018, 0xc000703800, 0xc00062a100, 0xc00036eae0, 0x19cb4a0, 0x0, 0x0, 0x0)
	google.golang.org/[email protected]/server.go:1194 +0x52b
google.golang.org/grpc.(*Server).handleStream(0xc0003576c0, 0x13e4018, 0xc000703800, 0xc00062a100, 0x0)
	google.golang.org/[email protected]/server.go:1517 +0xd0c
google.golang.org/grpc.(*Server).serveStreams.func1.2(0xc000042380, 0xc0003576c0, 0x13e4018, 0xc000703800, 0xc00062a100)
	google.golang.org/[email protected]/server.go:859 +0xab
created by google.golang.org/grpc.(*Server).serveStreams.func1
	google.golang.org/[email protected]/server.go:857 +0x1fd

Error: The terraform-provider-azuredevops_v0.1.8 plugin crashed!

can be fixed. The output above should help diagnose the issue.

##[error]Bash exited with code '1'.
Finishing: Terraform_run

[xuzhang3]

korenlev Can you help check the kube_config, looks like you config miss the contexts block.

Example:

contexts:
- context:
  cluster: development
  namespace: frontend
  user: developer
name: dev-frontend