rancher-users #terraform-provider-rancher2

many-terabyte-68707

05/12/2023, 7:23 AM

Hi folks. Has anyone come across this before? I've provided all the necessary parameters in the provider configuration, but the error still shows. I'd appreciate any pointers. Error: Invalid provider configuration │ │ Provider "registry.terraform.io/rancher/rancher2" requires explicit configuration. Add a provider block to the root module and configure the provider's │ required arguments as described in the provider documentation. │ ╵ ╷ │ Error: [ERROR] Normalizing url: no api_url provided │ │ with provider["registry.terraform.io/rancher/rancher2"], │ on <empty> line 0: │ (source code not available)

curved-army-69172

05/24/2023, 7:48 AM

Hi, is it possible to use servergroups (antiaffinity) when deploying a k3s cluster on Openstack? if so how 😉

limited-eye-27484

06/01/2023, 6:37 PM

Hi all, I have a running Rancher cluster that hosts a few EKS clusters. I’m trying to apply some terraform that had previously been working fine to one of our EKS clusters, but now I am getting an error:

Copy code

Error: [ERROR] Creating Project: Cluster ID c-dzn98 is not active

  with module.test-ci.rancher2_project.project,
  on ../../../providers/rancher/modules/project/main.tf line 1, in resource "rancher2_project" "project":
   1: resource "rancher2_project" "project" {

The cluster itself is alive and fine as being reported by Rancher, any ideas as to what I can do in order to fix this Cluster ID not being detected as active from Terraform?

kind-air-74358

06/13/2023, 2:06 PM

Hi, I am a bit confused and hope someone is able to give a bit of explanation on the following. We use the terraform-provider-rancher2 to create a Kubernetes cluster using the rancher2_cluster_v2 resource. This resource contains multiple

rke_config.machine_pools

block containing each a

machine_config

block. Their I define the kind and name of a just created resource

rancher2_machine_config_v2

. As this machine_config_v2 doesn’t contain any specific machine_pool configuration I am re-using this created machine_config_v2 for another

machine_pool

. See the example below.

Copy code

resource "rancher2_machine_config_v2" "main" {

  generate_name   = "gen1-standard-4"
  fleet_namespace = "fleet-default"

  vsphere_config { 
    ... # some variables
  }
}

resource "rancher2_cluster_v2" "main" {
  name                                     = var.name
  fleet_namespace                          = "fleet-default"
  ...

  rke_config {
    machine_global_config = ...

    machine_pools {
      name                         = "control-plane-01"
      cloud_credential_secret_name = ...
      control_plane_role           = true
      etcd_role                    = true
      quantity                     = 3
      machine_config {
        kind = rancher2_machine_config_v2.main.kind
        name = rancher2_machine_config_v2.main.name
      }
    }

    machine_pools {
      name                         = "workers-01"
      cloud_credential_secret_name = ...
      worker_role                  = true
      quantity                     = 6
      machine_config {
        kind = rancher2_machine_config_v2.main.kind
        name = rancher2_machine_config_v2.main.name
      }
    }

    ...
  }

When I’m using the Rancher UI next to increase the number of machines (using the + / - buttons) in the workers-01 pool, Rancher increases actually the number of instances of the pool control-plane-01. Is this due to using the same

machine_config_v2

resource for both pools, or is this actually a bug in the Rancher interface? I know that when creating a cluster using the Rancher Ui, it creates for each machine_pool a separate instance of the machine_config_v2, but I don’t see any use for that when using Terraform to just create a re-usable machine_config.

few-park-83463

06/27/2023, 2:40 PM

Anyone get this error when trying to setup etcd in the rancher2_cluster_v2 module?

few-park-83463

06/27/2023, 2:40 PM

Copy code

╷
│ Error: Too many etcd blocks
│
│   on <http://main.tf|main.tf> line 133, in resource "rancher2_cluster_v2" "cluster_on_vsphere":
│  133:       content {
│
│ No more than 1 "etcd" blocks are allowed
╵

few-park-83463

06/27/2023, 2:40 PM

Copy code

dynamic "etcd" {
      for_each = var.etcd_backup
      content {
        snapshot_schedule_cron = etcd.snapshot_schedule_cron
        snapshot_retention     = etcd.snapshot_retention
        s3_config {
          bucket    = etcd.value.bucket
          endpoint  = etcd.value.endpoint
          cloud_credential_name = data.rancher2_cloud_credential.s3_cred.id
          folder = etcd.value.folder
          skip_ssl_verify = true
        }
      }
    }

brainy-morning-84675

07/05/2023, 3:08 PM

hello! 👋 I am trying to provision a RKE2 cluster on Azure and was wondering if there are any plans for rancher2_machine_config_v2 to support deployment to multiple availability zones so that for eg. server1 would be deployed to availability zone 1 server2 would be deployed to availability zone 2 .. etc. So far I have only managed to get the servers to be deployed as part of an availability set. Thanks!

kind-air-74358

07/10/2023, 8:02 PM

Hello, I have a Rancher cluster bootstrapped using the Rancher2 Terraform Provider. Now I want to add a Kubernetes Resource (GitRepo) to that cluster. Unfortunately the provider doesn’t support fleet resources (yet). What would be the easiest way to create these Kubernetes Resources? One of the options I see is to create a Terraform data rancher2_cluster_v2 object and write the content of

kube_config

to some local file on the system and use that one with the kubernetes provider. But that feels like a bit of too over-engineering.

ancient-car-21816

07/11/2023, 2:26 PM

We deploy our Rancher cluster via Terraform, and at one point had to raise a cluster_monitoring_input memory setting. Since moving to Rancher's updated alerting/monitoring system this setting has been deprecated, but at this point we can't find a way to remove it from tf/cluster state (meaning this change persists after we attempt to apply it). Any suggestions?

millions-ocean-48249

07/21/2023, 2:27 PM

Hi everyone, can someone help me with understanding how to properly reference the vapp_property settings in the rancher2_machine_config_v2 resource? Its expecting a type of list, but I am confused on how to properly reference the key:values in the list block.

Copy code

vapp_property = [
        "guestinfo.interface.0.ip.0.address:ip:vswitch-pg",
        "guestinfo.interface.0.ip.0.netmask:$${netmask:vswitch-pg}",
        "guestinfo.interface.0.route.0.gateway:$${gateway:vswitch-pg}",
        "guestinfo.dns.servers:$${dns:vswitch-pg}"
]

Ive been stuck on this and cant really find any concrete examples. ANY help is greatly appreciated.

limited-eye-27484

07/21/2023, 11:10 PM

Does the Rancher2_cluster terraform provider allow for specifying availability zones for an AWS EKS cluster? Neither of the AWS EKS examples here do https://registry.terraform.io/providers/rancher/rancher2/latest/docs/resources/cluster#creating-eks-cluster-from-ran[…]_v2-for-rancher-v25x-or-above , but the Azure AKS implementation appears to?

millions-ocean-48249

07/24/2023, 7:16 PM

I am seeing some odd behavior with the VM Hardware version when cloning from a template in the rancher2_machine_config_v2 resource... On the template I am cloning from, the VM Hardware version is set to "ESXi 7.0 U2 and later (VM version 19)", but when the nodes are spun up they are set to "ESXi 5.5 and later (VM version 10)". Anyone have any ideas with what could be happening here? My default VM hardware version is set to 7.0 U2 (19) in the vDC settings

faint-shampoo-17603

08/02/2023, 3:24 PM

@faint-shampoo-17603 has left the channel

millions-ocean-48249

08/07/2023, 4:31 PM

In reference to the following github issue: https://github.com/rancher/terraform-provider-rancher2/issues/1090 Does anyone have an idea on how to set the OS in the machine pool config in the rancher2_cluster_v2 config block?

best-address-42882

08/09/2023, 7:16 PM

HI I have an odd issue, when I adds new env var in my terraform config, deployment recreates a pod (sometimes doesn't happen anything) but new values is not shows up, on a same time I can modify my old values that already added to the config and exist in helm chart. Does anyone have any suggestion?

refined-analyst-8898

08/09/2023, 8:10 PM

@refined-analyst-8898 has left the channel

square-policeman-85866

08/15/2023, 8:19 AM

Hi within our config for our machine pool we need to set the quanity to 0. But the terraform run fails out sayin the it’s to be 1. Is there any way around this. We need this to be zero

kind-air-74358

08/16/2023, 8:22 AM

Hi, I am facing an issue when using using a

dynamic

for

machine_pools

in the

rancher2_cluster_v2

resource. We have the following resource defined

Copy code

variable "machine_pools" {
  type = map(object({
    quantity       = number
    cpu_count      = number
    memory_size    = number
}

resource "rancher2_cluster_v2" "main" {
  name                                     = var.name
  fleet_namespace                          = "fleet-default"
  ...

  rke_config {
    # Workers
    dynamic "machine_pools" {
      for_each = var.machine_pools

      content {
        name                         = machine_pools.key
        cloud_credential_secret_name = data.rancher2_cloud_credential.main.id
        worker_role                  = true
        quantity                     = machine_pools.value["quantity"]

        ...
        machine_config {
          kind = rancher2_machine_config_v2.workers[machine_pools.key].kind
          name = rancher2_machine_config_v2.workers[machine_pools.key].name
        }
      }
    }
  }
}

The input looks then something like

Copy code

machine_pools = {
    "cpu2-mem8" = {
      quantity    = 2
      cpu_count   = 2
      memory_size = 8192
    },
    "cpu4-mem8" = {
      quantity    = 2
      cpu_count   = 4
      memory_size = 8192
    }
}

But once we want to add the following machine_pool

Copy code

"cpu2-mem16" = {
  quantity    = 2
  cpu_count   = 2
  memory_size = 16384
}

the cluster resource is updated, but the machine_pools are updated as follows

Copy code

~ machine_pools {
  ~ name                           = "cpu4-mem8" -> "cpu2-mem16"
  # (11 unchanged attributes hidden)

  ~ machine_config {
      - kind = "xxx" -> null
      - name = "nc-np48-w-cpu4-mem8-v25tb" -> null
    }
 }
+ machine_pools {
  + ...
  + name                         = "cpu4-mem8"
  + ...
}

So basically it don’t change the first machine_pool, update the second machine_pool with the configuration of the newly added machine_pool and creates a new machine_pool for the already existing machine_pool but which is being replaced by the new pool. Shouldn’t the name of the machine_pool be leading instead of the lexical order in this case? Is this an issue from our side, or is the provider not dealing with this correctly. Any idea on how to fix this or to work around this?

alert-king-94787

08/31/2023, 4:20 PM

Hi Guys I'm facing an issue when I run

terraform apply

to bootstrap rancher2 I'm launching a infrastructure as code with terraform and following docs like k3s HA + Rancher + Cert-manager. Now I trying to run a rancher2 bootstrap but the execution return error in terraform-provider-rancher2_v3.1.1 plugin:

Copy code

Error: Plugin did not respond
│ 
│   with module.rancher-server.rancher2_bootstrap.admin,
│   on ../../../../../../../infrastructure-modules/modules/aws-rancher-server/rancher.tf line 4, in resource "rancher2_bootstrap" "admin":
│    4: resource "rancher2_bootstrap" "admin" {
│ 
│ The plugin encountered an error, and failed to respond to the plugin.(*GRPCProvider).ApplyResourceChange call. The plugin logs may contain more details.
╵
Releasing state lock. This may take a few moments...

Stack trace from the terraform-provider-rancher2_v3.1.1 plugin:

panic: interface conversion: interface {} is nil, not string

goroutine 70 [running]:
<http://github.com/rancher/terraform-provider-rancher2/rancher2.DoUserLogin({0xc0001260c0|github.com/rancher/terraform-provider-rancher2/rancher2.DoUserLogin({0xc0001260c0>?, 0xc0010df4b8?}, {0x222af22, 0x5}, {0xc0010c606a, 0x5}, {0x222a46e, 0x5}, {0x228a036, 0x21}, ...)
	/go/src/github.com/rancher/terraform-provider-rancher2/rancher2/util.go:154 +0x5f9
<http://github.com/rancher/terraform-provider-rancher2/rancher2.bootstrapDoLogin(0xc000b1f628|github.com/rancher/terraform-provider-rancher2/rancher2.bootstrapDoLogin(0xc000b1f628>?, {0x2221660?, 0xc000474180})
	/go/src/github.com/rancher/terraform-provider-rancher2/rancher2/resource_rancher2_bootstrap.go:275 +0x445
<http://github.com/rancher/terraform-provider-rancher2/rancher2.resourceRancher2BootstrapCreate(0x1edf280|github.com/rancher/terraform-provider-rancher2/rancher2.resourceRancher2BootstrapCreate(0x1edf280>?, {0x2221660?, 0xc000474180})
	/go/src/github.com/rancher/terraform-provider-rancher2/rancher2/resource_rancher2_bootstrap.go:27 +0x70
<http://github.com/hashicorp/terraform-plugin-sdk/helper/schema.(*Resource).Apply(0xc000423040|github.com/hashicorp/terraform-plugin-sdk/helper/schema.(*Resource).Apply(0xc000423040>, 0xc0010ce460, 0xc0010d4260, {0x2221660, 0xc000474180})
	/go/pkg/mod/github.com/hashicorp/terraform-plugin-sdk@v1.17.2/helper/schema/resource.go:320 +0x438
<http://github.com/hashicorp/terraform-plugin-sdk/helper/schema.(*Provider).Apply(0xc00042a000|github.com/hashicorp/terraform-plugin-sdk/helper/schema.(*Provider).Apply(0xc00042a000>, 0xc000b1f8d0, 0x224f874?, 0xf?)
	/go/pkg/mod/github.com/hashicorp/terraform-plugin-sdk@v1.17.2/helper/schema/provider.go:294 +0x70
<http://github.com/hashicorp/terraform-plugin-sdk/internal/helper/plugin.(*GRPCProviderServer).ApplyResourceChange(0xc00011c868|github.com/hashicorp/terraform-plugin-sdk/internal/helper/plugin.(*GRPCProviderServer).ApplyResourceChange(0xc00011c868>, {0xc0010d8070?, 0x4bffa6?}, 0xc0010d8070)
	/go/pkg/mod/github.com/hashicorp/terraform-plugin-sdk@v1.17.2/internal/helper/plugin/grpc_provider.go:895 +0x805
<http://github.com/hashicorp/terraform-plugin-sdk/internal/tfplugin5._Provider_ApplyResourceChange_Handler({0x2156980|github.com/hashicorp/terraform-plugin-sdk/internal/tfplugin5._Provider_ApplyResourceChange_Handler({0x2156980>?, 0xc00011c868}, {0x279c2b0, 0xc0010ca1b0}, 0xc0010d8000, 0x0)
	/go/pkg/mod/github.com/hashicorp/terraform-plugin-sdk@v1.17.2/internal/tfplugin5/tfplugin5.pb.go:3305 +0x170
<http://google.golang.org/grpc.(*Server).processUnaryRPC(0xc0002a01e0|google.golang.org/grpc.(*Server).processUnaryRPC(0xc0002a01e0>, {0x27a5f00, 0xc0007024e0}, 0xc00033b320, 0xc00062ae40, 0x3949e00, 0x0)
	/go/pkg/mod/google.golang.org/grpc@v1.53.0/server.go:1336 +0xd23
<http://google.golang.org/grpc.(*Server).handleStream(0xc0002a01e0|google.golang.org/grpc.(*Server).handleStream(0xc0002a01e0>, {0x27a5f00, 0xc0007024e0}, 0xc00033b320, 0x0)
	/go/pkg/mod/google.golang.org/grpc@v1.53.0/server.go:1704 +0xa2f
<http://google.golang.org/grpc.(*Server).serveStreams.func1.2()|google.golang.org/grpc.(*Server).serveStreams.func1.2()>
	/go/pkg/mod/google.golang.org/grpc@v1.53.0/server.go:965 +0x98
created by <http://google.golang.org/grpc.(*Server).serveStreams.func1|google.golang.org/grpc.(*Server).serveStreams.func1>
	/go/pkg/mod/google.golang.org/grpc@v1.53.0/server.go:963 +0x28a

Error: The terraform-provider-rancher2_v3.1.1 plugin crashed!

This is always indicative of a bug within the plugin. It would be immensely
helpful if you could report the crash with the plugin's maintainers so that it
can be fixed. The output above should help diagnose the issue.

alert-king-94787

08/31/2023, 4:20 PM

Copy code

kubectl --kubeconfig kube_config_ranhcer_server_k3s_temp.yaml get pods --all-namespaces                                                                                                                       ─╯
NAMESPACE             NAME                                       READY   STATUS      RESTARTS      AGE
kube-system           coredns-74448699cf-lr2hz                   1/1     Running     0             3d1h
kube-system           local-path-provisioner-597bc7dccd-4txlt    1/1     Running     0             3d1h
kube-system           helm-install-traefik-crd-sxlcb             0/1     Completed   0             3d1h
kube-system           metrics-server-667586758d-2vttm            1/1     Running     0             3d1h
kube-system           helm-install-traefik-hgg65                 0/1     Completed   2             3d1h
kube-system           svclb-traefik-126decc1-lg8bn               2/2     Running     0             3d1h
kube-system           traefik-7467b667d9-t62mq                   1/1     Running     0             3d1h
cert-manager          cert-manager-cainjector-557c547f54-ng22p   1/1     Running     0             4h5m
cert-manager          cert-manager-5674b9b755-zczgh              1/1     Running     0             4h5m
cert-manager          cert-manager-webhook-86868b95db-hfl2v      1/1     Running     0             4h5m
cattle-system         rancher-99ccc5df-f65hv                     1/1     Running     0             4h3m
cattle-fleet-system   fleet-controller-56786984f4-4tvjc          1/1     Running     0             4h
cattle-fleet-system   gitjob-845b9dcc47-bg8x6                    1/1     Running     0             4h
cattle-system         rancher-webhook-998454b77-8bwlf            1/1     Running     0             3h59m
kube-system           svclb-traefik-126decc1-72xqb               2/2     Running     1 (14m ago)   14m

alert-king-94787

08/31/2023, 4:20 PM

Copy code

terraform version
Terraform v1.5.6
on linux_amd64
+ provider <http://registry.terraform.io/hashicorp/aws|registry.terraform.io/hashicorp/aws> v5.10.0
+ provider <http://registry.terraform.io/hashicorp/helm|registry.terraform.io/hashicorp/helm> v2.11.0
+ provider <http://registry.terraform.io/hashicorp/local|registry.terraform.io/hashicorp/local> v2.4.0
+ provider <http://registry.terraform.io/hashicorp/random|registry.terraform.io/hashicorp/random> v3.5.1
+ provider <http://registry.terraform.io/hashicorp/template|registry.terraform.io/hashicorp/template> v2.2.0
+ provider <http://registry.terraform.io/hashicorp/time|registry.terraform.io/hashicorp/time> v0.9.1
+ provider <http://registry.terraform.io/hashicorp/tls|registry.terraform.io/hashicorp/tls> v4.0.4
+ provider <http://registry.terraform.io/rancher/rancher2|registry.terraform.io/rancher/rancher2> v3.1.1

alert-king-94787

08/31/2023, 4:21 PM

Copy code

terraform {
  required_providers {
    aws = {
        source = "hashicorp/aws"
        version = "~> 5.10.0"
    }
    helm = {
      source  = "hashicorp/helm"
      version = "2.11"
    }
    local = {
      source  = "hashicorp/local"
      version = "2.4.0"
    }
    rancher2 = {
      source  = "rancher/rancher2"
      version = "3.1.1"
    }
  }
  required_version = "~> 1.5.4"
}

alert-king-94787

08/31/2023, 4:23 PM

Copy code

variable "rancher_kubernetes_version" {
  default     = "v1.24.14+k3s1"
}

variable "cert_manager_version" {
  default     = "1.12.3"
}

variable "rancher_version" {
  default     = "2.7.5"
}

variable "k3s_version" {
  default = "v1.24.14+k3s1"
}

alert-king-94787

08/31/2023, 4:25 PM

Does anyone know what's going on?

kind-air-74358

09/04/2023, 1:59 PM

Hi, is there a way to create a

rancher2_role_template

using custom verbs? I.e I want to create a role which implements the verbs

updatepsa

and

manage-namespaces

but I got the error

expected rules.0.verbs.1 to be one of [* create delete deletecollection get list patch update view watch own use bind escalate impersonate], got manage-namespaces

Those verbs are specified by Rancher: https://github.com/rancher/webhook/blob/bd1de5c5620665de8e7158c9bbe83e09780cf9bc/docs.md

future-accountant-19088

09/08/2023, 5:28 PM

So I've got some nice terraform to spin up basically 9 VMs that represent 2 separate HA RKE2 clusters. On one cluster I have rancher installed and I then want to register the other cluster. It looks like rancher2 supports registering other clusters to your rancher but I'm not sure what the process would be. Any thoughts?

glamorous-painting-54907

09/22/2023, 11:08 AM

Does somebody know how I can import an auth_config_activedirectory, basically what is the address?

clean-furniture-7088

09/22/2023, 2:25 PM

Hi, does anyone know the reason why the terraform provider doesn't allow a machine_pool with quantity set to 0 ? https://github.com/rancher/terraform-provider-rancher2/blob/master/rancher2/schema_cluster_v2_rke_config_machine_pool.go#L99 It's possible to create and set empty machine pools in the rancher ui, we want to be able to temporarily scale down the cluster without deleting the machine pool config

few-park-83463

09/22/2023, 5:15 PM

I want to deploy multiple pools across different vSphere clusters using TF. The issue I have is the vDS has been defined per cluster and their is no vDS that is not across multiple clusters. From what I can see the network piece is defined at the end of

worker_node_vsphere_config

instead of being nested in each worker pool. When TF runs it can't interpolate the value if stored in each worker polls mapped block