为什么 ‘terraform import’ 没有读取我的配置文件？

huangapple
go评论106阅读模式
英文:

Why 'terraform import' did not read my configuration?

问题

我正在使用Go构建一个Terraform插件/提供程序。我在我的模式中有以下内容（链接）：

  1. Importer: &schema.ResourceImporter{
  2.   State: resourceKubernetesClusterNodePoolImport,
  3. }

这是函数的代码（链接）：

  1. func resourceKubernetesClusterNodePoolImport(d *schema.ResourceData, m interface{}) ([]*schema.ResourceData, error) {
  2. 	apiClient := m.(*civogo.Client)
  4. 	clusterID, nodePoolID, err := utils.ResourceCommonParseID(d.Id())
  5. 	if err != nil {
  6. 		return nil, err
  7. 	}
  9. 	log.Printf("[INFO] retriving the node pool %s", nodePoolID)
  10. 	resp, err := apiClient.GetKubernetesCluster(clusterID)
  11. 	if err != nil {
  12. 		if resp != nil {
  13. 			return nil, err
  14. 		}
  15. 	}
  17. 	d.Set("cluster_id", resp.ID)
  18. 	for _, v := range resp.Pools {
  19. 		if v.ID == nodePoolID {
  20. 			d.Set("num_target_nodes", v.Count)
  21. 			d.Set("target_nodes_size", v.Size)
  22. 		}
  23. 	}
  25. 	return []*schema.ResourceData{d}, nil
  26. }

当我在上面的函数顶部添加了这些调试行时，结果（reg和ok）为空且为false：

  1. reg, ok := d.GetOk("region")
  2. log.Printf("resourceKubernetesClusterNodePoolImport reg %+v\n", reg)
  3. log.Println("resourceKubernetesClusterNodePoolImport ok", ok)
  4. log.Println("resourceKubernetesClusterNodePoolImport d.Get()", d.Get("region"))
  5. if !ok {
  6. 	return nil, fmt.Errorf("[ERR] Please provide a region in your import configuration")
  7. }

跟踪信息如下：

  1. $ echo $TF_LOG
  2. INFO
  3. $ tf import civo_kubernetes_node_pool.my_pool 1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad
  4. 2021-08-12T11:22:30.552+0800 [INFO]  Terraform version: 1.0.3
  5. 2021-08-12T11:22:30.552+0800 [INFO]  Go runtime version: go1.16.4
  6. 2021-08-12T11:22:30.552+0800 [INFO]  CLI args: []string{"/usr/local/bin/terraform", "import", "civo_kubernetes_node_pool.my_pool", "1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad"}
  7. 2021-08-12T11:22:30.552+0800 [INFO]  Loading CLI configuration from /Users/zulh/.terraformrc
  8. 2021-08-12T11:22:30.553+0800 [INFO]  CLI command args: []string{"import", "civo_kubernetes_node_pool.my_pool", "1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad"}
  9. 2021-08-12T11:22:30.629+0800 [INFO]  Failed to read plugin lock file .terraform/plugins/darwin_amd64/lock.json: open .terraform/plugins/darwin_amd64/lock.json: no such file or directory
  10. 2021-08-12T11:22:30.631+0800 [INFO]  provider: configuring client automatic mTLS
  11. 2021-08-12T11:22:31.284+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: configuring server automatic mTLS: timestamp=2021-08-12T11:22:31.283+0800
  12. 2021-08-12T11:22:31.358+0800 [INFO]  provider: configuring client automatic mTLS
  13. 2021-08-12T11:22:31.404+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: configuring server automatic mTLS: timestamp=2021-08-12T11:22:31.404+0800
  14. 2021-08-12T11:22:31.478+0800 [WARN]  ValidateProviderConfig from "provider["registry.terraform.io/civo/civo"]" changed the config value, but that value is unused
  15. civo_kubernetes_node_pool.my_pool: Importing from ID "1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad"...
  16. 2021-08-12T11:22:31.479+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: 2021/08/12 11:22:31 resourceKubernetesClusterNodePoolImport reg: timestamp=2021-08-12T11:22:31.479+0800      <-- 这一行
  17. 2021-08-12T11:22:31.479+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: 2021/08/12 11:22:31 resourceKubernetesClusterNodePoolImport ok false: timestamp=2021-08-12T11:22:31.479+0800 <-- 这一行
  18. 2021-08-12T11:22:31.479+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: 2021/08/12 11:22:31 resourceKubernetesClusterNodePoolImport d.Get(): timestamp=2021-08-12T11:22:31.479+0800  <-- 这一行
  19. │ Error: [ERR] Please provide a region in your import configuration
  20. │
  21. │

问题是，在运行上述tf import命令之前，我已经在我的Terraform配置文件main.tf中添加了region（并保存了它）。这是我的main.tf文件：

  1. resource "civo_kubernetes_cluster" "my_cluster" {
  2.     name = "my-cluster"
  3. }
  5. resource "civo_kubernetes_node_pool" "my_pool" {
  6.     region = "LON1" <-- 这一行
  7. }

为什么我的Terraform插件/提供程序将区域视为空？

据我了解，Terraform在运行导入逻辑之前会先读取配置文件。我错了吗？我是否遗漏了什么？

为了完整起见，我正在使用Terraform v1.0.3，这是我的provider.tf文件...

  1. terraform {
  2.   required_providers {
  3.     civo = {
  4.       source  = "civo/civo"
  5.       version = "99.0.0"
  6.     }
  7.   }
  8. }
  10. provider "civo" {
  11.     token = "my-api-key-here"
  12. }

...其中99.0.0是我仅用于测试的版本。我发布这个问题时，此提供程序的最新/生产版本是0.10.9。

英文:

I'm building a Terraform plugin/provider using Go. I have the following in my schema (link):

  1. Importer: &amp;schema.ResourceImporter{
  2.   State: resourceKubernetesClusterNodePoolImport,
  3. }

And this is the function (link):

  1. func resourceKubernetesClusterNodePoolImport(d *schema.ResourceData, m interface{}) ([]*schema.ResourceData, error) {
  2. 	apiClient := m.(*civogo.Client)
  4. 	clusterID, nodePoolID, err := utils.ResourceCommonParseID(d.Id())
  5. 	if err != nil {
  6. 		return nil, err
  7. 	}
  9. 	log.Printf(&quot;[INFO] retriving the node pool %s&quot;, nodePoolID)
  10. 	resp, err := apiClient.GetKubernetesCluster(clusterID)
  11. 	if err != nil {
  12. 		if resp != nil {
  13. 			return nil, err
  14. 		}
  15. 	}
  17. 	d.Set(&quot;cluster_id&quot;, resp.ID)
  18. 	for _, v := range resp.Pools {
  19. 		if v.ID == nodePoolID {
  20. 			d.Set(&quot;num_target_nodes&quot;, v.Count)
  21. 			d.Set(&quot;target_nodes_size&quot;, v.Size)
  22. 		}
  23. 	}
  25. 	return []*schema.ResourceData{d}, nil
  26. }

When I added these debug lines at the top of function above, the results (reg and ok) were empty and false:

  1. reg, ok := d.GetOk(&quot;region&quot;)
  2. log.Printf(&quot;resourceKubernetesClusterNodePoolImport reg %+v\n&quot;, reg)
  3. log.Println(&quot;resourceKubernetesClusterNodePoolImport ok&quot;, ok)
  4. log.Println(&quot;resourceKubernetesClusterNodePoolImport d.Get()&quot;, d.Get(&quot;region&quot;))
  5. if !ok {
  6. 	return nil, fmt.Errorf(&quot;[ERR] Please provide a region in your import configuration&quot;)
  7. }

Traces:

  1. $ echo $TF_LOG
  2. INFO
  3. $ tf import civo_kubernetes_node_pool.my_pool 1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad
  4. 2021-08-12T11:22:30.552+0800 [INFO]  Terraform version: 1.0.3
  5. 2021-08-12T11:22:30.552+0800 [INFO]  Go runtime version: go1.16.4
  6. 2021-08-12T11:22:30.552+0800 [INFO]  CLI args: []string{&quot;/usr/local/bin/terraform&quot;, &quot;import&quot;, &quot;civo_kubernetes_node_pool.my_pool&quot;, &quot;1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad&quot;}
  7. 2021-08-12T11:22:30.552+0800 [INFO]  Loading CLI configuration from /Users/zulh/.terraformrc
  8. 2021-08-12T11:22:30.553+0800 [INFO]  CLI command args: []string{&quot;import&quot;, &quot;civo_kubernetes_node_pool.my_pool&quot;, &quot;1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad&quot;}
  9. 2021-08-12T11:22:30.629+0800 [INFO]  Failed to read plugin lock file .terraform/plugins/darwin_amd64/lock.json: open .terraform/plugins/darwin_amd64/lock.json: no such file or directory
  10. 2021-08-12T11:22:30.631+0800 [INFO]  provider: configuring client automatic mTLS
  11. 2021-08-12T11:22:31.284+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: configuring server automatic mTLS: timestamp=2021-08-12T11:22:31.283+0800
  12. 2021-08-12T11:22:31.358+0800 [INFO]  provider: configuring client automatic mTLS
  13. 2021-08-12T11:22:31.404+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: configuring server automatic mTLS: timestamp=2021-08-12T11:22:31.404+0800
  14. 2021-08-12T11:22:31.478+0800 [WARN]  ValidateProviderConfig from &quot;provider[\&quot;registry.terraform.io/civo/civo\&quot;]&quot; changed the config value, but that value is unused
  15. civo_kubernetes_node_pool.my_pool: Importing from ID &quot;1b272dc1-836c-4ea7-8f20-4569f20ddbdd:d0c22622-54b3-41c2-9eae-24591737d5ad&quot;...
  16. 2021-08-12T11:22:31.479+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: 2021/08/12 11:22:31 resourceKubernetesClusterNodePoolImport reg: timestamp=2021-08-12T11:22:31.479+0800      &lt;-- this line
  17. 2021-08-12T11:22:31.479+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: 2021/08/12 11:22:31 resourceKubernetesClusterNodePoolImport ok false: timestamp=2021-08-12T11:22:31.479+0800 &lt;-- this line
  18. 2021-08-12T11:22:31.479+0800 [INFO]  provider.terraform-provider-civo_v99.0.0: 2021/08/12 11:22:31 resourceKubernetesClusterNodePoolImport d.Get(): timestamp=2021-08-12T11:22:31.479+0800  &lt;-- this line
  19. ╷
  20. │ Error: [ERR] Please provide a region in your import configuration
  21. │
  22. │
  23. ╵

The problem is, I have added the region in my Terraform configuration main.tf file (and saved it) before I run the tf import command above. Here is my main.tf file:

  1. resource &quot;civo_kubernetes_cluster&quot; &quot;my_cluster&quot; {
  2.     name = &quot;my-cluster&quot;
  3. }
  5. resource &quot;civo_kubernetes_node_pool&quot; &quot;my_pool&quot; {
  6.     region = &quot;LON1&quot; &lt;-- this line
  7. }

Why my Terraform plugin/provider sees the region as empty?

From my understanding, Terraform will first read the configuration file before running the import logic. Am I wrong? Did I miss something here?

For the sake of completeness, I'm using Terraform v1.0.3 and this is my provider.tf file...

  1. terraform {
  2.   required_providers {
  3.     civo = {
  4.       source  = &quot;civo/civo&quot;
  5.       version = &quot;99.0.0&quot;
  6.     }
  7.   }
  8. }
  10. provider &quot;civo&quot; {
  11.     token = “my-api-key-here”
  12. }

...where 99.0.0 is the version that I built for my testing only. The latest/prod version for this provider is 0.10.9 when I posted this question.

答案1

得分: 1

Importer.State函数无法访问现有的配置，它会提供一个空的资源状态和id - 请参考代码。

作为解决方法，您可以通过下划线（或任何不与id或区域名称中的符号冲突的符号）将region传递给导入命令，例如：

  1. terraform import civo_kubernetes_node_pool.my_pool <poolid>_<region>

然后在导入函数中处理这个参数：

  1. func resourceKubernetesClusterNodePoolImport(d *schema.ResourceData, m interface{}) ([]*schema.ResourceData, error) {
  2.     apiClient := m.(*civogo.Client)
  4.     // 从提供的字符串中提取id和region
  5.     parts := strings.Split(d.Id(), "_")
  6. 	if len(parts) != 2 {
  7. 		return nil, errors.New("要导入一个池，请使用格式{pool_id}_{region}")
  8. 	}
  10.     poolId := parts[0]
  11. 	region := parts[1]
  13.     // 为资源设置新的ID
  14.     d.SetId(poolId)
  16.     // 设置region
  17.     d.Set("region", region)
  19.     // 完成导入的其余部分
  21.     ...
  22. }
英文:

Importer.State function does not have access to existing config, it is provided an empty resource state with id - see code.

As a workaround you can allow passing region to import command through underscore (or any symbol that does not conflict with symbols in ids or region names), like so:

  1. terraform import civo_kubernetes_node_pool.my_pool &lt;poolid&gt;_&lt;region&gt;

And handle this in you importer function:

  1. func resourceKubernetesClusterNodePoolImport(d *schema.ResourceData, m interface{}) ([]*schema.ResourceData, error) {
  2.     apiClient := m.(*civogo.Client)
  4.     // extract id and region from provided string
  5.     parts := strings.Split(d.Id(), &quot;_&quot;)
  6. 	if len(parts) != 2 {
  7. 		return nil, errors.New(&quot;To import a pool, use the format {pool_id}_{region}&quot;)
  8. 	}
  10.     poolId := parts[0]
  11. 	region := parts[1]
  13.     // Set new ID for resource
  14.     d.SetId(poolId)
  16.     // Set region
  17.     d.Set(&quot;region&quot;, region)
  19.     // Do the rest of the import
  21.     ...
  22. }

huangapple
  • 本文由 发表于 2021年8月12日 12:17:37
  • 转载请务必保留本文链接：https://go.coder-hub.com/68751548.html
  • go
  • hashicorp
  • terraform
  • terraform-provider-kubernetes
  • terraform0.12+
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定