使用Terraform创建OKE集群
首先
我想使用Terraform来创建OKE（容器引擎 for Kubernetes）的集群。
前提条件
VCN和子网已经按照此示例创建完毕。
使用Terraform在预先创建的虚拟云网络（VCN）和子网中进行OKE集群的配置。
创建一个栈
因为创建TF文件是很困难的，所以我们可以通过OCI控制台设置群集，并在最后点击保存为“堆栈”而不是“创建群集”。
（我们将使用自定义创建，而不是使用快速创建来一起创建网络资源（例如VCN），仅创建群集）。
从创建的堆栈中下载Terraform配置。
下载的Terraform配置是Zip格式的，解压后即为main.tf文件。
编辑和创建配置文件
主.tf文件
为了使main.tf文件具有通用性，已经下载的main.tf文件中直接写入了各个变量。因此，我们需要将这些变量通过另外一个文件进行管理。
我已经根据以下方式进行了编辑。
-
- 1行目
provider は別ファイルで定義するので、コメントアウトします。
3-26行目
クラスタを作成する部分
各変数はvar.変数名に書き換え
28-62行目
ノードプールを作成する部分
各変数はvar.変数名に書き換え
29行目
3-26行目で作成したクラスタのcluster_idを引き継いでいます。
3行目で宣言してる変数？からcluster_idを引き継いでいるらしい（調べ切れてない）
1 # provider "oci" {}
2
3 resource "oci_containerengine_cluster" "generated_oci_containerengine_cluster" {
4 cluster_pod_network_options {
5 cni_type = var.cluster_pod_network_options
6 }
7 compartment_id = var.compartment_id
8 endpoint_config {
9 is_public_ip_enabled = var.is_public_ip_enabled
10 subnet_id = var.endpoint_config_subnet_id
11 }
12 kubernetes_version = var.kubernetes_version
13 name = var.cluster_name
14 options {
15 kubernetes_network_config {
16 services_cidr = var.kubernetes_network_config_cider
17 }
18 persistent_volume_config {
19 }
20 service_lb_config {
21 }
22 service_lb_subnet_ids = [var.service_lb_subnet_ids]
23 }
24 type = var.cluster_type
25 vcn_id = var.vcn_id
26 }
27
28 resource "oci_containerengine_node_pool" "create_node_pool_details0" {
29 cluster_id = "${oci_containerengine_cluster.generated_oci_containerengine_cluster.id}"
30 compartment_id = var.compartment_id
31 initial_node_labels {
32 key = "name"
33 value = var.nodepool_name
34 }
35 kubernetes_version = var.kubernetes_version
36 name = var.nodepool_name
37 node_config_details {
38 node_pool_pod_network_option_details {
39 cni_type = var.cluster_pod_network_options
40 max_pods_per_node = var.max_pods_per_node
41 pod_subnet_ids = [var.pod_subnet_ids]
42 }
43 placement_configs {
44 availability_domain = var.availability_domain
45 subnet_id = var.node_subnet_id
46 }
47 size = var.node_count
48 }
49 node_eviction_node_pool_settings {
50 eviction_grace_duration = var.eviction_grace_duration
51 is_force_delete_after_grace_duration = var.is_force_delete_after_grace_duration
52 }
53 node_shape = var.node_shape
54 node_shape_config {
55 memory_in_gbs = var.memory_in_gbs
56 ocpus = var.ocpus
57 }
58 node_source_details {
59 image_id = var.image_id
60 source_type = var.source_type
61 }
62 }
提供者.tf文件 .tf
与认证相关的文件。与以往一样。
1 provider "oci" {
2 tenancy_ocid = var.tenancy_ocid
3 user_ocid = var.user_ocid
4 private_key_path = var.private_key_path
5 fingerprint = var.fingerprint
6 region = var.region
7 }
8
9 terraform {
10 required_providers {
11 oci = {
12 source = "oracle/oci"
13 }
14 }
15 }
variables.tf文件
变量的定义。
对于由不同类别的默认值不会变化太多的值，我们设置了默认值。
1 ## provider
2 variable "tenancy_ocid" {
3 type = string
4 }
5 variable "user_ocid" {
6 type = string
7 }
8 variable "private_key_path" {
9 type = string
10 }
11 variable "fingerprint" {
12 type = string
13 }
14 variable "region" {
15 type = string
16 default = "ca-toronto-1"
17 }
18
19 ## OKE Cluster
20 variable "cluster_pod_network_options" {
21 type = string
22 default = "OCI_VCN_IP_NATIVE"
23 }
24 variable "compartment_id" {
25 type = string
26 }
27 variable "is_public_ip_enabled" {
28 type = bool
29 default = true
30 }
31 variable "endpoint_config_subnet_id" {
32 type = string
33 }
34 variable "kubernetes_version" {
35 type = string
36 default = "v1.27.2"
37 }
38 variable "cluster_name" {
39 type = string
40 }
41 variable "kubernetes_network_config_cider" {
42 type = string
43 default = "10.96.0.0/16"
44 }
45 variable "service_lb_subnet_ids" {
46 type = string
47 }
48 variable "cluster_type" {
49 type = string
50 default = "BASIC_CLUSTER"
51 }
52 variable "vcn_id" {
53 type = string
54 }
55
56 ## Node Pool
57 variable "nodepool_name" {
58 type = string
59 }
60 variable "max_pods_per_node" {
61 type = number
62 default ="31"
63 }
64 variable "pod_subnet_ids" {
65 type = string
66 }
67 variable "availability_domain" {
68 type = string
69 default = "TGjA:CA-TORONTO-1-AD-1"
70 }
71 variable "node_subnet_id" {
72 type = string
73 }
74 variable "node_count" {
75 type = number
76 }
77 variable "eviction_grace_duration" {
78 type = string
79 default = "PT60M"
80 }
81 variable "is_force_delete_after_grace_duration" {
82 type = bool
83 default = false
84 }
85 variable "node_shape" {
86 type = string
87 default = "VM.Standard.E4.Flex"
88 }
89 variable "memory_in_gbs" {
90 type = number
91 default = 16
92 }
93 variable "ocpus" {
94 type = number
95 default = 1
96 }
97 variable "image_id" {
98 type = string
99 default = "ocid1.image.oc1.ca-toronto-1.aaaaaaaaihukvlwwkkf4oklbg7mwxhxxe3nng5llyvckzk3lz6r2f4rlrf5a"
100 }
101 variable "source_type" {
102 type = string
103 default = "IMAGE"
104 }
terraform.tfvars文件
设置各个变量的值。实际上，变量的内容已经写在里面了。
1 ## provider
2 tenancy_ocid = ""
3 user_ocid = ""
4 private_key_path = ""
5 fingerprint = ""
6 # region = "" # default = "ca-toronto-1"
7
8 ## OKE Cluster
9 # cluster_pod_network_options = "" # default = "OCI_VCN_IP_NATIVE"
10 compartment_id = ""
11 # is_public_ip_enabled = "" # default = true
12 endpoint_config_subnet_id = ""
13 # kubernetes_version = "" # default = "v1.27.2"
14 cluster_name = ""
15 # kubernetes_network_config_cider = "" # default = "10.96.0.0/16"
16 service_lb_subnet_ids = ""
17 # cluster_type = "" # default = "BASIC_CLUSTER"
18 vcn_id = ""
19
20 ## Node Pool
21 nodepool_name = ""
22 # max_pods_per_node = "" # default ="31"
23 pod_subnet_ids = ""
24 # availability_domain = "" # "TGjA:CA-TORONTO-1-AD-1"
25 node_subnet_id = ""
26 node_count = ""
27 # eviction_grace_duration = "" # default = "PT60M"
28 # is_force_delete_after_grace_duration = "" # default = false
29 # node_shape = "" # default = "VM.Standard.E4.Flex"
30 # memory_in_gbs = "" # default = 16
31 # ocpus = "" # default = 1
32 # image_id = "" # default = "ocid1.image.oc1.ca-toronto-1.aaaaaaaaihukvlwwkkf4oklbg7mwxhxxe3nng5llyvckzk3lz6r2f4rlrf5a" (OL8.8)
33 # source_type = "" # default = "IMAGE"
我准备了以下4个文件。
$ ls -l
total 48
-rw-rw-r--. 1 opc opc 1713 Nov 8 04:23 main.tf
-rw-rw-r--. 1 opc opc 265 Nov 8 03:17 provider.tf
-rw-rw-r--. 1 opc opc 1101 Nov 8 05:51 terraform.tfvars
-rw-rw-r--. 1 opc opc 1797 Nov 8 05:21 variables.tf
运行Terraform
开始
$ terraform init
Initializing the backend...
Initializing provider plugins...
- Reusing previous version of oracle/oci from the dependency lock file
- Using previously-installed oracle/oci v5.19.0
Terraform has been successfully initialized!
You may now begin working with Terraform. Try running "terraform plan" to see
any changes that are required for your infrastructure. All Terraform commands
should now work.
If you ever set or change modules or backend configuration for Terraform,
rerun this command to reinitialize your working directory. If you forget, other
commands will detect it and remind you to do so if necessary.
计划
可以看出会创建两个资源，一个是集群，另一个是节点池。
$ terraform plan
Terraform used the selected providers to generate the following execution plan. Resource actions are indicated with the following symbols:
+ create
Terraform will perform the following actions:
・・・
Plan: 2 to add, 0 to change, 0 to destroy.
───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Note: You didn't use the -out option to save this plan, so Terraform can't guarantee to take exactly these actions if you run "terraform apply" now.
申请
$ terraform apply
Terraform used the selected providers to generate the following execution plan. Resource actions are indicated with the following symbols:
+ create
・・・
Plan: 2 to add, 0 to change, 0 to destroy.
Do you want to perform these actions?
Terraform will perform the actions described above.
Only 'yes' will be accepted to approve.
Enter a value: yes
・・・
oci_containerengine_cluster.generated_oci_containerengine_cluster: Creating...
oci_containerengine_cluster.generated_oci_containerengine_cluster: Still creating... [10s elapsed]
・・・
oci_containerengine_cluster.generated_oci_containerengine_cluster: Creation complete after 8m20s [id=ocid1.cluster.oc1.ca-toronto-1.aaaaaaaakoxuj73vvvv24rh2gpuhgqzei45xbkbq2oj5v5hb3c3t6j6pexka]
oci_containerengine_node_pool.create_node_pool_details0: Creating...
oci_containerengine_node_pool.create_node_pool_details0: Still creating... [10s elapsed]
・・・
oci_containerengine_node_pool.create_node_pool_details0: Creation complete after 3m9s [id=ocid1.nodepool.oc1.ca-toronto-1.aaaaaaaacje3bvvklnwnof4d3rlcs4vpvzjqjosdqlmytldzxn5fzpuihqmq]
Apply complete! Resources: 2 added, 0 changed, 0 destroyed.
尝试改变几次变量后，我们确认了OKE集群可以进行配置。不过，我们也注意到这种方法稍微缺乏通用性（例如无法设置虚拟节点）。
Terraform依然是一个很深奥的领域。
