大家可以看到，現在所有 s3.tf 的程式碼，都是直接 hardcode 的，那怎麼透過一些環境變數來動態改變設定呢？
建立 variables.tf
variable "region" {
description = "This is the cloud hosting region where your webapp will be deployed."
default = "ap-northeast-1"
}
variable "bucket" {
description = "default s3 bucket name."
default = "foobar"
}
variable "profile" {
description = "A named profile is a collection of settings and credentials that you can apply to a AWS CLI command."
default = "demo"
}接著打開 s3.tf:
resource "random_pet" "petname" {
length = 3
separator = "-"
}
resource "aws_s3_bucket" "b" {
bucket = "${var.bucket}-${random_pet.petname.id}"
acl = "private"
website {
index_document = "index.html"
}
}可以看到現在可以透過 ${var.bucket} 進行環境設定。
$ terraform apply
An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
+ create
Terraform will perform the following actions:
# aws_s3_bucket.b will be created
+ resource "aws_s3_bucket" "b" {
+ acceleration_status = (known after apply)
+ acl = "private"
+ arn = (known after apply)
+ bucket = (known after apply)
+ bucket_domain_name = (known after apply)
+ bucket_regional_domain_name = (known after apply)
+ force_destroy = false
+ hosted_zone_id = (known after apply)
+ id = (known after apply)
+ region = (known after apply)
+ request_payer = (known after apply)
+ website_domain = (known after apply)
+ website_endpoint = (known after apply)
+ versioning {
+ enabled = (known after apply)
+ mfa_delete = (known after apply)
}
+ website {
+ index_document = "index.html"
}
}
# aws_s3_bucket_object.index will be created
+ resource "aws_s3_bucket_object" "index" {
+ acl = "public-read"
+ bucket = (known after apply)
+ content_type = "text/html; charset=utf-8"
+ etag = (known after apply)
+ force_destroy = false
+ id = (known after apply)
+ key = "index.html"
+ server_side_encryption = (known after apply)
+ source = "content/index.html"
+ storage_class = (known after apply)
+ version_id = (known after apply)
}
# random_pet.petname will be created
+ resource "random_pet" "petname" {
+ id = (known after apply)
+ length = 3
+ separator = "-"
}
Plan: 3 to add, 0 to change, 0 to destroy.我們可以自訂不同的變數內容，如果有 dev 及 prod 環境，可以建立兩個檔案 dev.tfvars 及 prod.tfvars，內容格式如下:
bucket = "example"接著使用 apply 指令時需要加上額外的參數:
terraform apply -var-file=dev.tfvars可以看到底下結果:
Terraform will perform the following actions:
# aws_s3_bucket.b must be replaced
-/+ resource "aws_s3_bucket" "b" {
+ acceleration_status = (known after apply)
~ arn = "arn:aws:s3:::foobar-rightly-healthy-griffon" -> (known after apply)
~ bucket = "foobar-rightly-healthy-griffon" -> "example-rightly-healthy-griffon" # forces replacement
~ bucket_domain_name = "foobar-rightly-healthy-griffon.s3.amazonaws.com" -> (known after apply)
~ bucket_regional_domain_name = "foobar-rightly-healthy-griffon.s3.ap-northeast-1.amazonaws.com" -> (known after apply)
~ hosted_zone_id = "Z2M4EHUR26P7ZW" -> (known after apply)
~ id = "foobar-rightly-healthy-griffon" -> (known after apply)
~ region = "ap-northeast-1" -> (known after apply)
~ request_payer = "BucketOwner" -> (known after apply)
- tags = {} -> null
~ website_domain = "s3-website-ap-northeast-1.amazonaws.com" -> (known after apply)
~ website_endpoint = "foobar-rightly-healthy-griffon.s3-website-ap-northeast-1.amazonaws.com" -> (known after apply)
# (2 unchanged attributes hidden)
~ versioning {
~ enabled = false -> (known after apply)
~ mfa_delete = false -> (known after apply)
}
~ website {
# (1 unchanged attribute hidden)
}
}
# aws_s3_bucket_object.index must be replaced
-/+ resource "aws_s3_bucket_object" "index" {
~ bucket = "foobar-rightly-healthy-griffon" -> (known after apply) # forces replacement
~ etag = "f3a63be8f363c2478ffc79f169610d36" -> (known after apply)
~ id = "index.html" -> (known after apply)
- metadata = {} -> null
+ server_side_encryption = (known after apply)
~ storage_class = "STANDARD" -> (known after apply)
- tags = {} -> null
+ version_id = (known after apply)
# (5 unchanged attributes hidden)
}
Plan: 2 to add, 0 to change, 2 to destroy.
Changes to Outputs:
~ s3_webhost_url = "foobar-rightly-healthy-griffon.s3-website-ap-northeast-1.amazonaws.com" -> (known after apply)如果每個 html 檔案都要寫一個 config，這樣相當麻煩，透過底下方式可以讀取單一目錄直接設定，可以參考這篇設定『Uploading Multiple files in AWS S3 from terraform』
resource "aws_s3_bucket_object" "content" {
bucket = aws_s3_bucket.b.id
acl = "public-read"
content_type = "text/html; charset=utf-8"
for_each = fileset("content/", "*")
key = each.value
source = "content/${each.value}"
# etag makes the file update when it changes; see https://stackoverflow.com/questions/56107258/terraform-upload-file-to-s3-on-every-apply
etag = filemd5("content/${each.value}")
}透過 terraform 的 fileset 讀取全部資料。最後執行部署
$ terraform apply -var-file=dev.tfvars
random_pet.petname: Refreshing state... [id=rightly-healthy-griffon]
aws_s3_bucket_object.index: Refreshing state... [id=index.html]
aws_s3_bucket.b: Refreshing state... [id=example-rightly-healthy-griffon]
An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
+ create
- destroy
Terraform will perform the following actions:
# aws_s3_bucket_object.content["about.html"] will be created
+ resource "aws_s3_bucket_object" "content" {
+ acl = "public-read"
+ bucket = "example-rightly-healthy-griffon"
+ content_type = "text/html; charset=utf-8"
+ etag = "c00772d098b074161c79200aee996d0f"
+ force_destroy = false
+ id = (known after apply)
+ key = "about.html"
+ server_side_encryption = (known after apply)
+ source = "content/about.html"
+ storage_class = (known after apply)
+ version_id = (known after apply)
}
# aws_s3_bucket_object.content["index.html"] will be created
+ resource "aws_s3_bucket_object" "content" {
+ acl = "public-read"
+ bucket = "example-rightly-healthy-griffon"
+ content_type = "text/html; charset=utf-8"
+ etag = "f3a63be8f363c2478ffc79f169610d36"
+ force_destroy = false
+ id = (known after apply)
+ key = "index.html"
+ server_side_encryption = (known after apply)
+ source = "content/index.html"
+ storage_class = (known after apply)
+ version_id = (known after apply)
}
# aws_s3_bucket_object.index will be destroyed
- resource "aws_s3_bucket_object" "index" {
- acl = "public-read" -> null
- bucket = "example-rightly-healthy-griffon" -> null
- content_type = "text/html; charset=utf-8" -> null
- etag = "f3a63be8f363c2478ffc79f169610d36" -> null
- force_destroy = false -> null
- id = "index.html" -> null
- key = "index.html" -> null
- metadata = {} -> null
- source = "content/index.html" -> null
- storage_class = "STANDARD" -> null
- tags = {} -> null
}
Plan: 2 to add, 0 to change, 1 to destroy.這邊要注意 content_type 的設定，如果目錄底下有很多不同格式的檔案，這樣需要分別設定 content type，依照其他方式分類，像是 images/png 或 text/html 區分開來設定。這邊如果是換作 Pulumi 就可以用程式方式來透過副檔名來自動讀取 type，底下是 Go 語言範例:
mime.TypeByExtension(path.Ext(filepath.Join(site, name)))