cf-terraforming is a command line utility to facilitate terraforming your existing Cloudflare resources. It does this by using your account credentials to retrieve your configurations from the Cloudflare API and converting them to Terraform configurations that can be used with the Terraform Cloudflare provider.
This tool is ideal if you already have Cloudflare resources defined but want to start managing them via Terraform, and don't want to spend the time to manually write the Terraform configuration to describe them.
Usage:
cf-terraforming [command]
Available Commands:
access_application Import Access Application data into Terraform
access_policy Import Access Policy data into Terraform
access_rule Import Access Rule data into Terraform
account_member Import Account Member data into Terraform
all Import all Cloudflare resources into Terraform
custom_pages Import Custom Pages data into Terraform
filter Import Filter data into Terraform
firewall_rule Import Firewall Rule data into Terraform
help Help about any command
load_balancer Import a load balancer into Terraform
load_balancer_monitor Import a load balancer monitor into Terraform
load_balancer_pool Import a load balancer pool into Terraform
page_rule Import Page Rule data into Terraform
rate_limit Import Rate Limit data into Terraform
record Import Record data into Terraform
spectrum_application Import a spectrum application into Terraform
version Print the version number of cf-terraforming
waf_rule Import WAF Rule data into Terraform
worker_route Import a worker route into Terraform
worker_script Import a worker script into Terraform
zone Import zone data into Terraform
zone_lockdown Import Zone Lockdown data into Terraform
zone_settings_override Import Zone Settings Override data into Terraform
Flags:
-a, --account string Use specific account ID for import
-c, --config string config file (default is $HOME/.cf-terraforming.yaml)
-e, --email string API Email address associated with your account
-h, --help help for cf-terraforming
-k, --key string API Key generated on the 'My Profile' page. See: https://dash.cloudflare.com/?account=profile
-l, --loglevel string Specify logging level: (trace, debug, info, warn, error, fatal, panic)
-o, --organization string Use specific organization ID for import
-s, --tfstate Export tfstate for the given resource instead of HCL Terraform config (default ! See caveats below !)
-v, --verbose Specify verbose output (same as setting log level to debug)
-z, --zone string Limit the export to a single zone (name or ID)
Use "cf-terraforming [command] --help" for more information about a command.
A note on storing your credentials securely: We recommend that you store your Cloudflare credentials (API key, email, account ID, etc) as environment variables as demonstrated below.
You can use go run to build and execute the binary in a single command like so:
go run cmd/cf-terraforming/main.go --email $CLOUDFLARE_EMAIL --key $CLOUDFLARE_TOKEN --account $CLOUDFLARE_ACCOUNT_ID spectrum_application
will contact the Cloudflare API on your behalf and result in a valid Terraform configuration representing the resource you requested:
resource "cloudflare_spectrum_application" "1150bed3f45247b99f7db9696fffa17cbx9" {
protocol = "tcp/8000"
dns = {
type = "CNAME"
name = "example.com"
}
ip_firewall = "true"
tls = "off"
origin_direct = [ "tcp://37.241.37.138:8000", ]
}
See the currently supported resources below.
Use the all command to download everything and convert it into Terraform config.
go run cmd/cf-terraforming/main.go --email $CLOUDFLARE_EMAIL --key $CLOUDFLARE_TOKEN --account $CLOUDFLARE_ACCOUNT_ID all
By default, cf-terraforming will not output any log type messages to stdout when run, so as to not pollute your generated Terraform config files and to allow you to cleanly redirect cf-terraforming output to existing Terraform configs.
However, it can be useful when debugging issues to specify a logging level, like so:
go run cmd/cf-terraforming/main.go --email $CLOUDFLARE_EMAIL --key $CLOUDFLARE_TOKEN -a 1233455678d876bc764b5f763af7644411 -l="debug" spectrum_application
DEBU[0000] Initializing cloudflare-go API email=apicloudflare.com Account ID=e9e138b6x52ea331b359a2ddfc6a8 Organization ID= Zone name=example.com
DEBU[0000] Selecting zones for import
DEBU[0000] Zones selected:
DEBU[0000] Zone ID=81b06ss3228f488fh84e5e993c2dc17 Name=example.com
DEBU[0000] Importing zone settings data
For convenience, you can set the verbose flag, which is functionally equivalent to setting a log level of debug:
go run cmd/cf-terraforming/main.go --email $CLOUDFLARE_EMAIL --key $CLOUDFLARE_TOKEN -a 1233455678d876bc764b5f763af7644411 -v spectrum_application
- A Cloudflare account with resources defined (e.g. a few zones, some load balancers, spectrum applications, etc)
- A valid Cloudflare API key and sufficient permissions to access the resources you are requesting via the API
- A working installation of Go
$ go get -u github.com/cloudflare/cf-terraforming/...This will fetch the cf-terraforming tool as well as its dependencies, updating them as necessary.
Work is underway to support downloading your resources via cf-terraform as valid tfstate, which will allow you to skip importing and let Terraform become aware of and begin managing your resources immediately.
Currently, only the worker_route command supports the --tfstate flag, but support for downloading tfstate will steadily be added across the other resources supported by cf-terraforming.
To use this currently experimental feature, pass the --tfstate (-s) flag to your command like so:
$ go run cmd/cf-terraforming/main.go --email $CLOUDFLARE_EMAIL --key $CLOUDFLARE_API_KEY -z example.com -a $CLOUDFLARE_ACCOUNT_ID --organization $CLOUDFLARE_ORG_ID --tfstate worker_route
The output will look something like this:
{
"version": 1,
"terraform_version": "",
"serial": 0,
"lineage": "",
"modules": [
{
"path": ["root"],
"depends_on": [],
"outputs":
{},
"resources":
{
"cloudflare_worker_route.3489743985643594723e238ce":
{
"primary":
{
"id": "249653298b6041148b411b4723e238ce",
"attributes":
{
"enabled": "true",
"id": "249653298b6041148b411b4723e238ce",
"multi_script": "true",
"pattern": "example.com",
"zone": "example.com",
"zone_id": "z1b06143shshs3223e5ec83c2z1klop"
},
"meta":
{},
"tainted": false
},
"depends_on": [],
"deposed": [],
"provider": "provider.cloudflare",
"type": "cloudflare_worker_route"
},
"cloudflare_worker_route.zZ3exdfyt3e76274ef25dda436":
{
"primary":
{
"id": "9b3e8f2cc874zsde9274ef25dda58f36",
"attributes":
{
"enabled": "true",
"id": "9b3e8f2cc87444fb9274ef25dda58f36",
"multi_script": "true",
"pattern": "example.com/*",
"zone": "example.com",
"zone_id": "z1b06143shshs3223e5ec83c2z1klop"
},
"meta":
{},
"tainted": false
},
"depends_on": [],
"deposed": [],
"provider": "provider.cloudflare",
"type": "cloudflare_worker_route"
}
}
}]
}
This means that when you next run $ terraform plan, Terraform should see no difference between your defined resources and your current tfstate.
The following resources can be downloaded into Terraform HCL format right now. Some resources also support the --tfstate flag. Support across the remaining commands will be added over time.
| Resource | Generating HCL | Generating tfstate |
|---|---|---|
| access_application | ✔️ | ✔️ |
| access_policy | ✔️ | ✔️ |
| access_rule | ✔️ | ✔️ |
| account_member | ✔️ | ✖️ |
| custom_pages | ✔️ | ✖️ |
| filter | ✔️ | ✔️ |
| firewall_rule | ✔️ | ✔️ |
| load_balancer | ✔️ | ✖️ |
| load_balancer_pool | ✔️ | ✖️ |
| load_balancer_monitor | ✔️ | ✖️ |
| page_rule | ✔️ | ✖️ |
| rate_limit | ✔️ | ✔️ |
| record | ✔️ | ✔️ |
| spectrum_application | ✔️ | ✖️ |
| waf_rule | ✔️ | ✖️ |
| worker_route | ✔️ | ✔️ |
| worker_script | ✔️ | ✖️ |
| zone | ✔️ | ✔️ |
| zone_lockdown | ✔️ | ✔️ |
| zone_settings_override | ✔️ | ✖️ |