How to Monitor a URL with Terraform

Set up your first HTTP monitor with Oack's Terraform provider, add a Discord notification channel, and trigger a test alert — all from your terminal in about 10 minutes.

Prerequisites

  • Terraform ≥ 1.5 — install guide. This tutorial uses v1.5.7.
  • Oack account — sign up free and create your first team.
  • Account API key — go to Account Settings → API Keys in the dashboard and create a key. It starts with oack_acc_.
  • Account ID — visible on the Account Settings page.
  • Discord webhook URL — in your Discord server go to Channel Settings → Integrations → Webhooks → New Webhook, copy the URL.

Step 1 — Configure the Provider

Create a new directory for your project and add the Terraform configuration. We'll use environment variables for credentials so they don't end up in version control.

Terminal 
mkdir oack-monitoring && cd oack-monitoring

Export your Oack credentials:

Terminal 
export OACK_API_KEY="oack_acc_your_key_here"
export OACK_ACCOUNT_ID="your-account-uuid-here"

Create the provider configuration:

main.tf 
terraform {
  required_providers {
    oack = {
      source  = "oack-io/oack"
      version = "~> 0.3"
    }
  }
}

provider "oack" {
  # Reads OACK_API_KEY and OACK_ACCOUNT_ID from environment
}

Initialize the provider:

Terminal 
terraform init

You should see Terraform has been successfully initialized! and the Oack provider downloaded.

Step 2 — Create a Team and Monitor

Add a team and an HTTP monitor that checks httpbin.org every 60 seconds.

main.tf (append) 
resource "oack_team" "demo" {
  name = "Terraform Demo"
}

resource "oack_monitor" "httpbin" {
  team_id           = oack_team.demo.id
  name              = "httpbin health"
  url               = "https://httpbin.org/status/200"
  check_interval_ms = 60000
}

That's the minimum config — three required fields (team_id, name, url). Defaults handle everything else: GET method, 10 s timeout, failure threshold of 3, SSL & domain expiry monitoring enabled.

Apply:

Terminal 
terraform apply

Type yes to confirm. Terraform creates the team and monitor.

Step 3 — Verify the Monitor Is Running

Open the Oack dashboard. Your monitor will be at:

https://app-ru.oack.io/teams/<team_id>/monitors/<monitor_id>

You can grab the IDs from the Terraform state:

Terminal 
terraform output -json | jq

# Or read specific values:
terraform state show oack_team.demo
terraform state show oack_monitor.httpbin

Within a minute you'll see the first probe result — a green UP status with latency breakdown for DNS, TCP connect, TLS, TTFB, and total time.

Step 4 — Add a Discord Notification Channel

Create a Discord alert channel and link it to the monitor so you get notified on state changes.

main.tf (append) 
resource "oack_alert_channel" "discord" {
  team_id = oack_team.demo.id
  name    = "Discord Alerts"
  type    = "discord"

  config = {
    webhook_url = var.discord_webhook_url
  }
}

resource "oack_monitor_alert_channel_link" "httpbin_discord" {
  team_id    = oack_team.demo.id
  monitor_id = oack_monitor.httpbin.id
  channel_id = oack_alert_channel.discord.id
}

Add the variable so you can pass the webhook URL without hardcoding it:

variables.tf 
variable "discord_webhook_url" {
  type      = string
  sensitive = true
}

Apply with the webhook URL:

Terminal 
terraform apply -var="discord_webhook_url=https://discord.com/api/webhooks/YOUR_WEBHOOK_HERE"

The Discord channel is now linked to your monitor. When the monitor goes DOWN or recovers, you'll get a notification in your Discord channel.

Step 5 — Trigger a Test Alert (Status Code)

Let's force the monitor to fail by telling it to only accept 201 responses — but httpbin returns 200.

main.tf (update oack_monitor.httpbin) 
resource "oack_monitor" "httpbin" {
  team_id              = oack_team.demo.id
  name                 = "httpbin health"
  url                  = "https://httpbin.org/status/200"
  check_interval_ms    = 60000
  allowed_status_codes = ["201"]   # ← httpbin returns 200, this will fail
  failure_threshold    = 1         # ← fail on the first bad probe
}
Terminal 
terraform apply

Within 60 seconds the monitor will transition to DOWN and you'll see a Discord notification with the monitor URL, expected vs actual status code, and error details.

Step 6 — Trigger a Test Alert (Latency)

Another way to trigger an alert: set an impossibly low latency threshold. Change the monitor to accept any status but fail if response takes more than 1 ms:

main.tf (update oack_monitor.httpbin) 
resource "oack_monitor" "httpbin" {
  team_id              = oack_team.demo.id
  name                 = "httpbin health"
  url                  = "https://httpbin.org/status/200"
  check_interval_ms    = 60000
  latency_threshold_ms = 1         # ← 1 ms is impossible over the internet
  failure_threshold    = 1
}
Terminal 
terraform apply

The monitor will fail because no real HTTP request completes in 1 ms. You'll get another Discord alert showing the actual latency vs the 1 ms threshold.

Step 7 — Restore and Clean Up

Remove the restrictive settings to let the monitor recover:

main.tf (update oack_monitor.httpbin) 
resource "oack_monitor" "httpbin" {
  team_id              = oack_team.demo.id
  name                 = "httpbin health"
  url                  = "https://httpbin.org/status/200"
  check_interval_ms    = 60000
  allowed_status_codes = ["2xx"]
}
Terminal 
terraform apply

The monitor will recover to UP and you'll get a recovery notification in Discord with the downtime duration.

To tear everything down when you're done:

Terminal 
terraform destroy

What's Next