Resources, Dependencies, and Meta-Arguments¶

Overview¶

Declare managed resources, understand the dependency graph, and use count, for_each, lifecycle, and depends_on correctly.

A resource is something Terraform creates and updates through a provider. Addresses look like local_file.readme. Terraform builds a graph from references between resources; meta-arguments change how many instances exist and how replace/destroy behaves.

This is a core tutorial in Module 6 · Resources of the REBASH Academy Terraform for Cloud & DevOps Engineers series — written for Cloud, DevOps, Platform, and SRE engineers.

Prerequisites¶

Providers and the Terraform Plugin Model

Learning Objectives¶

By the end of this tutorial, you will be able to:

Write a resource block and read its address in state
Explain implicit vs explicit (depends_on) dependencies
Choose for_each over count for named instances
Apply a lifecycle rule intentionally

Architecture¶

This topic’s control points and relationships are shown below.

Theory¶

What it is¶

resource "local_file" "readme" {
  filename = "${path.module}/README.txt"
  content  = "managed by Terraform\n"
}

Terraform tracks this as local_file.readme in state. When another resource references local_file.readme.filename, Terraform gains an implicit dependency and creates the file first. Use depends_on only when there is a real ordering need that references cannot express (for example, an API side effect with no attribute link).

Meta-arguments (available on most resources):

Meta-argument	Purpose
`count`	Create N copies; instances are `resource.name[0]`
`for_each`	Create one instance per map/set key; `resource.name["key"]`
`depends_on`	Explicit edges in the graph
`lifecycle`	`create_before_destroy`, `prevent_destroy`, `ignore_changes`, …
`provider`	Select a non-default provider alias

Why it matters¶

Wrong dependencies cause apply-time races (subnet before VPC, DNS before load balancer). Prefer implicit links via attribute references — they document intent. Prefer for_each with stable keys (environment names, availability zones) so removing one item does not reshuffle indexes the way count often does. Lifecycle rules protect production databases from accidental destroy — and can also hide drift if you overuse ignore_changes.

How it works¶

Terraform evaluates expressions and builds a directed graph of resources.
Walk order respects dependencies; independent resources may create in parallel.
count / for_each expand one block into multiple instances in state.
On update, providers may update in place or force replace (destroy + create) depending on the attribute.
lifecycle tweaks replace/destroy behaviour for that resource.

for_each = toset(["dev", "stage"]) expands to local_file.env["dev"] and local_file.env["stage"] — stable keys beat count indexes when membership changes.

Key concepts and comparisons¶

Pattern	Prefer when
Implicit reference	Attribute of A is needed by B
`depends_on`	Ordering without a referenceable attribute
`count`	Simple N identical copies; indexes OK
`for_each`	Named, stable instance keys
`create_before_destroy`	Zero-downtime-style replace
`prevent_destroy`	Block accidental destroy plans
`ignore_changes`	Skip listed attributes (use sparingly)

Common pitfalls¶

Using count with a list that reorders — instance indexes shift and force replacements.
Sprinkling depends_on everywhere instead of wiring attributes.
Setting ignore_changes = all to “fix” drift — you stop managing the resource.
Assuming prevent_destroy blocks terraform destroy of the whole root without care — it blocks plans that would destroy that resource; design env teardown deliberately.

Hands-on Lab¶

Create a workspace for this tutorial.

mkdir -p ~/rebash-terraform/module-06 && cd ~/rebash-terraform/module-06

Focus: Explore implicit dependencies, count, and depends_on

Step 1 – Create chained resources¶

cat > main.tf <<'EOF'
terraform {
  required_providers {
    local = { source = "hashicorp/local", version = "~> 2.5" }
    null  = { source = "hashicorp/null", version = "~> 3.2" }
  }
}
resource "local_file" "base" {
  filename = "${path.module}/base.txt"
  content  = "base
"
}
resource "local_file" "child" {
  count    = 2
  filename = "${path.module}/child-${count.index}.txt"
  content  = "depends on ${local_file.base.filename}
"
}
resource "null_resource" "after" {
  depends_on = [local_file.child]
  triggers   = { stamp = timestamp() }
}
EOF
terraform init

Step 2 – Apply and inspect graph order¶

terraform apply -auto-approve
terraform state list
ls -1 child-*.txt base.txt
terraform graph | head -n 30

Final step – Cleanup note¶

terraform destroy -auto-approve
# Workspace kept for notes; remove with: rm -rf "$(pwd)" when finished

Validation¶

Lab commands run under ~/rebash-terraform/module-06/
You can explain each Theory section in your own words
You used modern tooling where it applies to this topic
You can describe one production failure mode for this topic

Code Walkthrough¶

Production practice for Resources, Dependencies, and Meta-Arguments always combines:

Inspect before you change (status, plan, logs, dry-run)
Prefer reversible, documented changes (Git, IaC, drop-ins, version pins)
Capture evidence (command output, pipeline logs) for handovers
Prefer current tools and APIs over legacy shortcuts
Least privilege — escalate credentials only when required

Keep runbooks short enough to follow under pressure. Automate checks; keep humans for judgement.

Security Considerations¶

Treat credentials and tokens for terraform as privileged — never commit them
Prefer short-lived auth (OIDC, roles, SSO) over long-lived keys
Validate blast radius before apply/deploy/delete operations
Restrict who can approve production changes
Collect audit logs; limit who can read sensitive traces

Common Mistakes¶

Using count with a list that reorders — instance indexes shift and force replacements.

Validate assumptions against the Theory section and official docs before changing production.

Sprinkling depends_on everywhere instead of wiring attributes.

Lab shortcuts (open security groups, admin roles, skip approvals) must not ship unchanged.

Changing production without a rollback path

Always know how to revert (previous artefact, prior release, state rollback, DNS failback).

Best Practices¶

Encode Resources, Dependencies, and Meta-Arguments changes as code and review them in pull requests
Pin versions (images, modules, actions, provider plugins)
Separate environments with clear promotion gates
Alert on symptoms with runbooks attached
Destroy lab resources; tag everything with owner and expiry where possible

Troubleshooting¶

Symptom	Likely cause	Fix
Auth / permission denied	Wrong identity, policy, or scope	Check caller identity, roles, and least-privilege policies
Timeout / no route	Network, DNS, security group, or endpoint	Trace path, DNS, and allow-lists before retrying
Drift / unexpected plan	Manual change or wrong state/workspace	Reconcile desired vs actual; avoid click-ops on managed resources
Pipeline/job red	Flaky step, cache, or missing secret	Read failing step logs; bisect recent workflow/config changes
Cost spike	Idle load balancer, NAT, oversized compute	Inventory billable resources; stop/delete labs promptly

Summary¶

Resources, Dependencies, and Meta-Arguments is essential for Cloud and DevOps engineers working with terraform. Practise the lab until the inspection and change path is muscle memory, then continue the track.

Interview Questions¶

How does Terraform infer dependencies between resources?
When do you need explicit depends_on?
What is the difference between count and for_each?
What operational risk does count index shifting introduce?
What does lifecycle { prevent_destroy = true } protect against?

Sample answer — question 2

Implicit dependencies come from references. depends_on is for hidden ordering (for example, API readiness) that references cannot express.

Sample answer — question 4

Removing an element from a count list can force replacement of later indexes. for_each with stable keys usually produces safer updates.