Terraform: "Error acquiring the state lock" means wait, then investigate

TL;DR: The most common scary error. It almost never means corruption; it means someone else holds the lock. Read the Lock Info. Never force-unlock because you are impatient.

The error

Error acquiring the state lock

What you are seeing

Error: Error acquiring the state lock

Error message: ConditionalCheckFailedException: ...
Lock Info:
  ID:        abc123
  Path:      myapp-terraform-state/dev/terraform.tfstate
  Operation: OperationTypeApply
  Who:       deploy-bot@ci-runner-4
  Version:   1.9.0
  Created:   2026-09-26 03:10:00 +0000 UTC

State locking prevents two writes from corrupting the state file. This error means another process holds the lock. Reads (plan, refresh-only) generally do not need the lock; writes (apply, state mv) do.

First response

1. Read the Lock Info. Who and Created tell you whether this is a live run or a corpse. A lock created two minutes ago by your CI pipeline is a live run: wait.
2. If it is your own pipeline, let the other run finish. Retrying apply in a loop does not help and can pile up queued runs.
3. Only terraform force-unlock [lock-id] when you have confirmed the holder is dead: crashed CI job, killed terminal, no terraform process running anywhere with that lock ID.

Rules for agents

1. Never force-unlock because you are impatient. Two writers with one state file is how state gets corrupted.
2. terraform plan -lock=false and terraform apply -lock=false disable locking entirely. This is a loaded gun: only for broken-lock emergencies you fully understand, never as a default.
3. Recurring stale locks point at infrastructure: CI jobs that get SIGKILLed, runners that die mid-apply. Fix the runner lifecycle, not the symptom.
4. After any force-unlock, run terraform plan and read it fully before applying. If the killed run half-applied, the plan shows it.

When to use this

• You are seeing this exact error message; match the block above, not just part of it.
• The failing call matches the scenario in the title: Terraform.
• You want the fastest verified fix before digging through logs.

When not to use this

• Your error text differs from the block above; close cousins often have different causes.
• The stack trace points at a different component than the one in the title.
• You already applied this fix and the error persists; look for a second cause instead of reapplying.

State locking docs: locks prevent concurrent writes; the error carries Lock Info (ID, Path, Operation, Who, Version, Created); force-unlock takes the lock ID; -lock=false disables locking. Verified against live docs 2026-09-26.