Terraform “Error acquiring the state lock”: Causes and Safe Fixes

Short answer: Terraform keeps a lock while it changes state. If another run is active, wait. If a run crashed or a CI job was cancelled, the lock stays behind, and you can remove it safely with terraform force-unlock after confirming nothing is running.

The error

terminal

Error: Error acquiring the state lock

Lock Info:
  ID:        3f2a9c1e-...
  Path:      my-tf-state/prod/terraform.tfstate
  Operation: OperationTypeApply
  Who:       runner@ci-12345
  Created:   2026-10-09 14:02:11 UTC

Why it happens

Cause	How to tell
Another plan or apply is running	The Who field shows an active CI job or a colleague
A CI job was cancelled or timed out	The job in Who is finished, but the lock remains
A local run was interrupted	Who shows your machine and an old Created time
Two pipelines target the same state	Locks appear whenever both run at the same time

Fix it safely

1. Read Who and Created in the error. Confirm that process is really gone (CI job finished, no terminal still running).
2. Remove the lock using the ID from the error:

unlock.sh

terraform force-unlock 3f2a9c1e-...

Prevent it

Serialize runs per state in CI and let Terraform wait briefly for a lock instead of failing immediately.

.github/workflows/terraform.yml

# GitHub Actions: one run at a time per environment
concurrency:
  group: terraform-prod
  cancel-in-progress: false

# and in the job
- run: terraform apply -lock-timeout=5m -auto-approve

With the S3 backend, Terraform 1.10 and later can lock using S3 itself (use_lockfile = true), without a DynamoDB table. Check the backend documentation for your version before migrating.

backend.tf

terraform {
  backend "s3" {
    bucket       = "my-tf-state"
    key          = "prod/terraform.tfstate"
    region       = "us-east-1"
    use_lockfile = true
  }
}