What causes "could not write lock info for %q: %s" in hashicorp/terraform?

Check write permission on the directory containing the state file: `ls -ld $(dirname )` and `touch /.write-test`.

could not write lock info for %q: %s

Error message

could not write lock info for %q: %s

What it means

Returned by LocalState.writeLockInfo (local_state.go:305) when ioutil.WriteFile fails to create/overwrite the lock info metadata file (.terraform.tfstate.lock.info) with mode 0600. writeLockInfo is the final step of Lock after the OS-level file lock (s.lock) succeeded; it persists the LockInfo record so other processes can identify the lock holder. Failure here means the lock is held but not documented.

Source

Thrown at internal/command/clistate/local_state.go:312

	}

	info := statemgr.LockInfo{}
	err = json.Unmarshal(infoData, &info)
	if err != nil {
		return nil, fmt.Errorf("state file %q locked, but could not unmarshal lock info: %s", s.Path, err)
	}
	return &info, nil
}

// write a new lock info file
func (s *LocalState) writeLockInfo(info *statemgr.LockInfo) error {
	path := s.lockInfoPath()
	info.Path = s.Path
	info.Created = time.Now().UTC()

	err := ioutil.WriteFile(path, info.Marshal(), 0600)
	if err != nil {
		return fmt.Errorf("could not write lock info for %q: %s", s.Path, err)
	}
	return nil
}

View on GitHub (pinned to c9def3e214)

Solutions

Check write permission on the directory containing the state file: `ls -ld $(dirname <statefile>)` and `touch <dir>/.write-test`.
Ensure the state directory exists and is writable by the Terraform process: `mkdir -p .terraform && chmod u+w .terraform`.
Free disk space if the volume is full.
If the directory was deleted mid-run, recreate it and re-run; call `terraform force-unlock` to clean up any half-acquired lock before retrying.

Example fix

# before
$ terraform plan
could not write lock info for "terraform.tfstate": open .terraform.tfstate.lock.info: permission denied

# after
$ chmod u+w .
$ terraform force-unlock <id>   # clear any partial lock
$ terraform plan

Defensive patterns

Strategy: validation

Validate before calling

// Ensure the state directory is writable before locking
stateDir := filepath.Dir(s.Path)
if err := os.MkdirAll(stateDir, 0755); err != nil {
    return fmt.Errorf("state dir unusable: %w", err)
}
probe := filepath.Join(stateDir, ".write-probe")
if err := ioutil.WriteFile(probe, []byte(""), 0600); err != nil {
    return fmt.Errorf("state dir not writable: %w", err)
}
os.Remove(probe)

Try / catch

// If writeLockInfo fails after acquiring the OS lock, attempt to release the OS lock to avoid a held-but-undocumented lock.
if err := s.writeLockInfo(info); err != nil {
    if uerr := s.unlock(); uerr != nil {
        log.Printf("warning: failed to write lock info (%v) and failed to roll back OS lock (%v)", err, uerr)
    }
    s.lockID = ""
    return fmt.Errorf("could not persist lock info: %w", err)
}

Prevention

Pre-flight check write permission on the state directory.
Keep the state directory on local, writable storage.
In CI, create the state directory explicitly before running Terraform.
Document the lock info file path so operators can inspect it.

When it happens

Trigger: ioutil.WriteFile returns an error: the state directory is read-only or missing, permission denied on the directory, disk full, path too long, or the process lost write permission between acquiring the file lock and writing the metadata.

Common situations: The state directory is on a read-only mount; another user owns the directory; the directory was deleted between locking and writing info; disk filled up; running in a sandbox that allows file locking but not file creation.

Related errors

AI-assisted analysis of hashicorp/terraform@c9def3e214 (2026-08-07). Data as JSON: /api/errors/b3b82e05845b4b16. Report an issue: GitHub.