What causes "Failed to override: %w %s " in hashicorp/terraform?

Re-read the wrapped error text; it usually names the run-state change (e.g. run was discarded).

Failed to override: %w %s

Error message

Failed to override: %w
%s

What it means

In the interactive override flow (backend_common.go:416-419), after the user is prompted to override a soft-failed policy, the confirm() helper returned an error that is not errRunOverridden. This wraps that error with the run URL. It indicates the override confirmation/prompt machinery itself failed (e.g. the run changed state during the prompt, a concurrent modification, or an input error).

Source

Thrown at internal/cloud/backend_common.go:418

				!pc.Actions.IsOverridable || !pc.Permissions.CanOverride {
				return fmt.Errorf("%s soft failed.\n%s", msgPrefix, runURL)
			}

			if op.AutoApprove {
				if _, err = b.client.PolicyChecks.Override(stopCtx, pc.ID); err != nil {
					return b.generalError(fmt.Sprintf("Failed to override policy check.\n%s", runURL), err)
				}
			} else if !b.input {
				return errPolicyOverrideNeedsUIConfirmation
			} else {
				opts := &terraform.InputOpts{
					Id:          "override",
					Query:       "\nDo you want to override the soft failed policy check?",
					Description: "Only 'override' will be accepted to override.",
				}
				err = b.confirm(stopCtx, op, opts, r, "override")
				if err != nil && err != errRunOverridden {
					return fmt.Errorf("Failed to override: %w\n%s\n", err, runURL)
				}

				if err != errRunOverridden {
					if _, err = b.client.PolicyChecks.Override(stopCtx, pc.ID); err != nil {
						return b.generalError(fmt.Sprintf("Failed to override policy check.\n%s", runURL), err)
					}
				} else {
					runURL := fmt.Sprintf(runHeader, b.Hostname, b.Organization, op.Workspace, r.ID)
					b.CLI.Output(fmt.Sprintf("The run needs to be manually overridden or discarded.\n%s\n", runURL))
				}
			}

			if b.CLI != nil {
				b.CLI.Output("------------------------------------------------------------------------")
			}
		default:
			return fmt.Errorf("Unknown or unexpected policy state: %s", pc.Status)
		}

View on GitHub (pinned to c9def3e214)

Solutions

Re-read the wrapped error text; it usually names the run-state change (e.g. run was discarded).
Start a fresh plan/apply; the failed run is typically no longer actionable.
Ensure no other user/process is concurrently mutating the same run during the prompt.
For automation, use -auto-approve to skip the interactive confirm path entirely.

Defensive patterns

Strategy: retry

Validate before calling

// Avoid concurrent mutation of the run during the override prompt.
// Serialize operations on a workspace with a lock.

Try / catch

// Re-fetch the run on confirm failure and retry if still overridable.
if err != nil && err != errRunOverridden {
    r, _ = b.client.Runs.Read(ctx, r.ID)
    if stillOverridable(r) { return retryOverride(ctx, r) }
    return err
}

Prevention

Prevent other users/processes from mutating the run while overriding.
Prefer -auto-approve in automation to skip the interactive confirm path.
Act promptly on override prompts to avoid race windows.

When it happens

Trigger: Reached only on the interactive path (not -auto-approve, b.input true). b.confirm(stopCtx, op, opts, r, "override") returned an err that is non-nil and != errRunOverridden. Distinguished from the API override failure at backend_common.go:422-424 which uses generalError.

Common situations: The run was discarded, errored, or externally modified while the user was answering the override prompt. Input stream error during confirmation. The run transitioned to a non-overridable state mid-prompt.

Related errors

AI-assisted analysis of hashicorp/terraform@c9def3e214 (2026-08-07). Data as JSON: /api/errors/d99cac74c2d692df. Report an issue: GitHub.