Restarting worker with directory cache gives "Destination directory already exists" / "Failed to hardlink newly cached directory"

[rejuvenile]

I see:

^[^[  2025-12-12T19:58:54.780199Z  WARN nativelink_worker::running_actions_manager: Directory cache failed, falling back to traditional download, digest: DigestInfo("2c6786e19c74725b23307422bc0d9bfab5ea04b130e56d31e726df03776ea082-166"), e: Error { code: InvalidArgument, messages: ["Destination directory already exists: /Users/user/Work/nativelink/work/df72a46e-f96f-4641-b17c-d8f8fa2bb0b5/work", "Failed to hardlink newly cached directory"] }

Most likely this is due to workers crashing with state still on disk.

[rejuvenile]

Actually, I nuked all state on disk and restarted, and yet my workers still say:

  2025-12-12T20:03:09.771676Z  WARN nativelink_worker::running_actions_manager: Directory cache failed, falling back to traditional download, digest: DigestInfo("40dc269c829e56d7969895bfa2ffa755cea90225aea4056ec9e9d30466144974-166"), e: Error { code: InvalidArgument, messages: ["Destination directory already exists: /Users/user/Work/nativelink/work/3b8f4423-5b9b-4f88-b83a-fa2e4473c855/work", "Failed to hardlink newly cached directory"] }
    at nativelink-worker/src/running_actions_manager.rs:295
    in nativelink_worker::local_worker::worker_start_action
    in nativelink_worker::local_worker::worker_start_action_ctx with operation_id: "3b8f4423-5b9b-4f88-b83a-fa2e4473c855", digest_function: BLAKE3
    in nativelink_worker::local_worker::run with shutdown_rx: broadcast::Receiver
    in nativelink::worker with name: "worker_0"

[MarcusSorealheis]

Directory cache needs to be marked as experimental and unstable. This already exists error is one of many issues others have highlighted.