vllm-project/vllm · error · TransferError

{len(errors)}/{len(transfers_to_wait)} transfers failed:

Error message

{len(errors)}/{len(transfers_to_wait)} transfers failed:

What it means

Error "{len(errors)}/{len(transfers_to_wait)} transfers failed: " thrown in vllm-project/vllm.

Source

Thrown at vllm/distributed/kv_transfer/kv_connector/v1/moriio/moriio_engine.py:696

                if status.Failed():
                    errors.append(
                        f"RDMA transfer failed: {status.Message()} "
                        f"(code={status.Code()})"
                    )
                    continue
                if timed_out:
                    errors.append(
                        f"RDMA transfer timed out after {timeout:.0f}s. "
                        f"Check NIC queue depth or reduce concurrency; "
                        f"adjust with kv_connector_extra_config.transfer_timeout."
                    )
                else:
                    still_waiting.append(status)
            remaining = still_waiting
            if remaining:
                time.sleep(0.001)
        if errors:
            raise TransferError(
                f"{len(errors)}/{len(transfers_to_wait)} transfers failed:\n"
                + "\n".join(errors)
            )

    def async_wait_reqid(self):
        assert self.notify_port is not None, "Notify port cannot be None"

        if self.notify_thread is not None:
            return

        def _async_wait():
            host = "*"
            path = make_zmq_path("tcp", host, self.notify_port)
            logger.info("Node starting to listen notify from path = %s", path)

            with zmq_ctx(zmq.ROUTER, path) as sock:
                while True:
                    try:

View on GitHub (pinned to c794754062)

Solutions

1. Inspect the per-transfer errors listed in the message and fix the underlying RDMA/MoRIIO transfer failures.

When it happens

Trigger: Raised at vllm/distributed/kv_transfer/kv_connector/v1/moriio/moriio_engine.py:696 when validation fails: transfers failed. Typically triggered by an incompatible or incomplete vLLM configuration, an unsupported platform/backend combination, or a runtime resource/dependency that is missing.

Common situations: Commonly encountered at vllm/distributed/kv_transfer/kv_connector/v1/moriio/moriio_engine.py:696 during vLLM startup/config validation or runtime setup when: (1) conflicting CLI flags or config fields are combined, (2) the current platform (CUDA/ROCm/CPU/XPU) or installed optional packages do not support the requested feature, or (3) a required value is absent or out of range. Resolve by correcting the configuration as described in the message, or by selecting a supported alternative.

AI-assisted analysis of vllm-project/vllm@c794754062 (2026-08-14). Data as JSON: /api/errors/fd196f2db52c8473. Report an issue: GitHub.