unslothai/unsloth · error · RuntimeError

sd-server connection lost during img_gen submit

Error message

sd-server connection lost during img_gen submit

What it means

The HTTP POST that submits an img_gen job failed at the transport layer (connection reset/refused, or httpx timeout) and the server is no longer alive, so the wrapper classifies it as the server having died during submit rather than transient network trouble. The full _died_message form appends the underlying transport exception and log tail. At submit time the connection was just verified alive, so a transport error here strongly implies a crash.

Source

Thrown at studio/backend/core/inference/sd_cpp_server.py:464

        Raises ``RuntimeError`` on submit/poll failures (including the server dying), with
        the log tail attached.
        """
        # Already stopped with the cancel event set: report cancellation (route 409), not a generic "server died" 500.
        if self._stopped or not self.is_alive():
            if cancel_event is not None and cancel_event.is_set():
                raise SdCppCancelled("sd-server generation was cancelled.")
            raise RuntimeError("sd-server is not running.")

        self._step_listener = on_step
        job_id: Optional[str] = None
        try:
            # Submit -> 202 Accepted + job id.
            try:
                resp = self._client.post(
                    f"{self.base_url}{_IMG_GEN_PATH}", json = payload, timeout = submit_timeout
                )
            except (*_TRANSPORT_ERRORS, httpx.TimeoutException) as exc:
                raise RuntimeError(self._died_message("img_gen submit", exc)) from exc
            if resp.status_code == 429:
                raise RuntimeError("sd-server job queue is full (HTTP 429).")
            if resp.status_code not in (200, 202):
                raise RuntimeError(
                    f"sd-server img_gen submit -> {resp.status_code}: {resp.text[:500]}"
                )
            try:
                job = resp.json()
            except ValueError as exc:
                raise RuntimeError(
                    f"sd-server img_gen returned a non-JSON submit response: {exc}"
                ) from exc
            if not isinstance(job, dict):
                raise RuntimeError(
                    f"sd-server img_gen returned an unexpected submit response type: {type(job)}"
                )
            job_id = job.get("id")
            if not job_id:

View on GitHub (pinned to 203007d190)

Solutions

1. Inspect the attached log tail/exception for the crash reason before retrying.
2. Restart the server, then resubmit the job once.
3. If it recurs on the same input, minimize the trigger (drop LoRAs, lower resolution) and capture the server stderr — it is a server-side crash, not client misuse.

try:
    blobs = server.img_gen(payload, ...)
except RuntimeError as e:
    if "connection lost during img_gen submit" in str(e):
        capture_server_tail(e)
        server = restart_server()
        blobs = server.img_gen(payload, ...)
    raise

When it happens

Trigger: Server process dies (OOM, segfault loading a LoRA, watchdog kill) between the alive-check and the POST, or during a slow submit; connection reset by peer surfaces as this error.

Common situations: Large payloads (many LoRAs, high-res params) making submit slow while the server crashes; memory pressure killing the server mid-request.

Related errors

• failed to spawn sd-server: {self._spawn_error}
• sd-server failed to become ready. Last output: {tail[:2000]}
• sd-server is not running.
• sd-server job queue is full (HTTP 429).
• sd-server img_gen submit -> {resp.status_code}: {resp.text[:

AI-assisted analysis of unslothai/unsloth@203007d190 (2026-08-15). Data as JSON: /api/errors/c40a2a768008c199. Report an issue: GitHub.