spacedriveapp/spacedrive · error · anyhow::Error

Failed to flush stream: {}

Error message

Failed to flush stream: {}

What it means

flush() on the send stream failed after a chunk was written. Everything written so far was buffered and could not be pushed to the transport, which means the connection is broken. Functionally identical to the write failures: the transport under the stream died and the transfer aborts at the current chunk.

Source

Thrown at core/src/ops/files/copy/strategy.rs:1225

				nonce,
				chunk_checksum: *chunk_checksum.as_bytes(),
			};

		let message_data = rmp_serde::to_vec(&chunk_message)?;

		send_stream.write_u8(0).await?;
		send_stream
			.write_all(&(message_data.len() as u32).to_be_bytes())
			.await
			.map_err(|e| anyhow::anyhow!("Failed to write message length: {}", e))?;
		send_stream
			.write_all(&message_data)
			.await
			.map_err(|e| anyhow::anyhow!("Failed to write chunk data: {}", e))?;
		send_stream
			.flush()
			.await
			.map_err(|e| anyhow::anyhow!("Failed to flush stream: {}", e))?;

		file_transfer_protocol.record_chunk_received(
			&transfer_id,
			chunk_index,
			bytes_read as u64,
		)?;

		bytes_transferred += bytes_read as u64;
		if let Some(callback) = progress_callback {
			callback(bytes_transferred, total_size);
		}

		chunk_index += 1;

		if chunk_index == 1 || chunk_index % 100 == 0 || chunk_index == total_chunks as u32 {
			ctx.log(format!(
				"Sent chunk {}/{} ({} bytes total)",
				chunk_index, total_chunks, bytes_transferred

View on GitHub (pinned to 6dfeccf211)

Solutions

1. Retry the copy job
2. Check peer connectivity and daemon health at the failure time
3. Reduce concurrent transfers if the peer is saturated
4. If reproducible on one link only, capture QUIC logs to identify which side closed

// Flush failure equals a dead connection: retry the transfer, never the flush alone.
match run_push_transfer().await {
    Ok(()) => Ok(()),
    Err(ref e) if e.to_string().contains("flush") => {
        tokio::time::sleep(std::time::Duration::from_secs(2)).await;
        run_push_transfer().await
    }
    Err(e) => Err(e.clone()),
}

When it happens

Trigger: Connection closed by peer or network while buffered chunk data was still pending; QUIC connection error surfacing at flush time; send-side buffer invalidation after a reset.

Common situations: Same environmental causes as mid-stream write failures: peer disconnects, network drops, device sleep events during large transfers.

Related errors

• Failed to write message length: {}
• Failed to write chunk data: {}
• Failed to open stream: {}
• Failed to finish stream: {}
• Transfer error: {}

AI-assisted analysis of spacedriveapp/spacedrive@6dfeccf211 (2026-08-16). Data as JSON: /api/errors/49a802029a67d0f4. Report an issue: GitHub.