What causes "failed to receive rebuild sink notify" in risingwavelabs/risingwave?

Look for an earlier error/exit of the write-log executor in the same actor's logs

failed to receive rebuild sink notify

Error message

failed to receive rebuild sink notify

What it means

While the sink executor consumes its write log (waiting on a rewind/rebuild future), it also listens on rebuild_sink_rx for messages from the writer (e.g. UpdateConfig from ALTER SINK CONFIG). If the channel closes with no message — the writer task is gone — recv returns None and the executor raises this error instead of silently proceeding.

Source

Thrown at src/stream/src/executor/sink.rs:897

                                    "failed to rewind the log reader"
                                );
                                Err(e)
                            }
                        }
                    } else {
                        Err(e)
                    }
                    .map_err(|e| StreamExecutorError::from((e, sink_param.sink_id)))?;
                }
            });
            let message = loop {
                select! {
                    result = &mut future => {
                        let Err(e): StreamExecutorResult<!> = result;
                        return Err(e);
                    }
                    result = rebuild_sink_rx.recv() => {
                        let message = result.ok_or_else(|| anyhow!("failed to receive rebuild sink notify"))?;
                        // Dropping the consumer costs a rewind, or a recovery when the log
                        // reader cannot rewind. Not worth it for an update that changes nothing.
                        if let RebuildSinkMessage::UpdateConfig(config) = &message
                            && !sink_config_has_changes(&sink_param.properties, config)
                        {
                            info!(
                                executor_id = %sink_writer_param.executor_id,
                                sink_id = %sink_param.sink_id,
                                "skip alter sink config because properties are unchanged"
                            );
                            continue;
                        }
                        break message;
                    }
                }
            };
            drop(future);
            match message {

View on GitHub (pinned to 6469eb736d)

Solutions

Look for an earlier error/exit of the write-log executor in the same actor's logs
Restart the actor so writer and reader halves are re-created together
Verify channel pairing in the sink executor split (writer holds tx, consumer holds rx) hasn't been broken by recent changes
Escalate to RisingWave maintainers if reproducible — it signals an unexpected writer shutdown

Defensive patterns

Strategy: try-catch

Try / catch

let message = match rebuild_sink_rx.recv().await {
    Some(m) => m,
    None => return Err(anyhow!("failed to receive rebuild sink notify")),
};

Prevention

Keep writer and reader halves created and dropped together in the executor split
Log writer-side exits so a closed rx is explainable
Add tests where the writer exits while the consumer is select!-waiting

When it happens

Trigger: rebuild_sink_rx.recv() returns None while execute_consume_log is select!-waiting: the paired writer (execute_write_log side) dropped rebuild_sink_tx, typically because the write-log executor terminated.

Common situations: Writer task exited due to a prior error, actor cancellation racing with log consumption, or misrouted channel wiring after executor restart.

Related errors

AI-assisted analysis of risingwavelabs/risingwave@6469eb736d (2026-09-11). Data as JSON: /api/errors/d890ac70228fcd28. Report an issue: GitHub.