diff --git a/snuba/subscriptions/executor_consumer.py b/snuba/subscriptions/executor_consumer.py
index afb94964266..18c67de7ff4 100644
--- a/snuba/subscriptions/executor_consumer.py
+++ b/snuba/subscriptions/executor_consumer.py
@@ -292,15 +292,27 @@ def poll(self) -> None:
                 )
             except QueryException as exc:
                 cause = exc.__cause__
-                if isinstance(cause, ClickhouseError):
-                    if cause.code in NON_RETRYABLE_CLICKHOUSE_ERROR_CODES:
-                        logger.exception("Error running subscription query %r", exc)
-                        self.__metrics.increment(
-                            "subscription_executor_nonretryable_error",
-                            tags={"error_type": str(cause.code)},
-                        )
-                    else:
-                        raise SubscriptionQueryException(exc.message)
+                # Retryable ClickHouse errors are re-raised so the consumer
+                # crashes and the message is retried. Every other failure
+                # (non-retryable ClickHouse error codes as well as non-ClickHouse
+                # causes such as QueryTooLongException) is non-retryable: log it,
+                # emit a metric and skip the message instead of submitting an
+                # unassigned transformed_message downstream.
+                if (
+                    isinstance(cause, ClickhouseError)
+                    and cause.code not in NON_RETRYABLE_CLICKHOUSE_ERROR_CODES
+                ):
+                    raise SubscriptionQueryException(exc.message)
+
+                logger.exception("Error running subscription query %r", exc)
+                error_type = (
+                    str(cause.code) if isinstance(cause, ClickhouseError) else type(cause).__name__
+                )
+                self.__metrics.increment(
+                    "subscription_executor_nonretryable_error",
+                    tags={"error_type": error_type},
+                )
+                continue
 
             self.__next_step.submit(transformed_message)
 
diff --git a/tests/subscriptions/test_executor_consumer.py b/tests/subscriptions/test_executor_consumer.py
index 15c66353650..9f1b90a5f78 100644
--- a/tests/subscriptions/test_executor_consumer.py
+++ b/tests/subscriptions/test_executor_consumer.py
@@ -1,6 +1,7 @@
 import json
 import time
 import uuid
+from concurrent.futures import Future
 from datetime import datetime, timedelta
 from typing import Iterator, Mapping, Optional
 from unittest import mock
@@ -42,6 +43,7 @@
     get_default_kafka_configuration,
 )
 from snuba.utils.streams.topics import Topic as SnubaTopic
+from snuba.web import QueryException, QueryTooLongException
 from tests.backends.metrics import Increment, TestingMetricsBackend
 
 
@@ -97,7 +99,7 @@ def on_partitions_assigned(partitions: Mapping[Partition, int]) -> None:
     # We need to wait for the consumer to receive partitions otherwise,
     # when we try to consume messages, we will not find anything.
     # Subscription is an async process.
-    assert assigned == True, "Did not receive assignment within 10 attempts"
+    assert assigned, "Did not receive assignment within 10 attempts"
 
     consumer_group = str(uuid.uuid1().hex)
     auto_offset_reset = "latest"
@@ -288,6 +290,55 @@ def test_execute_query_exception() -> None:
     strategy.join()
 
 
+@pytest.mark.redis_db
+@pytest.mark.clickhouse_db
+def test_poll_skips_non_retryable_query_exception() -> None:
+    """
+    A QueryException whose cause is not a retryable ClickhouseError (e.g.
+    QueryTooLongException) must be logged and skipped, not re-raised, and must
+    not fall through to submitting an unassigned transformed_message (which
+    previously raised UnboundLocalError). See SNUBA-9E1.
+    """
+    metrics = TestingMetricsBackend()
+    next_step = mock.Mock()
+
+    strategy = ExecuteQuery(
+        dataset=get_dataset("events"),
+        entity_names=["events"],
+        max_concurrent_queries=2,
+        stale_threshold_seconds=None,
+        metrics=metrics,
+        next_step=next_step,
+    )
+
+    exc = QueryException("boom")
+    exc.__cause__ = QueryTooLongException("query is too long")
+
+    future: "Future[object]" = Future()
+    future.set_exception(exc)
+
+    message = Message(BrokerValue("payload", Partition(Topic("test"), 0), 0, datetime(1970, 1, 1)))
+    result_future = mock.Mock()
+    result_future.future = future
+
+    strategy._ExecuteQuery__queue.append((message, result_future))  # type: ignore[attr-defined]
+
+    strategy.poll()
+
+    assert next_step.submit.call_count == 0
+    assert (
+        Increment(
+            "subscription_executor_nonretryable_error",
+            1,
+            {"error_type": "QueryTooLongException"},
+        )
+        in metrics.calls
+    )
+
+    strategy.close()
+    strategy.join()
+
+
 @pytest.mark.redis_db
 @pytest.mark.clickhouse_db
 def test_too_many_concurrent_queries() -> None: