Stopping multiple consumers in a short window freezes on GroupCoordinator._heartbeat_routine()
- 主要言語
- Python
- スター
- 1.4k
- フォーク
- 269
- 平均マージ
- 1日 1時間
- マージ済み PR(30日)
- 6
説明
**Describe the bug**
I have several hundred consumers all connected to the same singular topic with autocommit disabled. Each consumer is part of a unique individual group. When I call `AIOKafkaConsumer.stop()` on each of the consumers within a for loop the task doing the stopping eventually halts completely. When I probe into the task I find that it is stuck waiting on `GroupCoordinator._heartbeat_routine()` and it remains this way indefinitely.
```
wait_for=
# It also seems to happen with GroupCoordinator._commit_refresh_routine(), GroupCoordinator._commit_refresh_task and AIOKafkaClient._sync_task,
```
**Expected behaviour**
GroupCoordinator._heartbeat_routine() should not prevent `AIOKafkaConsumer.stop()` from completing.
**Environment (please complete the following information):**
- aiokafka version (`python -c "import aiokafka; print(aiokafka.__version__)"`): 0.7.1
- kafka-python version (`python -c "import kafka; print(kafka.__version__)"`): 2.0.2
- Kafka Broker version (`kafka-topics.sh --version`): 2.6.0
- Other information (Confluent Cloud version, etc.): Python 3.8.10
**Reproducible example**
```
```
コントリビューションガイド
評価
この issue はまだ評価されていません。