Object reference null while trying to produce to topic on higher loads during perf testing
See original GitHub issueDescription
We are doing performance testing and trying to produce around 24 million messages to a topic with Partition Count 36 and Replication Factor 3. We are calling ProduceAsync method to push messages. We are getting below exception on higher loads:
System.NullReferenceException: Object reference not set to an instance of an object.
at Confluent.Kafka.Impl.SafeKafkaHandle.marshalHeaders(IEnumerable1 headers) at Confluent.Kafka.Impl.SafeKafkaHandle.Produce(String topic, Byte[] val, Int32 valOffset, Int32 valLength, Byte[] key, Int32 keyOffset, Int32 keyLength, Int32 partition, Int64 timestamp, IEnumerable
1 headers, IntPtr opaque)
at Confluent.Kafka.Producer2.ProduceImpl(String topic, Byte[] val, Int32 valOffset, Int32 valLength, Byte[] key, Int32 keyOffset, Int32 keyLength, Timestamp timestamp, Partition partition, IEnumerable
1 headers, IDeliveryHandler deliveryHandler)
at Confluent.Kafka.Producer2.ProduceAsync(TopicPartition topicPartition, Message
2 message, CancellationToken cancellationToken)
How to reproduce
Checklist
- Confluent.Kafka 1.7.0.
Issue Analytics
- State:
- Created 2 years ago
- Comments:6 (3 by maintainers)
Top GitHub Comments
We went with other alternatives. I am closing the ticket for now.
@CwPiyush I tried to reproduce the issue locally on my system with 25M messages having headers, partition count= 36, and replication factor = 3 with the Benchmark test and didn’t face any troubles in Producer.
To further help in the investigation, can you please do: