[Bug Report]: Retry durable stops when a message is too big to be published

[martinhonovais]

Prerequisites

• I have searched issues to ensure it has not already been reported

Description

The retry durable polling stops working when a message is too big to be published on the topic.
Only the queue of the respective message should be affected.

{
    
Message_str: [KafkaFlow] Exception on queue RetryDurablePollingJob execution producing to retry topic,
    
Data: {
    
QueueId: 8db87b94-585d-46dd-a4b3-67ddc61589ce,
    
ItemId: 38c4a958-ff72-4eda-a5d3-6605e7c22104
},
    
Exception_str: Confluent.Kafka.ProduceException`2[System.Byte[],System.Byte[]]: Broker: Message size too large
   at Confluent.Kafka.Producer`2.ProduceAsync(TopicPartition topicPartition, Message`2 message, CancellationToken cancellationToken)
   at KafkaFlow.Producers.MessageProducer.InternalProduceAsync(IMessageContext context, Nullable`1 partition)
   at KafkaFlow.Producers.MessageProducer.InternalProduceAsync(IMessageContext context, Nullable`1 partition)
   at KafkaFlow.Producers.MessageProducer.<>c__DisplayClass10_0.<<ProduceAsync>b__0>d.MoveNext()
--- End of stack trace from previous location ---
   at KafkaFlow.Producers.MessageProducer.ProduceAsync(String topic, Object messageKey, Object messageValue, IMessageHeaders headers, Nullable`1 partition)
   at KafkaFlow.Producers.MessageProducer.ProduceAsync(String topic, Object messageKey, Object messageValue, IMessageHeaders headers, Nullable`1 partition)
   at KafkaFlow.Retry.Durable.Polling.Jobs.RetryDurablePollingJob.Execute(IJobExecutionContext context)
}

Steps to reproduce

1. Produce a message larger then the retry topic's capacity

Expected behavior

Only the queue with the big message should be affected. The remaining queues should be able to be retried.

Actual behavior

All queues stop to be retried

KafkaFlow Retry Extensions version

3.1.0