Question: how to mitigate broker unavailable at context start with kafka-clients >= 0.9

[aftermath02]

Hi,
First of all, thanks a lot for building this appender which is very useful !

Would it be possible to clarify the below statement coming from the documentation ?

If all brokers are not reachable when the logging context starts, or all brokers become unreachable for a longer time period (> metadata.max.age.ms), your appender will eventually block. This behavior is undesirable in general and can be migitated with kafka-clients 0.9 (see #16). Until then, you can wrap the KafkaAppender with logback's own AsyncAppender.

How can the problem of Kafka broker unavailable at context start be mitigated through kafka-clients >= 0.9 (without having to wrap the kafka appender with "async") ?

Note: when doing unit-testing with embedded-kafka (please refer to spring-kafka), the kafka-appender freezes the test (most probably because the broker is not available by the time the appender starts).

Thank you very much in advance for your expertise and your time.

Best Regards