GitHub user bossenti added a comment to the discussion: Distributed cluster 
support

> How to fill in multiple host names for adapter and sink?

As of now, we don't support this scenario for both. Although, this should be 
easy to implement, so feel free to file us a feature request.

> When the streams service restarts, it always receives this data from the 
> latest location, including the mqtt adapter, which leads to data loss during 
> the downtime. How can this be resolved?

I think you lost me here. How does the MQTT adapter come into play? Aren't we 
talking about Kafka?

> The data received through the adapter will enter the topic automatically 
> created by local Kafka, and then output through sink. What is the logic for 
> creating this topic in local Kafka? For example, how to set the number of 
> partitions?

The whole data handling in StreamPipes happens internally. There is no need to 
care about topic configuration by the end user. Currently, there is no option 
to influence the number of partitions used by StreamPipes afaik.

GitHub link: 
https://github.com/apache/streampipes/discussions/2912#discussioncomment-9928959

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]

Reply via email to