got it thanks! i was thinking mostly along the lines of KAFKA-3815 <https://issues.apache.org/jira/browse/KAFKA-3815>, which i like a lot. it would mean creating a cluster per connector. but if it also means creating 3 topics per connector then that makes things a little less convenient.
On Sun, Nov 27, 2016 at 1:24 AM, Ewen Cheslack-Postava <e...@confluent.io> wrote: > If you want independent clusters based on the same Kafka cluster, they need > independent values for config/offset/status topics. The Kafka Connect > framework doesn't provide its own sort of namespacing or anything, so if > you used the same topics in the same cluster, the values between different > Connect clusters targeted at the same Kafka cluster would become conflated. > > -Ewen > > On Sat, Nov 26, 2016 at 7:59 PM, Koert Kuipers <ko...@tresata.com> wrote: > > > if i were to run multiple distributed connect clusters (so with different > > group.id) does each connect cluster need its own offset.storage.topic, > > config.storage.topic and status.storage.topic? or can they safely be > shared > > between the clusters? > > > > thanks! > > koert > > >
