`enable.auto.commit` is a Consumer config and does not apply to Kafka
Stream.
In Kafka Streams, you basically always have auto commit enabled, and you
can control how frequently commits happen via `commit.interval.ms`.
Also on `close()` Kafka Streams would commit offsets.
-Matthias
On 5/31/22 12:29 PM, Luca wrote:
Hi Andy,
The defaults are sensible enough that, under normal operational conditions,
your app should pick up from where it left. To dig a little more into this, I
suggest you look into `auto.offset.reset` and `enable.auto.commit` options.
In case, you do need to reprocess everything, kafka streams comes with a handy
reset tool. You can read about it here:
https://kafka.apache.org/32/documentation/streams/developer-guide/app-reset-tool.html
Luca
On Tue, May 31, 2022, at 5:17 PM, andrew davidson wrote:
Thanks Luca
This is exactly what I was looking for.
On a related note let's say I stop and restart my application. What would I
have to do so that the I do not re process events?
I am still working through the kstreams 101 tutorial. I have not gotten to the
DSL tutorials yet
Andy
On 5/30/22, 11:16 PM, "Luca" <c...@lucapette.me> wrote:
Hi Andy,
If I understand your problem correctly, you want a "foreach" terminal
operation. You can check out the API here:
https://kafka.apache.org/32/documentation/streams/developer-guide/dsl-api.html
Luca
On Tue, May 31, 2022, at 6:37 AM, Andy wrote:
> All the Kstream examples I have found demonstrate how to use map, filter,
> and join on streams. The last step they typically user to() to
> publish/produce the results to a new stream
>
> How can I get the data out of the stream? For example I need to send the
> data to a legacy data that can not use kafka. Or maybe I want to plot the
> data,…
>
> I looked at the java doc and did not find anything
>
> Any idea what I should “google” to to find a code example?
>
> Kind regards
>
> Andy
>
lucapette.me
lucapette.me
