that's sorts things out
בתאריך יום ה׳, 1 ביולי 2021, 18:24, מאת Marcus Schäfer <
marcus.schae...@gmail.com>:
> Hi,
>
> > Hi Marcus,
> > What Ran meant is that please run the following command to check the
> > result.
> >
> > bin/kafka-topics.sh --describe --topic TOPIC_NAME --bootstrap-server
Hi,
> Hi Marcus,
> What Ran meant is that please run the following command to check the
> result.
>
> bin/kafka-topics.sh --describe --topic TOPIC_NAME --bootstrap-server
> localhost:9092
Ah, from the kafka tooling. Sorry I was confused because the aws
api name their commands also
Hi Marcus,
What Ran meant is that please run the following command to check the
result.
bin/kafka-topics.sh --describe --topic TOPIC_NAME --bootstrap-server
localhost:9092
Thanks.
Luke
On Thu, Jul 1, 2021 at 8:06 PM Ran Lupovich wrote:
> I meant kafka-topics describe command , I want to
I meant kafka-topics describe command , I want to make sure your topic
really have 20 partitions
בתאריך יום ה׳, 1 ביולי 2021, 14:12, מאת Marcus Schäfer <
marcus.schae...@gmail.com>:
> Hi,
>
> > How many partitions do you have for the topic? Can you issue --describe
> > please and share, thanks
Hi,
> How many partitions do you have for the topic? Can you issue --describe
> please and share, thanks
I've assigned 20 partitions:
aws kafka describe-configuration-revision --region eu-central-1 --arn "..."
--revision 1
base64 --decode ...
auto.create.topics.enable=false
How many partitions do you have for the topic? Can you issue --describe
please and share, thanks
בתאריך יום ה׳, 1 ביולי 2021, 13:31, מאת Marcus Schäfer <
marcus.schae...@gmail.com>:
> Hi,
>
> > Your first understanding is correct, provided each “consumer” means a
> > “consumer thread”
>
> all
Hi,
> Your first understanding is correct, provided each “consumer” means a
> “consumer thread”
all right, thanks
> IMO, Second understanding about message distribution is incorrect because
> there is something called as max poll records for each consumer. Its 500 by
> default. And the time
On which messages goes to which partition, it is decided by producer, so
using round robin assignor doesn't help here in theory.
And for producer, according to https://kafka.apache.org/documentation/
- The DefaultPartitioner now uses a sticky partitioning strategy. This
means that records
Hi Marcus,
Your first understanding is correct, provided each “consumer” means a
“consumer thread”
IMO, Second understanding about message distribution is incorrect because
there is something called as max poll records for each consumer. Its 500 by
default.
And the time between 2 polls is also
Hi,
I was reading through the docs and would like to make use
of kafka's dynamic partition assignment feature in a way
that partitions gets distributed evenly across all consumers.
To my understanding the RoundRobinAssignor would be sufficient
and I left the problem of rebalance if a consumer
10 matches
Mail list logo
