looks like some config error. Can you upload initial logs for both the
servers?
One user is sufficient inter broker communication.
On Wed, Aug 2, 2017 at 11:04 AM, Alexei Levashov <
alexei.levas...@arrayent.com> wrote:
> Hello Manikumar,
>
> I appreciate your advice , thank you.
>
> I tried to us
Hi Jan,
Thanks for taking the time to put this together, appreciated. For the
benefit of others would you mind explaining a bit about your motivation?
Cheers,
Damian
On Wed, 2 Aug 2017 at 01:40 Jan Filipiak wrote:
> Hi all,
>
> after some further discussions, the best thing to show my Idea of
Hi Karan,
I noticed that you posted this to the Confluent Google Group as well. Let's
discuss it over there. This mailing list is only for Apache Kafka.
Thanks,
Ismael
On Wed, Aug 2, 2017 at 7:11 AM, karan alang wrote:
> Hello, here is the update ..
>
> when i ran script - kafka-preferred-repl
Hi Apurva,
For the ProduceRequest,
- The increase is from 470 ms to around 1.004 s.
- The average batch size (batch-size-avg) is around 320B.
- The linger time is 10ms.
However, the 99th percentile for OffsetCommit has increased from 1.08 to
2.8 seconds.
Best Regards,
Gaurav Abbi
On Tu
Hi Jan,
Thanks for the effort in putting your thoughts down on paper.
Comparing what I see from your proposal and what is presented in KIP-182,
one of the main differences is the exclusion of an`Materialized` instance
in the `KTable` methods.
Can you go into more detail why this is so and the s
Hello Stevo,
Thanks for your inputs. I think your comments can be further categorized
into the following points:
1. Is config request.timeout.ms leaky of internal implementations?
Your observation that "request.timeout.ms" is related only to NetworkClient
is right, but I think that does not mean
Hello,
>From Kafka Documentation - I understand that Authentication and encryption
can be enabled for inter-broker, broker-client exchanges. By exchanges, i
mean data transfer-related activities.
My questions are:
1) Is it common to have ONLY inter-broker and broker-client exchanges, but
only pl
Hi Bill,
totally! So in the original discussion it was mentioned that the
overloads are nasty when implementing new features. So we wanted to get
rid of them. But what I felt was that the
copy & pasted code in the KTableProcessors for maintaining IQ stores was
as big as a hurdle as the overloa
Hello Jan,
Thanks for your proposal. As Bill mentioned the main difference is that we
extract the user-customizable materialization logic out of the topology
building DSL workflow. And the main motivations are in two folds:
1) efficiency wise, it allows some KTables to not be materialized if
unne
Hello Manikumar,
I set log level to debug and couldn't define appropriate filtering so logs
are a bit verbose.
I still can not interpret the failure.
Thx,
-AL
On Wed, Aug 2, 2017 at 12:01 AM, Manikumar
wrote:
> looks like some config error. Can you upload initial logs for both the
> servers?
>
10 matches
Mail list logo