Here is the document for 0.8:
https://cwiki.apache.org/confluence/display/KAFKA/A+Guide+To+The+Kafka+Protocol
They mention there is 1 backward incompatible change but I don't see it
pointed out.
On Mon, Nov 25, 2013 at 5:56 PM, Tom Brown wrote:
> This document describes the wire-format used by
ng on downgrading all our clients and I'll report back.
Sorry for the spam.
On Thu, Jun 13, 2013 at 12:31 PM, Brett Hoerner wrote:
> As an update, this continues to affect us.
>
> First I'd like to note ways in which my issues seems different than
> KAFKA-278,
>
> * I di
an you please try the workaround mentioned in the JIRA description?
>
> Thanks,
> Neha
>
>
> On Tue, Jun 4, 2013 at 4:21 PM, Brett Hoerner >wrote:
>
> > (version 0.7.2)
> >
> > For some reason, my producers are only picking up the partition on 1 of
> my
>
(version 0.7.2)
For some reason, my producers are only picking up the partition on 1 of my
2 brokers. I've been digging through the code, and I don't see any issues
given the state of my ZK nodes. The producer never seems to locate a
partition on Broker0, even though ZK clearly states that it has
Has anyone ever seen something like this on the Kafka (0.7.2) broker side?
I'm running JDK7, and that class most definitely exists
Once this started it just spammed non-stop (for everything each producer
sent to the box, I think) and Kafka was totally wedged. I bounced and it
came back online imme
