4=5>
> for the change of the KIP. I will come up with a solution to throttle
> replica movement when a broker is moving a "leader" replica.
Thanks. It looks great.
>
> On Tue, Jan 24, 2017 at 3:30 AM, Alexey Ozeritsky <aozerit...@yandex.ru>
> wrote:
>
>>
tOffset).
In this case the write speed to topicPartition.move and topicPartition.log will
be the same so this will allow us to move many partitions to one disk.
>
> And to answer your question, yes topicpartition.log refers to
> topic-paritition/segment.log.
>
> Thanks,
> Dong
>
>
23.01.2017, 22:11, "Dong Lin" <lindon...@gmail.com>:
> Thanks. Please see my comment inline.
>
> On Mon, Jan 23, 2017 at 6:45 AM, Alexey Ozeritsky <aozerit...@yandex.ru>
> wrote:
>
>> 13.01.2017, 22:29, "Dong Lin" <lindon...@gmail.co
Hi
22.08.2016, 19:16, "Tom Crayford" :
> Hi,
>
> I don't think I understand *why* you need this. Kafka is by default a
> distributed HA system that balances data and leadership over nodes. Why do
> you need to change this?
In fact, kafka does not balance anything. It uses
Hi,
We have the similar solution that have been working in production since 2014.
You can see it here:
https://github.com/resetius/kafka/commit/20658593e246d2184906879defa2e763c4d413fb
The idea is very simple
1. Disk balancer runs in a separate thread inside scheduler pool.
2. It does not touch