Re: Kafka behaviour on contolled shutdown of brokers

2018-04-23 Thread Brett Rann
i) no.
ii) yes you do, no it won't.

:)

You used the word replace, not add. Is the final state of your cluster 6
nodes, or 12?

If you're replacing, you might want to consider just replacing one at a
time to avoid having to do reassignments:

a) stop 1 broker, lets say broker "1".
b) start up the 7th node, but as broker "1". It would start replicating the
assigned topics from the online brokers.

When everything is ISR again, continue in pairs with the remaining 5.

There's an issue of throttling though and cluster performance while that
happens, and if you're uncomfortable with handling that it may be easier to
do it the way you proposed along with throttled partition reassignments
using the reassignment tooling.


On Sat, Apr 14, 2018 at 2:31 AM, Ashutosh Bharti <
ashutosh.bha...@hotmail.com> wrote:

> I have a kafka cluster  with 6 brokers and over 60 topics, with a
> replication factor of either 2 or 3. We plan to replace all the existing
> brokers with new nodes.
>
>
>  I have 2 questions:
>
>  i) once we add 6 new nodes to the cluster making it total 12 nodes, and
> shut down old brokers one by one, will the topics move to new brokers
> automatically?
>
> ii) If not, we will have to move them using the reassignment tool, in that
> case do we need to '__consumer_offsets' topic as well, or kafka will take
> care of that itself?
>
> Regards,
> Ashutosh
>
>


Kafka behaviour on contolled shutdown of brokers

2018-04-13 Thread Ashutosh Bharti
I have a kafka cluster  with 6 brokers and over 60 topics, with a replication 
factor of either 2 or 3. We plan to replace all the existing brokers with new 
nodes.


 I have 2 questions:

 i) once we add 6 new nodes to the cluster making it total 12 nodes, and shut 
down old brokers one by one, will the topics move to new brokers automatically?

ii) If not, we will have to move them using the reassignment tool, in that case 
do we need to '__consumer_offsets' topic as well, or kafka will take care of 
that itself?

Regards,
Ashutosh