Perhaps, it's a matter of semantics.
But, I think I'm not talking only about failure, but normal operation.
It's normal to take a cluster down for maintenance, or code update. And
this should be done a rolling restart manner (1 server at a time).
The reason for replication, is to increase relia
Hi, all
I'm using kafka-0.7.2
I'd like to commit offset every time reading a message block.
"autocommit.interval.ms" property is not enough.
To do that, should I use SimpleConsumer, and write a commit logic manually?
(i.e. zkUtils.commitOffset(groupId, messageAndOffset.offset() )
Is there no way
You have two choices.
-- Do what you say, and write your own consumer, based on the
SimpleConsumer. Handle all commits, ZK accesses, and balancing yourself.
-- Use a ConsumerConnector for every partition, and call commitOffsets()
explicitly when you have processed a message. This does a commit fo
Thank you for your reply.
The 2nd one is what I want.
I has tested with it, and it works fine. Thanks.
I hope it supports 'consumerConnector.gracefulShutdown()' to finalize a
connector gracefully (with some user properties). :-)
On Mon, Oct 28, 2013 at 10:16 AM, Philip O'Toole wrote:
> You
You can find the api doc in http://kafka.apache.org/documentation.html#api
The 0.8 beta1 release is available from
http://kafka.apache.org/downloads.html
Thanks,
Jun
On Sat, Oct 26, 2013 at 1:57 PM, Smith, Kieran R <
kieran.r.sm...@jpmorgan.com> wrote:
> Hi,
>
> Is there an eta on the 0.8 Kaf
