You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@zookeeper.apache.org by Narayanan A R <na...@gmail.com> on 2012/05/29 20:32:21 UTC

ZK disk usage

Hi,

How does ZK persistence work. Is it synchronous? The documentation says it
is a change journal like implementation. I assume a write request will not
return until the data is replicated to all the machines in the cluster and
each machine has successfully persisted to the disk?

Regards,
ARN

Re: ZK disk usage

Posted by Michi Mutsuzaki <mi...@cs.stanford.edu>.

Hi ARN,

ZooKeeper returns success when write is persisted (i.e. fsync'ed to
disk) to the majority of the cluster.

http://zookeeper.apache.org/doc/r3.1.2/zookeeperInternals.html#sc_guaranteesPropertiesDefinitions

--Michi

On Tue, May 29, 2012 at 11:32 AM, Narayanan A R
<na...@gmail.com> wrote:
> Hi,
>
> How does ZK persistence work. Is it synchronous? The documentation says it
> is a change journal like implementation. I assume a write request will not
> return until the data is replicated to all the machines in the cluster and
> each machine has successfully persisted to the disk?
>
> Regards,
> ARN