Are you using Curator’s InterProcessMutex? If so, then you don’t need to worry 
about sync. syncs are implied in every write operation in ZooKeeper. There’s 
almost never a reason to manually do a sync.


> On Oct 13, 2016, at 4:47 AM, liusammy <> wrote:
> Hi guys,
> as you know, zk has feature that write action is confirm after majority of zk 
> nodes successfully write. To read data, client needs to send sync request to 
> zk cluster node then get the newest data.
> currently I use curator to do distributed lock framework.
> I wonder whether or not curator does the “sync” action before trying to lock 
> a znode? Please point out my misunderstanding.
> thanks a lot!
> ​

Reply via email to