There have been discussions in the past about using a URL/IP address to
discover server configuration. See ZOOKEEPER-146 and ZOOKEEPER-1031 that are
considered as a part of ZOOKEEPER-107. But you seem to be proposing a scheme
where a client will discover a URL.
What is this URL used for? What is the information available to the client
before initiating the discovery protocol? Can you elaborate more on the use
On Fri, Apr 29, 2011 at 5:11 AM, Michele Girolami <
[EMAIL PROTECTED]> wrote:
> HI all,
> I started using ZooKeeper few weeks ago... I was wondering if ZooKeper
> implements such a mechanism in order to let the clients discover the
> ZooKeper URL. In other world I'd like to implements a ZooKeeper client
> without specify the ZooKeeper URL, but discover the URL dynamically. I've
> taken a look at all the ZooKeeper documentation but I didn't found such a
> Thank you for your precious comments.