zookeeper-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Eric Newton <eric.new...@gmail.com>
Subject Re: Using Zookeeper in a legacy system
Date Mon, 04 Aug 2014 17:35:26 GMT
I don't understand your question.  I suspect others do not either.

If you could explain your goal precisely, we could offer some advice.

"wanted to get each node"

What on each node? Accumulo?  A client of accumulo? Or just a zookeeper
client?

"then programmatically starts Acc and sets the config items?"

You can configure accumulo through the accumulo shell.  These changes will
be stored in zookeeper, and shared across all nodes.

That's my best guess... but feel free to re-post your questions to clarify.

-Eric


On Fri, Aug 1, 2014 at 10:15 AM, Geoffry Roberts <threadedblue@gmail.com>
wrote:

> All,
>
> I discovered ZK when I set up my first Accumulo db.  Since I've been
> reading, but everything I see tells me how to code a ZK client.  Before I
> jump into such a thing, I'd like to gain a better understanding of exactly
> what it is I need to do.
>
> Using Acc as an example, my current installation has each Acc node still
> reading from it's respective config directory .  Q: If I wanted to get each
> node reading from ZK, am I going to be writing some kind of script that
> reads ZK, then programmatically starts Acc and sets the config items?  or
> is there an easier way?
>
> I also have multiple tomcat and mysql instances that could benefit from
> this.
>
> Thanks
>
> --
> There are ways and there are ways,
>
> Geoffry Roberts
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message