Given your configuration I believe you have another zookeeper process that is 
to be running on locally but on ports 2191:2196? If that is the case, can you 
start that instance with the same configuration file and not get an error? Or 
will that only start if you comment out the server1 line (the opposite of what 
you did for this instance)?

I don’t see you setting the client port in this configuration. If you are 
running two instances on the same machine one needs to be running on a 
different –client-port. Is a different client-port set in server.2’s 

From: Andor Molnar <>
Reply-To: "" <>
Date: Thursday, May 17, 2018 at 2:37 PM
To: "" <>
Subject: Re: Error Starting 2 Node Cluster on Redhat


What does it write on the console exactly?
"...already running as process...?

Check the ZooKeeper PID file for existence.


On Thu, May 17, 2018 at 11:14 AM, THADC <
> wrote:

> Hi, I have a very simple setup (borrowed from Solr Ref Guide "Setting Up a
> Zookeeper Ensemble). For zoo.cfg, I have:
> *tickTime=2000
> initLimit=10
> syncLimit=5
> dataDir=/opt/zooDeployTest/zookeeper/zooDat
> server.1=localhost:2190:2195
> server.2=localhost:2191:2196*
> when I execute ./bin/ start zoo.cfg, it says zookeeper has
> started, but it really hasn't (<> status shows 
> its not running).
> However, if I comment out the server.2 line, it starts just fine. Any ideas
> what the problem might be?
> --
> Sent from: 

Reply via email to