You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by Seth Weiner <se...@gmail.com> on 2010/03/19 16:36:57 UTC
trouble with ec2 scripts
Hey folks,
I'm trying to fire up the ec2 contrib scripts to launch a simple hbase
cluster. The scripts seem to run fine, as all of the security groups and
instances get created and started. However, when I login to the cluster and
try any operations in the hbase shell I get an error indicating the master's
not running:
NativeException: org.apache.hadoop.hbase.MasterNotRunningException: null
Looking at the pid file in /tmp it seems like the process is actually
running. Any pointers on where to start looking for the root cause?
Thanks!
-Seth
Re: trouble with ec2 scripts
Posted by Andrew Purtell <ap...@yahoo.com>.
Hi Seth,
Very simple. You have to wait for EC2 to launch all of the slaves
before your cluster will come up. The master in your case is
waiting for DFS to initialize. Check the master log under
/mnt/hbase/logs if you want to see the "Waiting for DFS to
initialize..." messages.
The master will also wait until the requested number of slave
instances have all checked in before deploying user tables.
Hope that helps,
- Andy
> From: Seth Weiner <se...@gmail.com>
> Subject: trouble with ec2 scripts
> To: hbase-user@hadoop.apache.org
> Date: Friday, March 19, 2010, 8:36 AM
> Hey folks,
>
> I'm trying to fire up the ec2 contrib scripts to launch a
> simple hbase
> cluster. The scripts seem to run fine, as all of the
> security groups and
> instances get created and started. However, when I
> login to the cluster and
> try any operations in the hbase shell I get an error
> indicating the master's
> not running:
>
> NativeException:
> org.apache.hadoop.hbase.MasterNotRunningException: null