You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@bigtop.apache.org by David Starina <da...@gmail.com> on 2014/04/14 21:25:49 UTC
Problem running HBase
Hi,
I'm new to the Hadoop world. I have tried installing Hadoop + HBase +
Zookeeper on a single node via Bigtop (in a virtual machine CentOS). I'm
trying to create a table in HBase, and get the following error:
client.HConnectionManager$HConnectionImplementation: Check the value
configured in 'zookeeper.znode.parent'. There could be a mismatch with the
one configured in the master.
I checked the configuration in hbase-site.xml and it is empty. How should I
configure it to run on a single node?
Best regards,
David Starina
Re: Problem running HBase
Posted by Sean Mackrory <ma...@gmail.com>.
You're welcome. I suspect the problem may have been trying to start the
RegionServer in stand-alone mode, when only the Master is run. It was
probably looking for something the Master would do if the Master was
running in distributed mode.
On Mon, Apr 14, 2014 at 2:58 PM, David Starina <da...@gmail.com>wrote:
> I have changed the configuration as you have advised (using localhost for
> hostname) (I have not modified the Zookeeper config). Got an error when
> restarting hbase-master, but after restarting the machine, it works!
>
> Thank you, your help is very much appreciated.
>
> Best regards,
> David
>
>
> On Mon, Apr 14, 2014 at 9:41 PM, Sean Mackrory <ma...@gmail.com>wrote:
>
>> By default, HBase starts in a "standalone" mode which just involves one
>> daemon, and will not use an external instance of zookeeper. With no
>> configuration, this is started by just running 'service hbase-master
>> start'. I prefer to use turn on distributed mode and just run everything on
>> the same box. I usually just have this in hbase-site.xml:
>>
>> <configuration>
>> <property>
>> <name>hbase.cluster.distributed</name>
>> <value>true</value>
>> </property>
>> <property>
>> <name>hbase.rootdir</name>
>> <value>hdfs://`hostname`:8020/hbase</value>
>> </property>
>> </configuration>
>>
>> With this configuration, `service hbase-master start` and `server
>> hbase-regionserver start` you should get a usable service as long as
>> ZooKeeper is running on the local host as well. I've never run into the
>> 'zookeeper.znode.parent' problem, but this should at least allow you to add
>> other configuration tweaks if needed. Have you modified the Zookeeper
>> configuration in /etc/zookeeper/conf at all? For single node operation you
>> shouldn't need to. Just 'service zookeeper-server init' and 'service
>> zookeeper-server start' (and the former isn't strictly necessary, I think).
>>
>>
>> On Mon, Apr 14, 2014 at 1:25 PM, David Starina <da...@gmail.com>wrote:
>>
>>> Hi,
>>>
>>> I'm new to the Hadoop world. I have tried installing Hadoop + HBase +
>>> Zookeeper on a single node via Bigtop (in a virtual machine CentOS). I'm
>>> trying to create a table in HBase, and get the following error:
>>>
>>> client.HConnectionManager$HConnectionImplementation: Check the value
>>> configured in 'zookeeper.znode.parent'. There could be a mismatch with the
>>> one configured in the master.
>>>
>>> I checked the configuration in hbase-site.xml and it is empty. How
>>> should I configure it to run on a single node?
>>>
>>> Best regards,
>>> David Starina
>>>
>>
>>
>
Re: Problem running HBase
Posted by David Starina <da...@gmail.com>.
I have changed the configuration as you have advised (using localhost for
hostname) (I have not modified the Zookeeper config). Got an error when
restarting hbase-master, but after restarting the machine, it works!
Thank you, your help is very much appreciated.
Best regards,
David
On Mon, Apr 14, 2014 at 9:41 PM, Sean Mackrory <ma...@gmail.com> wrote:
> By default, HBase starts in a "standalone" mode which just involves one
> daemon, and will not use an external instance of zookeeper. With no
> configuration, this is started by just running 'service hbase-master
> start'. I prefer to use turn on distributed mode and just run everything on
> the same box. I usually just have this in hbase-site.xml:
>
> <configuration>
> <property>
> <name>hbase.cluster.distributed</name>
> <value>true</value>
> </property>
> <property>
> <name>hbase.rootdir</name>
> <value>hdfs://`hostname`:8020/hbase</value>
> </property>
> </configuration>
>
> With this configuration, `service hbase-master start` and `server
> hbase-regionserver start` you should get a usable service as long as
> ZooKeeper is running on the local host as well. I've never run into the
> 'zookeeper.znode.parent' problem, but this should at least allow you to add
> other configuration tweaks if needed. Have you modified the Zookeeper
> configuration in /etc/zookeeper/conf at all? For single node operation you
> shouldn't need to. Just 'service zookeeper-server init' and 'service
> zookeeper-server start' (and the former isn't strictly necessary, I think).
>
>
> On Mon, Apr 14, 2014 at 1:25 PM, David Starina <da...@gmail.com>wrote:
>
>> Hi,
>>
>> I'm new to the Hadoop world. I have tried installing Hadoop + HBase +
>> Zookeeper on a single node via Bigtop (in a virtual machine CentOS). I'm
>> trying to create a table in HBase, and get the following error:
>>
>> client.HConnectionManager$HConnectionImplementation: Check the value
>> configured in 'zookeeper.znode.parent'. There could be a mismatch with the
>> one configured in the master.
>>
>> I checked the configuration in hbase-site.xml and it is empty. How should
>> I configure it to run on a single node?
>>
>> Best regards,
>> David Starina
>>
>
>
Re: Problem running HBase
Posted by Sean Mackrory <ma...@gmail.com>.
By default, HBase starts in a "standalone" mode which just involves one
daemon, and will not use an external instance of zookeeper. With no
configuration, this is started by just running 'service hbase-master
start'. I prefer to use turn on distributed mode and just run everything on
the same box. I usually just have this in hbase-site.xml:
<configuration>
<property>
<name>hbase.cluster.distributed</name>
<value>true</value>
</property>
<property>
<name>hbase.rootdir</name>
<value>hdfs://`hostname`:8020/hbase</value>
</property>
</configuration>
With this configuration, `service hbase-master start` and `server
hbase-regionserver start` you should get a usable service as long as
ZooKeeper is running on the local host as well. I've never run into the
'zookeeper.znode.parent' problem, but this should at least allow you to add
other configuration tweaks if needed. Have you modified the Zookeeper
configuration in /etc/zookeeper/conf at all? For single node operation you
shouldn't need to. Just 'service zookeeper-server init' and 'service
zookeeper-server start' (and the former isn't strictly necessary, I think).
On Mon, Apr 14, 2014 at 1:25 PM, David Starina <da...@gmail.com>wrote:
> Hi,
>
> I'm new to the Hadoop world. I have tried installing Hadoop + HBase +
> Zookeeper on a single node via Bigtop (in a virtual machine CentOS). I'm
> trying to create a table in HBase, and get the following error:
>
> client.HConnectionManager$HConnectionImplementation: Check the value
> configured in 'zookeeper.znode.parent'. There could be a mismatch with the
> one configured in the master.
>
> I checked the configuration in hbase-site.xml and it is empty. How should
> I configure it to run on a single node?
>
> Best regards,
> David Starina
>