Grokbase Groups Hive user June 2011
FAQ
Hi all,

I have setup a hadoop cluster with 4 nodes running on Centos 5.4 machines.
Right now I had configured hive in the master node (hadoop server node) and
i am able to run hive queries via jdbc client (web application)
successfully.

Is it required to configure the hive server in the same machine running as
master? Ie.. Do we need to configure hadoop and hive server in the same
node?

I wish to setup hive server in a separate machine and hadoop server in the
other. Is it possible?

If so, kindly give me your suggestions.

Thank you,
--



Regards,
Sangita

--



Regards,
Sangeetha S
SSNCE.

Search Discussions

  • Ravi . at Jun 5, 2011 at 4:57 am
    It's possible to setup separate node as hive client and it's very common
    practice.

    It's not required to setup Hive client on the any of the Hadoop master nodes
    or slave nodes (NameNode ,JobTracker ,DataNode)
    You can setup Hive client on a separate node which can connect to the
    NameNode and JobTracker.
    To use separate machine as a hive client copy Hadoop configuration files
    from cluster on client machine and export it in env before using hive.

    On Sat, Jun 4, 2011 at 2:23 AM, sangeetha s wrote:

    Hi all,

    I have setup a hadoop cluster with 4 nodes running on Centos 5.4 machines.
    Right now I had configured hive in the master node (hadoop server node) and
    i am able to run hive queries via jdbc client (web application)
    successfully.

    Is it required to configure the hive server in the same machine running as
    master? Ie.. Do we need to configure hadoop and hive server in the same
    node?

    I wish to setup hive server in a separate machine and hadoop server in the
    other. Is it possible?

    If so, kindly give me your suggestions.

    Thank you,
    --



    Regards,
    Sangita

    --



    Regards,
    Sangeetha S
    SSNCE.
  • Sun, Michael at Jun 5, 2011 at 5:23 am
    Do you need to copy all hive jars into hardoop lib?


    From: Ravi .
    Sent: Sunday, June 05, 2011 12:56 AM
    To: user@hive.apache.org <user@hive.apache.org>; sangeetha s <sangee.sha@gmail.com>
    Subject: Re: configuring Hive server and Hadoop server in separate machines


    It's possible to setup separate node as hive client and it's very common practice.

    It's not required to setup Hive client on the any of the Hadoop master nodes or slave nodes (NameNode ,JobTracker ,DataNode)
    You can setup Hive client on a separate node which can connect to the NameNode and JobTracker.
    To use separate machine as a hive client copy Hadoop configuration files from cluster on client machine and export it in env before using hive.


    On Sat, Jun 4, 2011 at 2:23 AM, sangeetha s wrote:


    Hi all,

    I have setup a hadoop cluster with 4 nodes running on Centos 5.4 machines. Right now I had configured hive in the master node (hadoop server node) and i am able to run hive queries via jdbc client (web application) successfully.

    Is it required to configure the hive server in the same machine running as master? Ie.. Do we need to configure hadoop and hive server in the same node?

    I wish to setup hive server in a separate machine and hadoop server in the other. Is it possible?

    If so, kindly give me your suggestions.

    Thank you,
    --



    Regards,
    Sangita

    --



    Regards,
    Sangeetha S
    SSNCE.





    This message, including any attachments, is the property of Sears Holdings Corporation and/or one of its subsidiaries. It is confidential and may contain proprietary or legally privileged information. If you are not the intended recipient, please delete it without reading the contents. Thank you.
  • Ravi . at Jun 5, 2011 at 5:41 am
    No, But you need to Hadoop and Hive binary files to client. Just copy the
    Hadoop home and Hive home dir contents to client and export it as
    HADOOP_HOME and HIVE_HOME.
    On Sat, Jun 4, 2011 at 10:23 PM, Sun, Michael wrote:

    Do you need to copy all hive jars into hardoop lib?

    *From*: Ravi .
    *Sent*: Sunday, June 05, 2011 12:56 AM
    *To*: user@hive.apache.org <user@hive.apache.org>; sangeetha s <
    sangee.sha@gmail.com>
    *Subject*: Re: configuring Hive server and Hadoop server in separate
    machines

    It's possible to setup separate node as hive client and it's very common
    practice.

    It's not required to setup Hive client on the any of the Hadoop master
    nodes or slave nodes (NameNode ,JobTracker ,DataNode)
    You can setup Hive client on a separate node which can connect to the
    NameNode and JobTracker.
    To use separate machine as a hive client copy Hadoop configuration files
    from cluster on client machine and export it in env before using hive.

    On Sat, Jun 4, 2011 at 2:23 AM, sangeetha s wrote:

    Hi all,

    I have setup a hadoop cluster with 4 nodes running on Centos 5.4 machines.
    Right now I had configured hive in the master node (hadoop server node) and
    i am able to run hive queries via jdbc client (web application)
    successfully.

    Is it required to configure the hive server in the same machine running
    as master? Ie.. Do we need to configure hadoop and hive server in the same
    node?

    I wish to setup hive server in a separate machine and hadoop server in the
    other. Is it possible?

    If so, kindly give me your suggestions.

    Thank you,
    --



    Regards,
    Sangita

    --



    Regards,
    Sangeetha S
    SSNCE.
    This message, including any attachments, is the property of Sears Holdings
    Corporation and/or one of its subsidiaries. It is confidential and may
    contain proprietary or legally privileged information. If you are not the
    intended recipient, please delete it without reading the contents. Thank
    you.
  • Sun, Michael at Jun 5, 2011 at 5:49 am
    Great! Thx


    From: Ravi .
    Sent: Sunday, June 05, 2011 01:40 AM
    To: user@hive.apache.org <user@hive.apache.org>; Sun, Michael
    Cc: sangee.sha@gmail.com <sangee.sha@gmail.com>
    Subject: Re: configuring Hive server and Hadoop server in separate machines


    No, But you need to Hadoop and Hive binary files to client. Just copy the Hadoop home and Hive home dir contents to client and export it as HADOOP_HOME and HIVE_HOME.


    On Sat, Jun 4, 2011 at 10:23 PM, Sun, Michael wrote:


    Do you need to copy all hive jars into hardoop lib?


    From: Ravi .
    Sent: Sunday, June 05, 2011 12:56 AM
    To: user@hive.apache.org <user@hive.apache.org>; sangeetha s <sangee.sha@gmail.com>
    Subject: Re: configuring Hive server and Hadoop server in separate machines


    It's possible to setup separate node as hive client and it's very common practice.

    It's not required to setup Hive client on the any of the Hadoop master nodes or slave nodes (NameNode ,JobTracker ,DataNode)
    You can setup Hive client on a separate node which can connect to the NameNode and JobTracker.
    To use separate machine as a hive client copy Hadoop configuration files from cluster on client machine and export it in env before using hive.


    On Sat, Jun 4, 2011 at 2:23 AM, sangeetha s wrote:


    Hi all,

    I have setup a hadoop cluster with 4 nodes running on Centos 5.4 machines. Right now I had configured hive in the master node (hadoop server node) and i am able to run hive queries via jdbc client (web application) successfully.

    Is it required to configure the hive server in the same machine running as master? Ie.. Do we need to configure hadoop and hive server in the same node?

    I wish to setup hive server in a separate machine and hadoop server in the other. Is it possible?

    If so, kindly give me your suggestions.

    Thank you,
    --



    Regards,
    Sangita

    --



    Regards,
    Sangeetha S
    SSNCE.




    This message, including any attachments, is the property of Sears Holdings Corporation and/or one of its subsidiaries. It is confidential and may contain proprietary or legally privileged information. If you are not the intended recipient, please delete it without reading the contents. Thank you.




    This message, including any attachments, is the property of Sears Holdings Corporation and/or one of its subsidiaries. It is confidential and may contain proprietary or legally privileged information. If you are not the intended recipient, please delete it without reading the contents. Thank you.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupuser @
categorieshive, hadoop
postedJun 4, '11 at 9:24a
activeJun 5, '11 at 5:49a
posts5
users3
websitehive.apache.org

People

Translate

site design / logo © 2021 Grokbase