sqoop-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Narasimha Tadepalli <Narasimha.Tadepa...@actian.com>
Subject RE: Can I run only sqoop1 client without prerequisite of Hadoop?
Date Thu, 22 Jan 2015 20:18:18 GMT
Hi Abraham

Thanks for the response. I do know that sqoop1 does write to local fs. But my main question
is can I only run the sqoop1 without having Hadoop installed in the system?

Thanks
Narasimha

From: Abraham Elmahrek [mailto:abe@cloudera.com]
Sent: Wednesday, January 21, 2015 6:36 PM
To: user@sqoop.apache.org
Subject: Re: Can I run only sqoop1 client without prerequisite of Hadoop?

Hey there,

I'm assuming you'd like to use Sqoop to transfer to a local file such that you can transport
it out of your closed environment? If so, I'd check out "local" fs support: https://sqoop.apache.org/docs/1.4.5/SqoopUserGuide.html#_using_generic_and_specific_arguments.
Essentially, you can write to the local file system with that.

AFAIK, Sqoop1 doesn't support FS => HDFS data transfers. In Sqoop2, such a general data
transfer use case is being worked on. You can use HDFS "put" in the mean time I'd imagine.

-Abe

On Wed, Jan 21, 2015 at 4:20 PM, Narasimha Tadepalli <Narasimha.Tadepalli@actian.com<mailto:Narasimha.Tadepalli@actian.com>>
wrote:
We have very complex distributed environment where our Hadoop and oracle database are in two
different private network infrastructures. We wanna use sqoop1 to import the oracle database
in avro and then transport that data close to HDFS environment. And then export using sqoop1
into Hadoop systems. Is there a way I can only run sqoop1 without Hadoop prerequisite in the
system? May be by just adding few dependency jars to the path?

Thanks
Narasimha

Mime
View raw message