Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Sqoop >> mail # user >> Import data to HDFS using Sqoop2


Copy link to this message
-
Re: Import data to HDFS using Sqoop2
Yanting,

I'm sorry, I'm a bit confused. The database you are using here is called
"invoice". Do you also have a schema called "invoice"? Try removing the
"Schema name" from your job and see if that works for you.

http://www.postgresql.org/docs/9.1/static/ddl-schemas.html

-Abe
On Tue, Sep 3, 2013 at 6:34 PM, Yanting Chen <[EMAIL PROTECTED]> wrote:

> However, with Sqoop 1.4.4, all schema and table names are lower-case, and
> it is successful to make import.
>
> As a result, I am not sure if case sensitivity will be a problem in Sqoop
> 2.
>
> *This is the connection I created:*
> 1 connection(s) to show:
> Connection with id 1 and name ds_msg_log (Created 9/2/13 10:54 AM, Updated
> 9/3/13 6:35 PM)
> Using Connector id 1
>   Connection configuration
>     JDBC Driver Class: org.postgresql.Driver
>     JDBC Connection String: jdbc:postgresql://MY_IP:MY_PORT/invoice
>     Username: hadoop
>     Password:
>     JDBC Connection Properties:
>   Security related configuration options
>     Max connections: 123
>
> *This is the job I created:*
> Job with id 3 and name ds_msg_log (Created 9/2/13 11:37 AM, Updated 9/3/13
> 5:57 PM)
> Using Connection id 1 and Connector id 1
>   Database configuration
>     Schema name: invoice
>     Table name: ds_msg_log
>     Table SQL statement:
>     Table column names:
>     Partition column name:
>     Boundary query:
>    Output configuration
>     Storage type: HDFS
>     Output format: TEXT_FILE
>     Output directory: /user/root/ds_msg_log_sqoop2
>   Throttling resources
>     Extractors:
>     Loaders:
>
> *The command I use to start this job is:*
> submission start --jid 3
>
>
> On Wed, Sep 4, 2013 at 9:24 AM, Abraham Elmahrek <[EMAIL PROTECTED]> wrote:
>
>> Sorry, I missed that you are using PostGreSQL. I think case sensitivity
>> will be a problem with that database.
>>
>>
>> On Tue, Sep 3, 2013 at 6:23 PM, Abraham Elmahrek <[EMAIL PROTECTED]>wrote:
>>
>>> What database are you using? Are table names or schema names case
>>> sensitive in your database? Sqoop quotes all names, so case sensitivity
>>> makes a difference here.
>>>
>>> Also, could you provide the corresponding Sqoop 1.4.4 command that
>>> worked for you?
>>>
>>> -Abe
>>>
>>>
>>> On Tue, Sep 3, 2013 at 6:06 PM, 陳彥廷(Yanting Chen) <
>>> [EMAIL PROTECTED]> wrote:
>>>
>>>> I am pretty sure that the schema "invoice" exists because I
>>>> successfully imported data to this schema using Sqoop 1.4.4
>>>>
>>>> Also, I already remove "*" when creating job.
>>>>
>>>> The following is the table in invoice schema:
>>>>     TABLE DS_MSG_LOG
>>>>     (
>>>>         MESSAGE_ID VARCHAR2(23) NOT NULL,
>>>>         LOGIN_ID VARCHAR2(30),
>>>>         ACPT_DTS TIMESTAMP(6),
>>>>         DLVR_DTS TIMESTAMP(6),
>>>>         SENDER_ID VARCHAR2(30),
>>>>         SENDER_VAC_ID VARCHAR2(39),
>>>>         RECEIVER_ID VARCHAR2(30),
>>>>         RECEIVER_VAC_ID VARCHAR2(39),
>>>>         STATUS VARCHAR2(1),
>>>>         MESSAGE_TYPE VARCHAR2(8),
>>>>         FLOW_TYPE VARCHAR2(5),
>>>>         SERVICE_TYPE VARCHAR2(1),
>>>>         SOURCE_FILE_NAME VARCHAR2(150),
>>>>         ARCHIVE_FILE_NAME VARCHAR2(250),
>>>>         ARCHIVE_CHAR_COUNT NUMBER,
>>>>         DECRYPT_FILE_NAME VARCHAR2(250),
>>>>         DECRYPT_CHAR_COUNT NUMBER,
>>>>         RESP_FILE_NAME VARCHAR2(250),
>>>>         RESP_CHAR_COUNT NUMBER,
>>>>         RESP_FLAG VARCHAR2(1),
>>>>         RTG_SEQ VARCHAR2(8),
>>>>         RESENT_FLAG VARCHAR2(1) DEFAULT 'N',
>>>>         TOTAL_INV_COUNT NUMBER,
>>>>         CONSTRAINT PK_DS_MSG_LOG PRIMARY KEY (MESSAGE_ID)
>>>>     )
>>>>
>>>>
>>>> On Wed, Sep 4, 2013 at 1:46 AM, Abraham Elmahrek <[EMAIL PROTECTED]>wrote:
>>>>
>>>>> Hey User,
>>>>>
>>>>> It looks like the schema "invoice" does not exist in your database.
>>>>> Could you please provide your database schema? Also, "*" is unnecessary
>>>>> when specifying "Table column names". If you leave it blank it will import
>>>>> all columns by default.
>>>>>
>>>>> -Abe
>>>>>
>>>>>
>>>>> On Tue, Sep 3, 2013 at 3:03 AM, 陳彥廷(Yanting Chen) <