Quantcast
Channel: Pentaho Community Forums
Viewing all 16689 articles
Browse latest View live
↧

Reading data from external process or using 'Execute a process' step

January 30, 2015, 7:33 am
$
0
0
Hello,

I have a program which creates a report out of our database. It used to write to a file,
but I added an option that instead it can also write its output to standard output. The
goal is to use Kettle as a filter. While I realize that I could write the file, then let Kettle
read the contents from the (zipped) file, I would really like to forego the use of the file.

So I created a transformation which starts the program through the 'Execute a process'
step. This succeeds. What does not succeed is to get the output back as a stream of
rows. I have tried all possible combinations for the output line delimiter ($[0A], $[0D],
$[0A,0D] and $[0D,0A]) but it insists on returning all output in a single line. However,
I do need to filter from about 1000000 lines.

Are there other options for reading the output from a process in a transformation? I
searched, but found none. Using Java to link a process seems to defeat the purpose
of Kettle, that is, to have a relatively easy environment to link up and process data
streams. Besides, the program generating the data is written in C#, so if I have to
resort to Java to make Kettle work, I might as well proceed to drop Kettle for the
job and write the rest of the process in C#.

Regards,

Jurgen

(I wanted to add my .ktr file as an example, but the firewall here does not let me)
↧
Search

How to install IBM Websphere MQ Consumer

January 30, 2015, 7:39 am
$
0
0
hi,

I'm noob in pentaho, but i need comunicate with Websphere MQ, searching in internet, i found a input step called Websphere MQ consumer, with the documentation for installing, in this documentation indicate install the libraries for working the plug in, but not indicate how to install the step, is this downloadble for the pentaho site?, the link of documentation is this:

http://wiki.pentaho.com/display/EAI/...M+Websphere+MQ

please i need a help with this, thanks!!


regards
↧
↧

Out of index error: move to next sheet!

January 30, 2015, 9:41 am
$
0
0
Hi, I've got a transformation that is failing. The input step is from an .xls spreadsheet, and although the transformation says it runs successfully, I'm getting a "Out of index error: move to next sheet!" error, and no data is being written at the output. Also, I can preview the data fine from the input step.

Has anyone seen this?

Thanks,

Rick

Spoon - Kettle 4.1.0
↧

Can we build live dashboard in Pentaho

January 30, 2015, 9:55 am
$
0
0
Just wondering if we can build live dashboard in Pentaho, which can update automatically, say every minute?

I noticed some fancy dashboard in Kibana which can update automatically. Can we implement similar dashboard in Pentaho?

Thanks.
↧

Converting Integer to String but getting error couldn't convert string to Integer

January 30, 2015, 1:06 pm
$
0
0
Hi friends,

I am reading data from CSV file and given data types for all columns. One column has integer data type and CSV file has values as (#3 not correct Int type). So getting error in CSV Input' step which reads data. But the trasnformation continue to execute and won't stop. Now, I want to store that value in String datatype (a new variable) but its giving error saying 'Couldn't convert string to integer'. I am using 'Meta-Data' tabb of 'Select / Rename values'

Steps following:
1. Get all CSV file names (using Get file names)
2. Read the CSV file data (using CSV Input task) - Here it will give error about type conversion
Get file data.jpg
3. Now I am using 'Select / Rename values' task to change the data type from Integer to String. Created a new variable in Metadata tab but getting error "Unexpected conversion error while converting value [Code] to an Integer"
Fields2.jpg

Is there anyway I can get column's error value into new column with String data type.

Thanks,
Nilesh
Attached Images
↧
↧

failover

January 30, 2015, 11:31 pm
$
0
0
Hi,

I have configured clusters with one master and 4 slaves. Now how to handle if any of the slave or master node gets failed how can we recover the data?
How to handle node failover issue using Pentaho Data Integration.
↧

folder

January 30, 2015, 11:34 pm
$
0
0
How to create a new folder in BI Server :

For example by default it will have two folders namely
1) Home
2) Public

Suppose if i want to create one more directory with Samples how can i do it. Finally in biserver it has to be like
1) Home
2) Public
3) Samples

Please Suggest me
↧

run batch commande with kettle

February 1, 2015, 10:39 am
$
0
0
Hi all,
I want to run a batch file with dynamic arguments,

I got the arguments from CDE dashboard,
I use 5.2.0 version of Pentaho EE,
I am on windows, and i have no ssh.

Wich step in Pentaho DI (kettle) can be used in this case ?

Thanks.
↧
Search

Connect the PDI CE to Hadoop

February 1, 2015, 7:04 pm
$
0
0
Hi all,

We want to do a demo which load the data from local file to Hadoop system.
does the PDI community version 5.1.0 can do that? if you have an experience, could you attach a sample?


Many thanks,
chen
↧
↧

chart not displayed in pentaho report

February 1, 2015, 10:26 pm
$
0
0
Hi,

I am a fresher to penatho reporting.Here i have created one report that displayes line chart depending on postgres sql query. Its working fine in PRD tool. But I am trying to generate html output by passing prpt file to java code and it generates only text an not displaying line chart. I want show line chart in html and pdf output and i am using penta 3.8 version.


Can anyone help me to solve this problem?.
If you have any source code to display chart from web application using prpt file, could you please send me?


thanks in advance
Ashok
↧

Same job different tables

February 2, 2015, 1:46 am
$
0
0
Hi, I'm new in this, and I'm in my first project.

I need to extract data from a db and put it a fact and dimension db.

My first step is CDC.
I decide to do it in a stage database that have for each table that i need, two tables with only the original field that i need and a hash field.
One is for the last load and one for the actual load.

I have a job that do this: get data from original database, add the hash field, put data in "actual load" table, do a merge rows (diff) using pk and hash with the "last data" table and than i use Sinchronize after merge. Everything work fine.

But I have 10 different table where i need to do this, each one need same work but have different field.
How can do this? There is a sample to take inspiration?
My first solution is create 10 transformations, one for each table, but I think that will exists something to do this.

Thank all for help and sorry for spanglish...
↧

Questions about new installations

February 2, 2015, 5:53 am
$
0
0
Hello, I had to upgrade a Pentaho to server to latest version, so I decide to install a new version to testing.


I have a Debian Wheezy virtual machine and I make this steps:


1. Install Java:


Code:
apt-get install openjdk-7-jre-headless

2. Download and execute Business Analytics Platform:


Code:
wget http://sourceforge.net/projects/pentaho/files/Business%20Intelligence%20Server/5.2/biserver-ce-5.2.0.0-209.zip
unzip biserver-ce-5.2.0.0-209.zip
biserver-ce/
./start-pentaho.sh
3. Access to the web interface with "admin" user:


http://ServerIP:8080/pentaho/Home

I have some newby quiestions:

1. Is Debian a good OS to run Pentaho?
2. Is OpenJDK good Java software? Is better to install Oracle Java?
3. The web interface is different that we have, is it possible to change it? Themes?

I will appreciate any help, this is my first Pentaho installation.

Regards.
↧

Message box

February 2, 2015, 6:15 am
$
0
0
Is it possible to get a message box when a user select a wrong parameter. I have 2 parameters (Start Date and End Date). Both can be selected with a date picker. Is it possible that the user get a message in a message box when he selects a End Data which is before the Start Date?
↧
↧

Link from sub reports elements to a report element

February 2, 2015, 7:03 am
$
0
0
In my report I am trying to link the sub-reports to an element in the parent report. In this case the element is a group header $(fieldname) which html-anchor is already linked to the report 'table of contents'. I tried _PARENT but I cannot make it work, it navigates only to the first record of the data list instead to its parent.

Anybody knows how to solve this.
Thanks in advance.
Regards
↧

CDE Parameter

February 2, 2015, 9:15 am
$
0
0
Hello,
I used the OLAP Selector Wizard and OLAP chart wizard to create a pie chart to show some data. Things seem to work and the parameter Year works just fine. I now have a Query component that is working with SQL query data source and I want to pass the same parameter to it and I have been unable to get it to work.

This what i have tried

- Created a Query Component.
- Assigned its listener to listen to Year parameter
- In the parameters of query component i copied the same ARGS and VAL as pie chart (i.e. filter_academic_yearParameter, filter_academic_yearParameter)
- In the data source, I created a new sql over sqljndi data source and placed my query. The where clause looks like "Where AcademicYear = ${academic_year}"
- In the data source parameters I have name as filter_academic_yearParameter and value as a static value of 2012

however selecting the drop down - does not change anything on this query component value. I have no idea what am I missing here. Any help is greatly appreciated.


Thanks,
↧
Search

Hive Partition on date(from timestamp)

February 2, 2015, 10:01 am
$
0
0
I have partition the data according to date.

In Hive if i have a EMP table with 5 columns
EMP : name string,
age int,
salary decimal,
loc string,
join_date timestamp.

I want to load the same data in EMP_HISTORY table partition on join_date(only date and avoid timestamp)...
but in my EMP_HISTORY i want 5 columns should appear and join_date column should contain data with date & timestamp but partition should happen only on date(from join_date).
How can i proceed.
--------------------
EMP_HISTORY : name string,
age int,
salary decimal,
loc string,
join_date timestamp.
PARTITION BY(join_date only date)..
ultimately i want only 5 columns as per EMP table and partition on date of join_date.
↧

[Useful Tips] PDI Spoon easter egg - Building transformations faster

February 2, 2015, 10:21 am
$
0
0
I could not believe when someone showed this to me last week, in a "oh, did you know.." kind'a a way.

My reaction was immediate - "How come I didn't know about this?!?! Freakin' amazing!"

Turns out, there's a very very cool "easter egg" in PDI. Instead of dragging the element (transformation step or job entry) to the spoon canvas, we can double click on that element and it will be placed on the canvas already connected with the previously selected element!

Screenshoting it: Put a step on the canvas

Put a step on the canvas, select it
Double click on some other step - magic happens!

The step will appear connected to the selected one

And there's more. What do we all do after putting a step in the canvas? Yeah, we double click on it to configure the step. So now, try to hold the shift key while you double click. If you're a heavy kettle user by now you should be saying something on the lines of "holy cow!!" :)

Shift-double-click will immediately open the step dialogue

And that's it for today's section of "Useful Tips". Have fun!


-pedro




More...
↧
↧

ERROR ...Row Preview - ADDRESS_ID None : Unknown type 0 specified.

February 2, 2015, 2:28 pm
$
0
0
Greetings experts,

When calling a stored procedure with the "Call DB Procedure" transformation, I get this error:

Quote:
2015/02/02 16:04:50 - Row Preview - ERROR (version 5.2.0.0, build 1 from 2014-09-30_19-48-28 by buildguy) : org.pentaho.di.core.exception.KettleValueException:
2015/02/02 16:04:50 - Row Preview - ADDRESS_ID None : Unknown type 0 specified.
All of the parameters are String:

Code:
CREATE OR REPLACE PROCEDURE SCOTT.Get_Addr_Id
(
  P_Address_Line_1      VARCHAR2
 , P_Address_Line_2      VARCHAR2
 , P_City                VARCHAR2
 , P_Zipcode            VARCHAR2
 , P_Address_Id  IN OUT VARCHAR2
)
IS ...
and I call it like this:
DBAddr.jpg

I have tried everything. Originally the "ADDRESS_ID" was type "NUMBER" in the database, but I was getting a parameter type mismatch so I changed to string.

Any help is appreciated.
:)
Attached Images
↧

Color kml shapes on the CDE Map Component?

February 2, 2015, 8:02 pm
$
0
0
Hello all,

I am working with various maps. I have a kml file with shapes of the US States. The shapes are on the map, the shapes are shaded based on customers in each state. The colors, however, are Green, Yellow, and Red, which seems to be the default. Does anybody know how I can change the colors of the shapes shading? I filled out the "Colormap" parameter on the Map Component settings but that did nothing. The Pentaho sample map dashboard with shapes also has the ugly Green, Red, Yellow colors.

Any help would be appreciated.
↧

User console login error

February 2, 2015, 11:15 pm
$
0
0
Hi all,
I am getting problem during accessing user console login.
how can i access or login for Penatho dashboard.
Even user name:- admin and password as :- password is not working.
Can anyone help me out?
For reference see attached screenshot.

Thanks
Sandeep
Attached Images
↧
Viewing all 16689 articles
Browse latest View live
Search