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

Logging problem in transformations

August 26, 2015, 8:57 pm
$
0
0
Hi,

I have developed transformations in spoon and now I want to execute the same through java. I have set the logging parameters such as Log Connection, Log Table name etc;for transformation from the transformation settings. When I run the transformation from Spoon,the logging works and transformation ends successfully.
But when I execute the same KTR in java, I get the following error:

Exception in thread "main" java.lang.IncompatibleClassChangeError: Class org.pentaho.di.core.logging.TransLogTable does not implement the requested interface org.pentaho.di.core.logging.LogTableCoreInterface
at org.pentaho.di.core.database.Database.writeLogRecord(Database.java:3166)
at org.pentaho.di.trans.Trans.beginProcessing(Trans.java:2286)
at org.pentaho.di.trans.Trans.prepareExecution(Trans.java:974)
at org.pentaho.di.trans.Trans.execute(Trans.java:607)
at com.nri.dwh.etl.LoadandExecuteTransformations.runTransformation(LoadandExecuteTransformations.java:31)

Moreover, is there a way to redirect the logs in a log file instead of database? I am using logback.xml which is almost similar to log4j.xml.
I have attached the logback.xml file herewith.
Attached Files
↧
Search

[Problem] Pentaho Community 5.4 - Data Integration problem with germen umlaut

August 26, 2015, 11:53 pm
$
0
0
Hey there,

if I launch Spoon 5.4.0.1-130 i cloudn't see german umlaut/charset (ä,ö,ü,ß) on my result-rows (ETL) and if turn the preferred language (in Spoon) to german i could only see a questionmark for an umlaut on the gui - for my ETL it is real important the show umlaut in the result row.

the ETL is very simple:

1 Step: CouchDB Input (in my CouchDB a can see umlaut)
2 Step: Text File Outpu (with encoding ISO-8859-15)

System:
Windows 8.1 (new installation)
Java 1.6.0_45-b06 (Java 6)
Pentaho Data Integration 5.4.0.1-130 (Spoon)

I hope that anybody can help me to solve this annoying problem.
↧
↧

Get value Estimation errors in Weka API

August 27, 2015, 1:44 am
$
0
0
Hi all,

I'm working with Weka API and I don't know how to get the "Estimation errors" value in C45 algorithm with Weka. Please help!
Attached Images
↧

Pentaho ODBC Connection - OBI Server

August 27, 2015, 5:51 am
$
0
0
Hello Experts,

I am trying to connect to OBI Server using Pentaho Oracle ODBC connection

Pentaho_OBI.jpg

As per the attached image, my Test Connection is good. when i try to import tables I am not seeing any tables listed.
ImportTables.jpg

I am struggling with this for a very long time. your suggestion will be much appreciated!!

Thank you!
Sangeetha
Attached Images
↧

CDA Security

August 27, 2015, 5:55 am
$
0
0
Hi, all users with permissions for execute the cda file, look source xml (dont change and save, but look),
its expose business logical, and very bad. How solve problem ?

Other way to, parameters and functions controlls are expose in page source, and can hacking for
get data hacking.

Att,
↧
↧

Excel with questions

August 27, 2015, 6:04 am
$
0
0
hello, i'm new with Pentaho and i have some questions if it is possible do some things with this "Suite". Should any help me? i have attached a file to answer my question. Could any edit this file if my questions are corrects?
Attached Files
↧

Pentaho Support Experience

August 27, 2015, 1:22 am
$
0
0
I am not sure where to post here so decided to update this thread.

Got in touch with Pentaho local reps as well as with UK office mid July. In the end of the same month managed to get a quick and somewhat unorganised demo which didn't cover few key aspects we specifically requested because something wasn't right with the environment. Agreed to follow up with another session later on. Haven't heard back from them since then, calls not returned, emails not answered.

Same time in parallel started talking to London office. They failed to call me on agreed time followed by another unsuccessful attempt to arrange a call initiated by me again.

In the meantime other vendors keep the line open and do everything they can to sell their solutions while I can't even get a written quote for PDI EE.
/rant off

Does anyone have experience with Pentaho support? How would you rate them?
↧

Writing a custom header to a Text File

August 27, 2015, 2:47 am
$
0
0
Hi Team,

I want to add/prepend 1 static row (which contains 3 fields) in my output csv file which has data (22 fields) in comma separated format.

How i can achieve that, please provide your inputs.
↧
Search

Kettle Transformation Deadlock Issue

August 27, 2015, 7:50 am
$
0
0
Hello, I have a transformation which takes around 9 Million rows as input and outputs unique rows. As part of the processing logic I had to branch into separate flows and then use the Merge Join step to merge the flows back together. Please see the original version (Original_Flow.ktr). With this setup I noticed that the flow just hangs randomly. After referencing the Tx deadlock article http://wiki.pentaho.com/display/EAI/...tion+Deadlocks I modified the flow to (Modified_Flow.ktr) and also increased the "No. of Rows in Rowset" to 50000. This flow ran perfectly fine for a while. When I ran the same flow for a smaller dataset I got the same hang behavior again. 1) How do we debug the flow to find out on which step the flow is hanging? 2) I have configured my JVM to 20GB so can I increase the "No. of Rows in Rowset" to say 100,000 and still be able to run the flow for input sizes ranging from 1 million to 10 million? Does increasing the "No. of Rows in Rowset" value work for smaller input load sizes? 3) What other alternatives can you suggest to avoid the deadlocks? Thanks.Original_Flow.ktrModified_Flow.ktr
Attached Files
↧
↧

Stand Alone CDA

August 27, 2015, 9:02 am
$
0
0
Is it possible to have a stand-alone CDA that is not part of Pentaho's add-on?
↧

Environment for debugging Kettle

August 27, 2015, 9:52 am
$
0
0
Hi,
I'm new to developing in Eclipse so please be lenient with me.

For the developing of a custom Kettle-Step I need to set up an Eclipse environment where I can develop and debug a new transformation step.

I did the following:
  • downloaded the source of the Branch 5.4.0.3 from git
  • ran "ant clean-all resolve create-dot-classpath" as described in the README.txt and it finished with BUILD SUCCESSFUL

Now I would like to compile and run/debug in Eclipse:
So I opened the Project by importing the "Existing Projects into Workspace" with checked option "Search for nested projects" and got all the Kettle-Core,-DB Dialog,-Engine,-UI and PlugIn-Projects and the 'pentaho-kettle'-project. (All projects on the same level!)
As the README.txt says there should be a launch configuration for Run and Debug... but there isn't.

To build the whole project from Eclipse I tried "Build Project" on the 'pentaho-kettle'-project and got tons of errors.
So I tried to run the build.xml as Ant-Task (right click on build.xml ->'run as' ->'Ant Build') got several warnings but also a BUILD SUCCESSFUL!

But I have still no launch configuration and when I try to run the Launch-File (pentaho-kettle-5.4.0.3.launch) it comes up with "Launch configuration pentaho-kettle-5.4.0.3 references non-existing project pentaho-kettle-5.4.0.3."
I guess I messed something up with the project import in Eclipse. Can anybody help me?

My Environment looks like this:
  • Eclipse Luna (4.4.2)
    (with the Apache Ivy PlugIns(Ivy Ant Tasks, IvyDE, IvyDE ResolveVisualizer) and some SVN-Plugins)
    Compiler compliance level is set to 1.7.
  • JDK 7u80
  • ANT 1.9.6

Thanks in advance!
↧

Report Designer - data source question....

August 27, 2015, 9:57 am
$
0
0
Can the data source be a parameterized Stored Procedure ?
↧

System.loadLibrary for dll native methods results in java.lang.UnsatisfiedLinkError

August 27, 2015, 10:18 am
$
0
0
Hi,

I have a dll file in C:\Windows\system32 that is used by a jar. The class in the jar that needs this DLL has several static native methods, and uses the System.loadLibrary() method inside a static initializer. This was throwing an java.lang.UnsatisfiedLinkError: no **** in java.library.path exception.

Finally I was able to get it working by copying the dll and ini file into the libswt/win64 folder.

I couldn't find any documentation on it, so hopefully this helps someone.
↧
↧

Pentaho 5.X - Clean repository jackrabit

August 27, 2015, 12:30 pm
$
0
0
Hi, I have same projects in solutions, but size repository speed increase,
its version artefacts. How clean respositry ?
↧

Adding a new parameter throws an error in the report designer

August 27, 2015, 2:13 pm
$
0
0
I have a designed a report which works well. I am trying to add another sql query in it. Previewing the sql query works without any hiccups.
But when I try to add the parameter for the same query, I get two different errors depending on which mysql-connector I am using.

Earlier I was using the mysql-connector-java-5.0.8-bin and the error
was

- org.pentaho.reporting.engine.classic.core.ReportDataFactoryException: Failed at query: (a few lines down )

Caused by: java.sql.SQLException: Stopped by user.


Then I changed the mysql-connector to mysql-connector-java-5.1.36-bin. The error changed to :-

- org.pentaho.reporting.engine.classic.core.ReportDataFactoryException: Failed at query:

Caused by: com.mysql.jdbc.exceptions.MySQLTimeoutException: Statement cancelled due to timeout or client request


Any suggestions would be helpful. I am using Pentaho 5.3.0.0-213 on Windows 8.1 Pro.
Thanks
↧
Search

Sample Hadoop job Pentaho MapReduce - wordcount fails in MapReduce job

August 27, 2015, 3:06 pm
$
0
0
We have a PDI 5.4 client we need to use with Cloudera 5.4 (using YARN for MapReduce, I think) on a Linux server. We can download files to HDFS (on the Cloudera server) and run ETL (i.e., without map-reduce) on data on that server. However, any job that includes MapReduce fails inexplicably, with only an obscure error message (see below). We've followed instructions found on:
http://wiki.pentaho.com/display/BAD/...ro+and+Version

We installed the cdh54 shim. And made configuration changes accordingly.

See screens below, Job Metrics and logging screens.

(the error message is: Aborting job.
Attached Images
↧

User Defined Java Class - Assignment conversion not possible

August 27, 2015, 8:43 pm
$
0
0
Getting the following error when i click the "Test Class" button on my User Defined Java Class.

org.codehaus.janino.CompileException: Line 61, Column 2: Assignment conversion not possible from type "java.lang.Object" to type "Processor$TeamObject[]"

I'm not sure how to proceed and googling this error hasn't been much help.

pdi-ce-5.4.0.1-130

Code:


import com.google.gson.Gson;
import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;


String perfidField;
String nameField;
String shortNameField;
String awayTeamField;
String homeTeamField;
String primaryField;
String jsonField;


public boolean processRow(StepMetaInterface smi, StepDataInterface sdi) throws KettleException
{
    // First, get a row from the default input hop
    //
    Object[] r = getRow();


    // If the row object is null, we are done processing.
    //
    if (r == null) {
        setOutputDone();
        return false;
    }


    // Let's look up parameters only once for performance reason.
    //
    if (first) {
        perfidField=getParameter("PERFID_FIELD");
        nameField=getParameter("NAME_FIELD");
        shortNameField=getParameter("SHORTNAME_FIELD");
        awayTeamField=getParameter("AWAYTEAM_FIELD");
        homeTeamField=getParameter("HOMETEAM_FIELD");
        primaryField=getParameter("PRIMARY_FIELD");
        jsonField = getParameter("JSON_FIELD");
        first=false;
    }


    // It is always safest to call createOutputRow() to ensure that your output row's Object[] is large
    // enough to handle any new fields you are creating in this step.
    //
    Object[] outputRow = createOutputRow(r, data.outputRowMeta.size());


    String jsonIn = get(Fields.In, jsonField).getString(r);
   
   
    // Set the value in the output field
    //
    //String name = firstname+" "+lastname;
    //get(Fields.Out, nameField).setValue(outputRow, name);
 
    // putRow will send the row on to the default output hop.
    //
    //putRow(data.outputRowMeta, outputRow);


    com.google.gson.Gson gson = new com.google.gson.Gson();
    TeamObject[] obj = gson.fromJson(jsonIn, TeamObject[].class);
   
    for(int i=0;i<obj.length;i++ ) {
                get(Fields.Out, nameField).setValue(outputRow, obj[0].name);
                putRow(data.outputRowMeta,outputRow);
            }


    return true;
}


public class TeamObject {


    public int id;
    public String short_name;
    public String name;
    public int home_venue_id;
    public boolean primary;
    public boolean away_team;
    public boolean home_team;


}
↧
↧

css issue

August 28, 2015, 12:15 am
$
0
0
Hi,

We are using ckeditor to generate HTML code fragments. These HTML code fragments are to be added to Pentaho to generate pdfs.

The HTML is displayed correctly in browser windows (chrome, firefox, IE) but not in pentaho. Some help will be appreciated.

A sample of HTML fragment:

<!-- Start -->

&nbsp;
<table border="0" cellspacing="0" cols="2" frame="void" rules="none">
<colgroup>
<col width="232" />
<col width="96" />
</colgroup>
<tbody>
<tr>
<td align="JUSTIFY" bgcolor="#00ae00" colspan="2" height="37" rowspan="2" sdnum="1033;1033;General" style="border-bottom:
#000000 3px solid; border-left: #000000 3px solid; border-top: #000000 3px solid; border-right: #000000 3px solid" valign="top"
width="327">
<font color="#000000" face="Calibri">All existing rating agencies selected for customer will be displayed.</font></td>
</tr>
<tr>
</tr>
<tr>
<td align="JUSTIFY" bgcolor="#00ae00" colspan="2" height="37" rowspan="2" sdnum="1033;1033;General" style="border-bottom:
#000000 3px solid; border-left: #000000 3px solid; border-top: #000000 3px solid; border-right: #000000 3px solid" valign="top">
<font color="#000000" face="Calibri">All existing customer rating will be displayed.</font></td>
</tr>
<tr>
</tr>
<tr>
<td align="JUSTIFY" bgcolor="#00ae00" colspan="2" height="119" rowspan="2" sdnum="1033;1033;General" style="border-bottom:
#000000 3px solid; border-left: #000000 3px solid; border-top: #000000 3px solid; border-right: #000000 3px solid" valign="top">
<font color="#000000" face="Calibri">For example if &ldquo;Moodys&rdquo; is displayed in rating column then ratings related to moodys
will be available in the drop down. Similarly ratings will be available for S &amp; P, Fitch etc. Existing Rating and Current rating cannot be
same. </font></td>
</tr>
<tr>
</tr>
</tbody>
</table>


<!-- End -->

Thanks and Regards,
Navin
↧

Get Log Information for transformation

August 28, 2015, 1:53 am
$
0
0
Hi all,

I am a beginner of pentaho ETL. I am using pdi-ce 5.4. I created some simple transformation (job entry) and have a high level job to trigger them one by one.

However, I have trouble on gathering the execution information such as the job status, start time, end time. My expectation is to capture those ETL job information to a centrailized log file like below. Does anybody have suggestion ? Thanks.

##################################
Job Log - WORKFLOW_EXTRACTION

Job #: 001
Job name: EXTRACT_CUSTOMER
Status: Finished
Started: 12:30 AM
On date: 8/28/2015
Last ran: 12:30 AM
On date: 8/28/2015
Elapsed time: 00:00:02


Job #: 002
Job name: EXTRACT_PRODUCT
Status: Failed
Started: 12:30 AM
On date: 8/28/2015
Last ran: 05:05 PM
On date: 8/28/2015
Elapsed time: 16:35:00

##################################
↧

MDX Query result join(?)

August 28, 2015, 4:40 am
$
0
0
Hello,

For a client I'm currently implementing an Analyser report in which he want to display the gains/losses of contracts per period (years/months/weeks). For the implementation this means that I need to have a column counting the number of records which started in that period. And for the other column I need to count the number of records which have an end_date in that period.

I can easily do these two things separately in MDX:
This shows me the gains
Quote:
SELECT
{[contract_start_date].[year].MEMBERS}
ON ROWS,
{[Measures].[value]}
ON COLUMNS
FROM Cube
This shows me the losses:
Quote:
SELECT
{[contract_end_date].[year].MEMBERS}
ON ROWS,
{[Measures].[value]}
ON COLUMNS
FROM Cube
But what I actually want in my report is:
Quote:
Year | contract_gains | contract_losses
_______________________________________
2015 | 10 | 2
2016 | 15 | 4
Can anyone help me to solve this issue?
↧
Viewing all 16689 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>