February 7, 2014, 7:49 am
Could you please help me find the size of the largest cluster ?
Class attribute: flightbooking
Classes to Clusters:
0 1 2 <-- assigned to cluster
0 100 0 | senior
8 20 4 | student
0 70 0 | worker
Cluster 0 <-- No class
Cluster 1 <-- student
Cluster 2 <-- No class
Incorrectly clustered instances : 182.0 46.6667 %
↧
February 7, 2014, 9:07 am
Hi,
SO: Linux
PDI: 4.4.0
Java(TM) SE Runtime Environment (build 1.6.0_20-b02)
I was building a job with several transformations (lets say 2) that uses a "Load file content in memory" step with content from url. This setup works great with only ONE transformation that uses a "Load file content in memory" but when a job chain 2 (or more) transformations with this kind a step (even as a serial call) the second transformation dont call the proper content (It uses the content from the first trnasformation).
I have been able to replicate it in a very simple way...
- Startup PDI;
- Create a New transformation (lets call it t1);
- Add the step "Load file content in memory";
- Add a file (url1);
- Preview data from the step;
- Add output file;
- Create a New transformation (lets call it t2);
- Add the step "Load file content in memory";
- Add a file (url2);
- Preview data from the step;
- Add output file;
(***) This shall return the same data for both...
- Create a Job;
- Add [Start];
- Add transformation [T1];
- Add transformation [T2];
- Add Sucess;
If u save all... close PDI and open the t2 (only) it preview the proper data.
On top of that.... if u run the job as:
[START]---->[T1]---->[T2]---->[Sucess]
At [T2] execution it will read [T1] data...
The funny part is that running from shell (kitchen) it seens to work proper....
I made some search and didn't find any about it! It is a bug or this behavior is the expected?
↧
↧
February 7, 2014, 12:52 pm
Hello all,
My team is working on a project to embed Kettle in a webapp. That process is going well, but we have written several custom plugins that are not available to the Kettle execution process, so our jobs/transformations cannot load and execute. We are using 5.0.1.
Can someone please provide guidance on how to programmatically make Kettle aware of our plugins?
Thanks!
Clint
↧
February 8, 2014, 9:50 pm
Hello Experts,
I am using kettle to migrate Oracle schema (to another Oracle schema) and stuck with this scenario -
CREATE OR REPLACE TYPE ssn AS OBJECT(ssn NUMBER(10));
-- table not getting migrated when type is referenced as column
CREATE TABLE person
(
NAME VARCHAR(40) NOT NULL,
refssn ssn NOT NULL
);
Please guide how this migration scenario can be handled through kettle.
Thank you.
↧
February 8, 2014, 11:53 pm
hello,
I know that we have in schema workbench Dimensions and dimension usage, How about Metadata, can I use the same principle here as well.
meaning , Can I have general business Views and then use them in more than Business Model ?
if not, what is the best way to copy the same views from one model to anther ?
thank you
Amanda
↧
↧
February 9, 2014, 5:46 am
I have OlapConnection instance, and i try clear schema cache:
RolapConnection privateConnection = s.OlapConnection.unwrap(RolapConnection.class);
CacheControl cacheControl = privateConnection.getCacheControl(null);
cacheControl.flushSchemaCache();
but the dimension elements is not updated. How to clear all cache of OlapConnection?
↧
February 9, 2014, 6:37 am
Hi,
I want to execute kettle transformation using java API.I am using pentaho 5.0.1 version and i have imported the kettle jobs into the solution repository.I have the following statements
KettleEnvironment.init();
JobMeta jobMeta = new JobMeta("testrepo\\test.kjb", null);
Job job = new Job(null, jobMeta);
job.start();
job.waitUntilFinished();
But it is throwing an error message as the kettle is trying to find the test.kjb in the tomcat bin folder instead of the solution repository.Can someone help in resolving this issue.
↧
February 9, 2014, 4:02 pm
Im new to this so I am trying to do a simple job> trying to export the whole database in 1 excel file where the Sheetname = TableName
So I tried to use Get Table Names but it only brings 47 tables.. I have 164 in total.
I am using Table Input .... generating table names via "SHOW TABLES" SQL statement.
That gives me the list, then I normalize it and Sort Rows then Get Rows from result
All good
Now I am trying to use My Bulk Loader
I want to load all tables... So how do I reference Target Table as row values from my SQL statement?
Im thinking ${internal.step.name} but I have no idea how to use this variable? Do I put in the step name after it?
Then on my Excel export, how do I reference the same way the results by sheetname/tablename and get data accordingly?
↧
February 9, 2014, 7:51 pm
i have one table its for sales target and the target amount insert for 1 year , and i have to divide 12 (12 month) to get amount each month
how i can get 12 row with 12 result from amount 1 year and divide 12 ?
↧
↧
February 9, 2014, 9:51 pm
Hi All,
I want to hide X and Y axis values in the line graph.
Please let me know how to do this.
Thanks,
malibu
↧
February 10, 2014, 12:20 am
Hello!
I have a task to calculate trend based on our cube data. It's seems that there is a special MDX function for that task - LinRegSlope(). But I've got a strange result, even on SteelWheelsSales cube.
My MDX is below.
Code:
with set Years4 as {LastPeriods(4, [Time].[All Years])}
member [Measures].[R] as LinRegSlope([Years4], [Measures].[Quantity])
select {[Measures].[R]} ON COLUMNS,
Hierarchize(Union({[Time].[All Years]}, [Time].[All Years].Children)) ON ROWS
from [SteelWheelsSales]
Result in that case is 'NaN'. Does anybody use this function on Mondrian?
↧
February 10, 2014, 1:23 am
Hi,
I would like to present count of records in dashboard - like: "Count of records: #". I am using text component and I assume, there has to be some javascript code in expression property
# value can be retrieved from a data source.
Is there a way, how to do it? I found only posts related to get parameter values, but no info about getting data from datasource.
Could you please help?
Thank you
Magdalena
↧
February 10, 2014, 2:06 am
Good morning, I am testing Pentaho Data Integration to see if it would meet my data integration needs.
I need to do a java transformation using a complex program that we are using
I can't find a good documentation explaining how I can use third party libraries.
I have created a lib ext folder, which I have added to the class path.
I now have a conflict with the libraries. I am using JPA, Spring, ...
It seems that Pentaho is using JPA1.0 while my dependencies are on JPA 2.0, which has been deployed to the lib ext folder.
What is the way to manage this in Pentaho. Nothing clear in the documentation. Can we manage the libs, as in a typical application server?
Thanks and regards
Gilles
↧
↧
February 10, 2014, 3:35 am
Hello,
Is there a way to disable cache for all reports? I have many reports and cant change one by one...
Maybe some config in BI Server that removes that?
Thanks
↧
February 10, 2014, 3:49 am
Guys
I am unable to build my pentaho projects as pentaho maven repository at this location
http://repository.pentaho.org/artifactory
is down since more than 8 hours.
any clue as to what could be the reason? or repository has moved to a new location?
Or if admin of the repository is online then please have a look and make the system up.
Utkarsh
↧
February 10, 2014, 4:27 am
Hi, let's assume that I have typical time tame with typical time hierarchy defined on it - Year->Month->Day. Now if I take the year and show it in the pentaho tool I will have only unique values for years visible and its ok for me, but if I now take only month, the values seem to repeat for each year, so instead of having an aggregation only through month I get data aggregated through month and year. How should i define this dimension in order to have separate aggregation for all elements of the hierarchy? please have in mind that I still need the usage of all date functions available in mondrian.
↧
February 10, 2014, 8:19 am
Hi
I am having database connection problems. I'd be grateful for some help.
I am testing a connection in Metadata Editor and and get Invalid JNDI connection BlueDW : No suitable driver found for jdbc:mysql://dw1.uat.paypoint.net:3306
My JNDI name is BlueDW, the extract from jdbc.properties is
BlueDW/type=javax.sql.DataSourceBlueDW/driver=com.mysql.jdbc.Driver
BlueDW/url=jdbc:mysql://dw1.uat.paypoint.net:3306
BlueDW/user=****
BlueDW/password=****
I have copied mysql-connector-java-5.1.17.jar into meta-data-editor/ metadata-editor/jdbc/ and meta-data-editor/lib/
Here is more error detail:
org.pentaho.di.core.exception.KettleDatabaseException:
Error occured while trying to connect to the database
Invalid JNDI connection BlueDW : No suitable driver found for jdbc:mysql://dw1.uat.paypoint.net:3306
at org.pentaho.di.core.database.Database.normalConnect(Database.java:415)
...
at java.lang.reflect.Method.invoke(Method.java:606)
at org.pentaho.commons.launcher.Launcher.main(Launcher.java:110)
Caused by: org.pentaho.di.core.exception.KettleDatabaseException:
Invalid JNDI connection BlueDW : No suitable driver found for jdbc:mysql://dw1.uat.paypoint.net:3306
at org.pentaho.di.core.database.Database.initWithNamedDataSource(Database.java:435)
at org.pentaho.di.core.database.Database.connectUsingClass(Database.java:457)
at org.pentaho.di.core.database.Database.normalConnect(Database.java:399)
... 39 more
JNDI Name : BlueDW
↧
↧
February 10, 2014, 10:01 am
I have a newly installed CE 5.0.1 on Linux. The sample reports in Steel Wheels do not execute from the PUC. The PUC just downloads the file to my desktop client.
Is the community version not able to execute .prpt files from the PUC?
or
Did I miss setting something up during installation?
↧
February 10, 2014, 10:28 am
Hi,
I've been trying to get my head round the new jackrabbit repository and digging through the code have run into things like org.pentaho.platform.repository2.mt.RepositoryTenantManager. Pretty exciting!
I can see a few tests around this stuff, but it's not exposed in the REST API. I know MT was on the roadmap for 5 at one point, but can't find anything in the release notes. Does anyone know the status of this?
I'm in the process of implementing something along the lines of
BISERVER-6330 Need reference implementations for tenant-aware datasource service. Any pointers to using this new stuff greatly appreciated!
Matt
↧
February 10, 2014, 10:28 am
Hi everyone,
I was wondering if someone could help me answer this question.
how do you recognize noisy attributes from a data set ?
Thank you
↧