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

pass variables into kitchen job from cmd line

December 5, 2014, 5:55 pm
$
0
0
Hi all

The issue is i want to use a variable in all pdi jobs but i dont want to display it in kettle.properties file. I dont want to edit each job by adding param options or setvariables or getvariables .

So , i am trying to pass variables to kitchen job from command line like -D TEMP_PASSWORD=1234. In the job, i am writing output of TEMP_PASSWORD by using ${TEMP_PASSWORD}. But i am not getting output as 1234 , it is showing as ${TEMP_PASSWORD}.

I even tried passing kettle variables with a new value, but that doesn't worked too. The job is using the old values that are in kettle.properties file.


Can someone please help me, I really need to complete this task.


Note: pentaho version is 5.0.7

Thanks in advance
↧
Search

Java use 2core at 100%

December 6, 2014, 1:08 am
$
0
0
Hello,

Kettle is running version 5 stable on Debian stable. I have 3 jobs very simple. Get files, ftp, move files, delete files and finish. Jobs run every 10 minutes and after 2 days java use 2 core at 100% :

when I use htop I can see : thise process "java -Xmx512 -XX:MaxPermSize=256m" using 100% and running 10 times. No log error. How can I debug or fixes thise issue ?

I have 8Gb of Ram, and mem is free. No swap.


Thanks
↧
↧

Calculating Z-Score

December 6, 2014, 2:13 am
$
0
0
The statistics step in Pentaho PDI computes the mean and standard deviation but does not include a calculation for z-score

I tried to do it by using the statistics step to compute the mean and standard deviation of a field and then feed the input file through again calculating the z-score by subtracting the mean of the field from the value of the field, divided by the standard deviation.

I need to compute the z-score for several fields in a file. I then need to write out the file including the z-score of those fields.

But how do you put two inputs into one step? I need to put the data file in but I also need to put the statistics in too (the mean and standard deviation) in the same step to compute the z-score.

I know you can put multiple files (for example, csv file) to be read into one step but all the files must have the same format

How does one read in, two different files, into a Transformation step?

Thanks for help!
↧

Non-numeric sql fields as parameter combo

December 6, 2014, 5:27 am
$
0
0
We can pass numeric fields as parameter combo as per the below code :
SELECT
DISTINCT brand_name AS "Brand Name",
CASE
WHEN ${param_Measure}='unit_sales' THEN sum(unit_sales)
ELSE sum(store_sales)

END AS "Sales"

FROM product p

I want to pass non-numeric(Sql Column) fields as combo parameter .Is it possible to do . Below is the code am using :

select
case when ${Dimension} = 'gender_code'
then gender_code
when ${Dimension} = 'Grade_code'
then Grade_code
end as "dimension",
sum(leave_count) as leave_count
from hcm_leave_itf_tb
group by ${Dimension}

Kindly give a solution


↧

Can I make Pentaho more tolerant of SQLite locking?

December 6, 2014, 7:39 am
$
0
0
Hi Pentaho gurus,

I want to build a fact table from a staging table and some pre-built dimension tables. All the tables are stored in the same SQLite database.

I have created a transformation to do this in five steps.

1. Read staging table (Table input)
2. Lookup dim1 key (Database value lookup)
3. Lookup dim2 key (Database value lookup)
4. Lookup dim3 key (Database value lookup)
5. Write fact table (Table output)

With the default settings the table output step always fals with an error like this:

Quote:
Unexpected error inserting row
Error comitting connection
database is locked
I suspect the cause is that PDI tries to parallelize the reads and writes, but SQLite requires an exclusive lock on the database to write. (See 3.0 Locking in SQLite documentation).

One workaround is to make the commit size of the table output step greater than the number of rows in the staging table. This feels dirty.

Is there an option that makes PDI write in serial, or wait longer for the locks?

Thanks,
Iain
↧
↧

Creating hyperlink tabs in dashboards

December 6, 2014, 8:21 pm
$
0
0
Hi,
Is there is anyway that i can create tabs in the dashboard which will be hyperlinks to other dashboards. Forexample there will be 4 dashboards with names A, B, C, D. Now i am in dashboard A. all Dashboards will contain tabs at the top of page with tabs A, B, C, D. On clicking tab B on dashboard A it should redirect me to dashboard B.
Thanks,
Rahul
↧

How to design a schema in mondrian4?

December 7, 2014, 4:02 am
$
0
0
Hi,
I want to design a schema in mondrian4,
I can user what tool to design the schema?
Where can i get the tool?
Where can i get the total Introduction to schema in mondrian4?

Looking forward to your answer.
Thank you!
↧

What's the newest link of Kettle sample projects?

December 7, 2014, 7:48 pm
$
0
0
Hi,everyone.
I have got a Kettle 4.4.0 version and I want to Running Jobs in my Java code.So I have read the document called "pdi_embed_extend_guide.pdf".It says I can get the sample projects which contains the sample code from "https://pentaho.box.com/extending-and-embedding-pdi",but this link can't open now!

Can anyone kind tell me the newest link?Thanks a lot!
↧
Search

Geo Map Analysis Report does not reflect data changes

December 7, 2014, 9:49 pm
$
0
0
  • As part of a POC we are developing, we implemented a ​Geo Map Report using Pentaho Analytic Report.
    We were through the whole process, essentially mapping data stored in a standard MS SQL Table, which contains coords data. The report finally displays those coords, alung with some other information added.
    However, if I change the coords of any specific record, and then I refresh the Report (by recalling it or press the refresh button on the browser), the geo position of its corresponding item does not change. I tried by open/closing the browser, clearing the browser cache, but I had no luck.
    What is it happening in this case? I tried to find something on pentaho forums, but I couldn't find any useful information.
    Should anybody may help here, I will really appreciate it.
    Regards,
    Mario







↧
↧

Some enlighten me a bit about pentaho please

December 8, 2014, 3:25 am
$
0
0
Hi guys,

might be the wrong forum, but somehow my question didn't fit anywhere else.

We just got a new ERP System and that one lacks some charts etc.
Now I would like to create some reports, charts, stats, whatever on a central point/website, so my boss can check them from time to time.

After some time I found pentaho and spagoBI, both making my head hurt. After pentaho looks better, I would like to start here, but really need a small kick in the right direction.

I don't need anything fancy, just working statistics. For the reports maybe some grouping/sum functions, that's it.

Is there any guide that can give a me a fast/easy overview over the single programs of pentaho? Or do I only need the BI Platform to reach my answers?
Now problem with reading into all the stuff and learning it, but honestly, I'm lost right now and don't know where to start.

Thanks so much :)
↧

PSW Bug: invalid handling of encoding in CDATA cripples schema files

December 8, 2014, 4:00 am
$
0
0
Pentaho Schema Workbench has a new annotation feature that uses CDATA-sections for the first time.

This however now reveals that PSW has (and apparently always had) a severe encoding problem. It loads schema XML files in UTF8 but always saves the files in ISO Latin-1, thus breaking the file on save.

So you open a schema file, add an annotation, type a non-ASCII character, save the file and close it. But when you try to re-open this file again, PSW claims that the file is now invalid and can't be opened. PSW saved the file in the wrong encoding and now complains over its own mistake.

Currently there is no work-around for the problem, which renders the whole annotation feature dangerously bugged for all non-English applications.

(I had wanted to report that as a bug, but couldn't find a public bug-tracker, since the Pentaho Jira seems to be private.)
↧

Create loops in a job

December 8, 2014, 3:24 pm
$
0
0
Hi all,

I am completely new to using Penthaho (Kettle), I am very impressed how powerful and friendly tool it is. However I am sure I am not using it to its full capacity. I have heard somewhere that I can create 'loops' for repetitive tasks.

I have a process --> get a file (Adobe server) from FTP --> create a table --> Bulk load --> Format Data --> Insert into AGG Table --> Drop the table (for bulkload). This process needs to be repeated for 4 channels. So the question is, do I have to repeat the process and change the table names or can I make it easier (clever).

I have search for loops in a search box but got no results. Any hint would be appreciated. I have also attached the file.

Thanks,

Jakub
Attached Files
↧

CARTE - Error Running Remote Transformation

December 8, 2014, 5:40 pm
$
0
0
Hi guys, I am kind of stuck here and need your help :D

I use my laptop with Kettle CE to create transformations. My data wharehouse and repository are in a Linux CentOS server hosting a PostgreSQL 9.2 db.

I followed the manual that I found in wikipedia and download the data-integration folder to the linux server.

Then I created a .xml file as follows :

cat carte.xml


<slave_config>
<slaveserver>
<name>LINUX Pentaho</name>
<hostname>127.0.0.1</hostname>
<port>8081</port>
<username>cluster</username>
<password>cluster</password>
</slaveserver>
</slave_config>

Then I use the following command to start the Carte Application on the server:

sh carte.sh carte.xml

Done, I use ps -ef and can see the process running. Also, I use netstat -putan and see port 8081 listening.

One more step, I created a TCP ALLOW rule on iptables deamon and restarted it.

Now I should be able to connect, right?

What I do, I open my transformation, create a slave server with the following configuration :

1.jpg

Then I launch the job using the remote server:

2.png

But then, I get the following error.

how to resolve it? Password was cluster

3.png
Attached Images
↧
↧

Kettle best approach for job

December 8, 2014, 9:50 pm
$
0
0
Hi guys,

very new to kettle, only 2 days in so I am finding my way around. I have set up a job which quite simple.

Pick up a file, process, and load. How ever that process is for 8 different channels so I would need to repeat that task 8 times. Is there a way so I can re use the same tasks only with different conditions?

(Loops, repeat...)

Any hint will be appreciated.

Thanks,

Jakub
Attached Files
↧

Convert binary data into string

December 9, 2014, 1:19 am
$
0
0
Hi All, I'm new here

I would like to ask some questions about transferring one data to another and vice versa, if the database' field on a table have a pattern of binary numbers which for example from monday to sunday = 1111110 one represent as their schedule and zero represent no schedule. How do I transfer or/and convert them using Pentaho and the output of it have a column from monday to sunday and the rows will have either Yes or No, yes represent as their is a schedule which in the binary is equals to 1 and no represents as no schedule which in the binary is zero.

I uploaded an image for you to see, in the left part is the binary which come from my database date pattern and in the right part is the output I want. Thank you for any help.
Attached Images
↧
Search

Saiku OLAP wizard - "OK" button not working in Pentaho 5.2 (CDE)

December 9, 2014, 1:49 am
$
0
0
Hello all,

unsing Pentaho 5.2 i can open the SAIKU OLAP Wizard and create an Analysis. But when clicking the "ok" button nothing happens. Anyone experiencing the same issue?

Thanks,
David
↧

Change chart query dinamically

December 9, 2014, 4:53 am
$
0
0
Hello.

I've created a dashboard in CDE. In this dashboard there's a form with 3 options and a chart whose query must change accordingly to the option I choose in the form. Alternatively I could render different charts according to the option (if the first solution isn't possible).

But I still didn't find a way to implement any of those solutions. Does anybody have a suggestion on how to solve this?

Thanks in advance!
↧
↧

CARTE - Error Running Remote Transformation

December 9, 2014, 5:35 am
$
0
0
Hello guys,

I am trying to run my transformations in a Linux CentOS server via Carte, but it does not work.

The server contains a PostgreSQL database, where the data will be inserted.

I copied the data-integration folder to the server and configured a carte.xml file, as follows:

<slave_config>
<slaveserver>
<name>LINUX Pentaho</name>
<hostname>127.0.0.1</hostname>
<port>8081</port>
<username>cluster</username>
<password>cluster</password>
</slaveserver>
</slave_config>


Then I used that file to start Carte up.

nohup sh carte.sh carte.xml &



After that, I checked with the netstat command that the service is running and listening.

netstat -putan

tcp 0 0 ::ffff:127.0.0.1:8081 :::* LISTEN 8507/java




Now, I went to Spoon interface (on a remote host) , opened my transformation and configured a Slave Server as in the image below:

1.jpg

Then, executed the transformation and asked to run in a remote host.

2.png

And got the following Error:

Permission Denied : Connect


So, what can be the cause of this Permission Denied? Port 8081 is opened in the iptables as in the rule below:

-A INPUT -m state --state NEW -m tcp -p tcp --dport 8081 -j ACCEPT


Thanks in advance.
Attached Images
↧

CCC Line Chart: How to make legend items selectable, but not chart points

December 9, 2014, 10:37 am
$
0
0
Hi everyone,

I am currently trying to make a dashboard to display solar power generation over the course of a day. Since there are many panels, I would like to easily be able to switch between which are displayed, without having to wait for the plot to re-render every time that I click on a legend item. I have written a function which I have bound to the selectionChangedAction which keeps a list of selected points and updates the chart when certain criteria are matched.

My question is: How can I allow the user to click on a legend item to make the data corresponding to that legend item "selected", but have the selection remain unchanged when they click an actual point on the curve?

Thank you very much!
↧

Report designer height controls

December 9, 2014, 7:34 pm
$
0
0
Hi

Is there is anyway that i can restrict the height of pentaho report designer page. I want to show only report header data. I want to compress all the space of page header, group, report footer and page footer. Is there is anyway to avoid the scroll bar when we publish the same report in dashboard.


Thanks,
Rahul
↧
Viewing all 16689 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>