I have a question, in pentaho 5.1, how can I deploy a cube without using the schema workbench? I'm kind of newbie in Pentaho.
Is there a cmd line? Java code? Or something like...
Thanks a lot!
You can do that in the User Console.
There is a menu Manage Data source... There you can upload your xml and refer to a database connection for it.
First, I suppose you have installed BA Server and have made at least fact table.
In case you don't know what the fact table is, or someone else is reading this answer, you can find brief explanation here.
Of course, it's better to have full Star Schema. You cannot create Snow Flake inside Pentaho User Console. You can create it with Pentaho Schema Workbench or by manually edit mondarian.xml.
Make sure that your JDBC driver is inside BA Server driver directory. Look! Open Pentaho User Console. It's by default at localhost:8080/pentaho or yourdomain.name:8080/pentaho and login as administrator
File -> New -> Data Source
Choose Data source
Type Choose fact table and define connections to dimensions (if exists)
Choose to modify cube on the end of data source wizard
.
Related
I created an schema in Schema Workbench and publish with no errors, but when I got in BI Server with the standard user admin, I choose New->jpivot, then it display the name of the schema I created but it does not display the cube. For reference the error I get from catalina.out is:
17:11:45,174 ERROR [PentahoDataSourceResolver] PentahoXmlaServlet.ERROR_0002 - IDatasourceService.UNABLE_TO_INSTANTIATE_OBJECT
org.pentaho.platform.api.data.DBDatasourceServiceException: javax.naming.NameNotFoundException: Name [Esquema Salario] is not bound in this Context. Unable to find [Esquema Salario].
Name [Esquema Salario] is not bound in this Context errors usually appear if you use JNDI name, which is not defined on your system. So, I assumed, that this is the name of the datasource, which you reference while publishing Mondrian schema files to BI server.
Xml file with Mondrian schema definition generated by Schema Workbench does not contain any information regarding how to connect to database. So, you need to specify these details when you upload your schema file on BI server (It's done on step 4 below).
But first you have to create the connection itself (steps 1-2):
Create new JDBC datasource:
Define connection parameters:
If cubes still don't appear after these steps, you may republish your cube: follow same steps as in step 1, but select "Analysis" instead of "JDBC" in the end.
Upload the xml file, generated by schema workbench and select the datasource, which you have created on step 2. .
If cube still does not appear - check your log again. If you see the same Name is not bound error, you may try to restart your BI server application (new connections usually get recongnized immediately, but if you had a connection with same name before, than you might need to restart tomcat).
If that does not work, than once again, check log files. I guess, you'll have some different error in this case.
I had the same problem as the OP (blank screen when clicking New View) with the latest version of Pentaho BI server 7.1 (latest at the moment) and even with the 6.0 version one, Pivot4J SNAPSHOT 1.0 plug in version (latest as of today), Schema Workbech 3.14 (latest as of today).
And as, in line with OP, my catalina.out log was also spitting the Name [DatasourceName] is not bound in this Context. Unable to find [DatasourceName].
After several trials and errors I noticed the problem showed up when I checked the "Register the XMLA Data Source" when publishing the schema on Schema Workbench. So to fix the problem I just unchecked it before publishing.
Another way to fix this is going to the Manage Datasources option on the BI server, Import Analysis, choosing the schema created by Schema Workbench, AND manually setting the data source parameter value EnableXmla to false and saving changes. Now the schema should show up when clicking on Create New > Pivot4J view.
I'm creating cubes (xml schemas) via schema-workbench or ivy schema editor.
When I'm publishing it, I would like to know where the schemas (mondrian.xml files) are actually saved,What is the location of these files?
Thanks,
Which version of Pentaho BA server are you using? Pre or Post 5.0?
Pre 5.0: you choose the file path when publishing. The path is under you ${BISERVER}/pentaho-solutions.
5.0 and beyond: there's no physical file, it's stored in Pentaho's Jackrabbit repository only.
If you create any Analysis in Mondrian 5.*, and you have a test server and prod server, don't export Mondrian schemas from one to another. For a mystical reason you won't be able to get rid of them later.
This was my case, when I started searching for the pentaho-solutions/system/olap/datasources.xml in order to delete the problematic mondrian scheme. This source just doesn't exist anymore.
All data is now saved with the help of Jackrabbit. Jackrabbit stores your Mondrian Schemas together with all other Reports and Analysis to the database (the path is given in the jackrabbit preference file). But in the database you can only see their IDs. So there is no chance to get rid of one object - you leave everything or delete everything by truncating the table. The main problem is, that in the same table are saved not only schemas, but also all other reports, which you have downloaded to the server.
I've created a new database using Microsoft SQL Server Management Studio, and now I want to interact with it through LabVIEW. I already have several VIs to interact with a previous database, using the database connectivity tool kit. This database was created by someone who has since left the project and I can't find it in anything but LabVIEW.
I'm quite experienced with LabVIEW, but completely new to and bewildered by databases.
Thank you in advance.
The first Connectivity Toolkit VI called should be Open Connection.
The existing code (VI) will either use a file or a string as an input.
If the input is a string, then you will need to create a new connection string compatible with your server. You can find common SQL Server strings at https://www.connectionstrings.com/sql-server-2008/
If the input is a file name, you can copy the .UDL file that is referenced and then modify the copied file by opening it (double click) and then select the OLE DB Provider for SQL Server and then set the connection options to point to your server, database etc. and then test the connection.
Basically the workflow you have to go through is the following:
Open connection
Execute your query
Fetch data (if needed)
Close connection
If you search for "Database" in the NI Example Finder shipped with Labview you will find a few good starting points.
In particular give a look to Database Connection.vi and Database Fetching.vi.
If you plan to use transactions try also Database Transaction.vi.
I found that the solution to my problem was to create a .udl file and use that as the file path for opening the database connection.
Here's the address that taught me how to do this:
http://msdn.microsoft.com/en-us/library/e38h511e(v=vs.71).aspx
Thank you to everyone who submitted answers, they certainly helped point me in the right direction.
I created a sample database in hive and some sample tables in it.when I am connecting to Hive from my BI tool 'Tableau'I am unable to view the databases that are created.Only the defualt database is shown.
Could anyone help me with the issue so that I can access the newly created database in Tableau.
Thank You,
Divya Teja Y.
First: Make sure that you are calling your hive server connection from the same directory (or referencing the same metastore) as the one in which you created the database/tables.
Second: When you open up the Cloudera Hadoop Connection, you should be able to see your database under "Step 4: Select a schema on the server" Click the magnifying glass.
I am using netbeans IDE 7.1. I am trying to generate entity classes from a database(sql server). I am able to setup the connection to this remote datasource but in the New Entity Classes from database Wizard the tables are not showing up and at the bottom it says select atleast one table.
I could execute queries and browse through the tables from the netbeans databases service.
Can someone help me to fix this..
Thanks
maybe your schema is not selected correctly, check on service windows->database->your connection(the bold schema is a default selected schema) if it wrong, right-click->properties, and change your schema to public, all the tables that you created are here normally
hope useful :)
I faced this issue and the only way I could get it to work was to create a new connection every time. Once i created a new connection it would show me the tables and I could use it, otherwise it would just leave the list empty
In the newer versions of NetBeans if you create a database connection with default schema, it will take mysql as schema name and generate the connection string.
So goto your DB connection in NetBeans and edit your connection url.
Example :
Change
jdbc:mysql://localhost:3306/mysql?zeroDateTimeBehavior=CONVERT_TO_NULL
to
jdbc:mysql://localhost:3306/YOUR_SCHEMA_NAME?zeroDateTimeBehavior=CONVERT_TO_NULL (add your schema name in the connection string and save)
I had to go into Glassfish console and create a special connnection pool and resource. The connection pool had to be res-type="java.sql.Driver" (not datasource). Then when creating my entity, I picked this new resource from the list and it worked, my tables showed up.