Hitachi Vantara Pentaho Community Forums
Results 1 to 4 of 4

Thread: configuration via command line

  1. #1

    Default configuration via command line

    Hi all!

    We just set up an application server (or, actually, a service provider did this for us) and the only way I can connect is via SSH and PuTTY (or SFTP). So, I downloaded Kettle to my local machine, transferred the file to the server, connected through PuTTY, unzipped it and ran the "chmod +x *.sh" command. So far, so good.
    Then, I have some experience from a previous assignment in how to use Kitchen and Pan on the command line, so no problem there either. But there's still some steps in between - configuration. Is there any way for me to create a file-repository on the server using the command line? How can I set up (or change) DB-connections?
    If all that doesn't work - how can I run Spoon locally and connect to the server installation?

    I'm sorry if this is documented somewhere, I just couldn't find it. Feels like all my googling just got me to pages explaining how to use Kitchen and Pan from the command line.

    Any help is much appreciated!
    Greetings!
    D

  2. #2
    Join Date
    Jun 2012
    Posts
    5,534

    Default

    I modify my local repositories.xml and transfer it to the server.
    The same with shared.xml for database connections.
    My DB connections are fully parameterized, each parameter set is a properties file containing names qualified by the connection identifier.
    The first transformation in my job reads the necessary properties files to set up corresponding variables.
    Properties files live on the server in a directory like /config/${STAGE}, so a job picks up the right properties files by evaluating Kettle variable ${STAGE}.
    This way I never have to change DB connection names.

    PS: If you want you can use -param:STAGE=prod on then command line instead of a Kettle enviroment variable - just make STAGE a top level job parameter.
    Last edited by marabu; 05-16-2014 at 08:31 AM.
    So long, and thanks for all the fish.

  3. #3

    Default

    Great, thanks marabu! I'll try this directly Monday morning! :-)

  4. #4

    Default

    Sorry to bother you again but I don't get this working.

    There's a user, say "my_user", with home directory "/home/my_user". In this directory, I got Kettle (in a folder named "data-integration"), the folder for the repository (called "kettle_repository" and simply copied over in its entirety from my local machine) and even .kettle is there. In the latter, I placed "<base_directory>/home/my_user/kettle_repository</base_directory>" that is supposed to point to the repository.
    So far, so good. In my interpretation when invoking Kitchen/Pan, it looks for .kettle, reads out the directory and looks there for the file. But this is what I get:
    Code:
    [my_user@dev-as1 data-integration]$ ./kitchen.sh --file=j_mainframe_main -rep=MyRepository
    2014/05/19 15:09:38 - Kitchen - Start of run.
    2014/05/19 15:09:38 - RepositoriesMeta - Reading repositories XML file: /home/my_user/.kettle/repositories.xml
    ERROR: Kitchen can't continue because the job couldn't be loaded.
    [my_user@dev-as1 data-integration]$
    So it gets as far as finding repositories.xml but then it doesn't find the job. Don't know if it came to the right folder. Note that the "Reading repositories XML file" log-line is missing if I don't specify -rep=MyRepository, so that got me a step further (I guess).
    Oh, how I love the administrative part of this job...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.