commiting changes

    This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

    • commiting changes

      good day!

      how to commit changes made in cube correctly? I'm running process which loads data into the cube, i could see them in corresponding .cube.log and via Excel f.e. ... but as i understand, data loaded resides in memory an corresponding .cube.data file is not affected. I've tried both

      Database::save()
      Connection()::save()

      to commit changes to the FS, but the only way to make it as i see - to use 'Save Database' menu item in Palo Service Manager UI (after that i see log files cleaned and data files growing)... that's not a good option for me :)

      please answer - should methods mentioned cause the same behavior? Or - are there any other reliable methods?

      Thanks,
      Axel
    • RE: commiting changes

      hi axel,

      I do not quite understand what your problem is, but in our application any changes, e.g. of a cell value, are directly passed to the palo server. When the application is closing we simply disconnect from the server. Maybe you should try the Cube#commitLog() method? However, we do not use it.

      regards,

      arnd
    • RE: commiting changes

      There is no problem - i do understand that changes are going into cube directly and you can see them immediately. But changes made are in memory and in .cube.log files, not in .cube.data files; so in case of server failure you'll need to wait while they will be rolled up... i've just wanted to make process of snap cube changes to disk more predictable :) Thanks, i've missed Cube::commitLog method - it does exactly i have wanted...

      Thanks,
      Axel