I have a simple knime flow that reads data from excel and sends the data to Tableau Server in .hyper format.
It runs sucessfully in KNIME Analytics platform ( I have installed the extensions and updated the path ) . So no problems there.
But the task fails in KNIME Server.
I get this error message
Tableau Writer (Hyper) 3:6 - WARNING: This nodes requires the ‘Tableau Hyper’ backend, but the active backend is: ‘Tableau TDE’ The Tableau backend can be configured in the Tableau preference page.
Send to Tableau Server (Hyper) 3:7 - WARNING: This nodes requires the ‘Tableau Hyper’ backend, but the active backend is: ‘Tableau TDE’ The Tableau backend can be configured in the Tableau preference page.
How can i update the KNIME server backend?
(Note; Both KNIME analytics platform and KNIME server are running in the same machine)
simply change the preferences of your executor. To do that, you have to start the KNIME Analaytics Platform that serves as your executor and change the preferences like you do in your local instance. You can also use that to add new extensions to the executor so that they are available for execution.
Depending on the changes you make have to restart KNIME Server in case there’s already an executor in the background running.
They are ( KNIME analytics, server) both local instances. Running on the same machine… I have changed the preferences of the Analytics platform which I use as the executor and it runs fine, I can create .hyper and publish to Tableau Server. There are no other executors.The flow fails in the KNIME Server. And I did restart the server multiple times.
do you use your KNIME Analytics Platform also as your KNIME Server Executor? You can simply check that if you have a look at the knime-server.config file stored at
-PATH-TO-SERVER-/workflow_repository/config
There you might use another KNIME Analytics Platform instance.
Also have a look at the documentation: https://docs.knime.com/2018-12/server_admin_guide/index.html#preferences-file
yes I do use the Analytics Platform as the executor. I installed the extension. I believe it is one of the options.
I don’t want to use another Analytics Platform. Just running few integration tests between KNIME server and Tableau Server. Want to keep it as simple as it can be.
I followed the instructions on that documentation you sent. I exported the preferences.epf file and pasted under …workflow_repository\config but still no joy. Getting the same error message.
one issue could be, that a restart of the server didn’t yield a restart of the executor. Thus, check if there are any running java-processes after the server has been shut down, or simply restart the PC. Furthermore, ensure that the preferences.epf file has the correct name and the right permissions.
Another possible workaround would be to remove the KNIME Tableau Integration - TDE Format from the executor. This way the option shouldn’t be available and ‘Tableau Hyper’ should be used as a default.