What is Dataiku Data Science Studio
Data Science Studio (DSS) is a software platform that aggregates all the steps and big data tools necessary to get from raw data to production ready applications.
It shortens the load-prepare-test-deploy cycles required to create data driven applications.
Thanks to its visual and interactive workspace, it is accessible to both Data Scientists and Business Analysts.
How to use this image
Upon startup, this image configures Data Science Studio to use port 10000 and
data directory /home/dataiku/dss
By default, this directory is initially empty, so Data Science Studio starts as
a fresh install.
It is however possible to mount an existing directory on this path, to have Data Science
Studio run off it. In this case, the initial startup triggers an initial install or an update
of this directory, depending whether it is initially empty or not.
Start a DSS instance on the default internal data directory, exposing DSS on the default port (10000)
docker run -p 10000:10000 -d dataiku/dss
Alternatively, you can start a DSS instance on a mounted data directory, exposing DSS on the default port (10000).
docker run -p 10000:10000 -v MYDIR:/home/dataiku/dss -d dataiku/dss
Open DSS in your browser (Chrome or Firefox)
For Linux, open
For Mac and Windows, you should connect to
http://ADDRESS:10000 where ADDRESS is the IP address of the VM. Find the IP address with the commands:
docker-machine ls docker-machine ip <Name>
DataIcu has version 4.0.7 - why in docker hub there is 4.0.2
My docker+dss container did not start correctly.
Here is an extract of the logs:
- Installation complete
- Next, start Data Science Studio using:
- '/home/dataiku/dss/bin/dss start'
- Dataiku Data Science Studio will be accessible on http://<SERVER ADDRESS>:10000
- You can configure Dataiku Data Science Studio to start automatically at server boot with:
- sudo -i "/home/dataiku/dataiku-dss-2.1.4/scripts/install/install-boot.sh" "/home/dataiku/dss" dataiku
[+] Checking dependencies
- Detected OS distribution : debian 8
- Checking required packages...
[+] Installing required R packages
[+] Installing R kernel for Jupyter
[InstallKernelSpec] Installed kernelspec ir in /home/dataiku/dss/jupyter-run/jupyter/kernels/ir
Error: Cannot open an HTTP server: socket.error reported errno.EPERM (1)
For help, use /home/dataiku/dataiku-dss-2.1.4/python.packages/supervisor/supervisord.py -h
Can you help me?
Do you have the last version of DSS? Is it with your email that it does not work? On the first install, you specified a Dataiku account (email/password) that is required to get a license. If you don't remember your password, please get in touch with our support team. I hope that helps.
Jeremy from Dataiku
After closing the Docker and starting it up again if I say I'm an existing user and give default admin/admin credentials it will say that Login Failed.
If I start over the registration it will say that the email has already been used.
What should I do?
Thanks for reaching out and sorry about the delay for the answer. Your problem should have been fixed. If you have a current issue with your licence, just delete the license file (
$DATA_DIR/config/license.json) and restart DSS.
If the problem persists, please open a ticket on support.dataiku.com.
Regards, Jeremy from Dataiku.
When I run the docker instance it works like a charm. However when I stop then start it again the web shows a licensing issue (even though it is a community edition).
"This copy of Data Science Studio is not properly licensed : Your license is not authorized on this node.
Please ensure your license file is properly configured or contact Dataiku to obtain a new license"
you are indeed correct. If starting the Docker process did not show an error message, the Data Science Studio should be up and running and accessible on http port 10000 of the Docker host.
You need of course to make sure you have launched the container with "-p 10000:10000" to make sure the container port has been mapped on the host (or -p MYPORT:10000 if you want to expose DSS on another port on the host).
Another issue comes to mind: if you are running Docker on OS X or Windows, the Docker host is not localhost but a VM running inside it (using boot2docker). You should then connect to http://ADDRESS:10000 where ADDRESS is the IP address of the VM, accessible through "boot2docker ip".
In any case, the best way to get help on a problem like this would be to open a ticket on support.dataiku.com. This would in particular ensure it is timely answered, and tracked until solved.
I'm very interested in experiencing your Data Science Studio. I followed your installation process which is - well - very straightforward and naively believed that I would simply need to go to http://localhost:10000 to open the Admin GUI. The docker process is up and running but unfortunately I can't open the application. Could you please tell me what I did wrong ?