API node not configured
Hello, I tried to deploy an API service in the Enterprise trial version. I created my infrastructure with details as shown below. How do I configure a static API node? Without this, I am not able to deploy an endpoint.
Answers
-
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,225 Dataiker
Hi @RohitRanga
,To install static API node: https://doc.dataiku.com/dss/latest/installation/custom/api-node.html
You will have a license Business or Enterprise license. https://www.dataiku.com/product/plans-and-features/
The free 14-day enterprise trial will not work for API Nodes.
If you are in touch with the Dataiku Sales team they can help with an appropriate license.
Thanks,
-
Thanks @AlexT
! Just to confirm, I would not be able to install an API node, deploy an end-point and use Test queries with the Enterprise trial, is that correct? -
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,225 Dataiker
That is correct, free 14-day enterprise trial can't be used to install an API node.
You can still use API Designer and test queries locally within API Designer on DSS with the Enterprise trial:
-
Oh, even though I was able to deploy the endpoint, it gives me a warning that no API nodes are configured. So I am not able to test queries.
-
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,225 Dataiker
You can test before deploying in the api designer under test queries.
-
I am getting this error (refer attachment)
-
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,225 Dataiker
Can you share your code?
-
Sure, here it is. I am just using the default code.
Things not clear:
- predict function asks for dataframe, but how to send a dataframe using test queries?
- any processing that needs to be done on the input, (for eg., getting embeddings from text) needs to be handled inside the predict function only?
- from the error, it seems like there is a failure with the development server deployment
-
@AlexT
It is now working! The only error I am facing now is when I try to access files (in local machine and managed folder) the same way I used to access inside the flow recipes. -
Alexandru Dataiker, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 1,225 Dataiker
That’s expected the same methods will not work you have use the api as if the endpoint Is external And provide an api key :
-
What is the difference between the project key and API key? I was able to create a project key from Project > Security > API key
-
Does API key refer to the personal API key? I found it inside User Profile and Settings