How to add an endpoint without using the GUI

[PC-1]

We would like to automate the adding of endpoints to Qlik Replicate without having to use the GUI. The automation is required to be non-interactive since it will run from a batch scheduler. I am aware that tasks can be imported outside of the GUI (via the JSON files) but cannot see how to create endpoints via an import.

Is it possible to do this ?

Thanks

PC

[SwathiPulagam]

Hi @PC-1 ,

There is no API call to create the endpoint so you have to submit feature requests or ideas in the below forum: 

https://community.qlik.com/t5/Suggest-an-Idea/idb-p/qlik-ideas

Thanks,

Swathi

[PGN · Accepted solution]

Whenever I import tasks outside the GUI, the endpoints as defined in the task are automatically created (or overwritten if they exist).  Not sure if that answers your question.

[PC-1]

The endpoints only appear in the task if they already exist. So I need to create those before creating the task. I'm not aware that you can create a task without having any endpoints. If you can then this would help.

[PGN · Accepted solution]

I guess I'm not following what your process is.  In the .json file, you can create tags for the endpoints.  When you export a .json file from the server using the GUI or any other method, it creates the entire set of tags needed.  You should be able to use that as a guide to create .json files for use on input.  

[PC-1]

ok thanks , this will help.

[Heinvandenheuvel]

I disagree.

When you look at the structure of a task, or 'export-all' json with  a capable JSON tool (for example Notepad++ ) you will readily see a structure:

"tasks" is an array of tasks. 

"databases" is an array of endpoints.

Either can be omitted or provided.

To create endpoints, just IMPORT a JSON with only the "databases" array, no "tasks" and provided those to be added or replaced. Note: You cannot 'delete' this way, only grow.

Yes you can add (or replace a task) without endpoint. This is critically important functionality when moving tasks between environments  where the task designer may not have or want (plausible deniability) the knowledge or credentials to replace an existing endpoint. It's even more critical when multiple tasks share the same (target) endpoint.

Hein