Configuring Amazon Authentication Node for Both Local and Server Use

[gcincilla]

Hello KNIME Community,

I am in the process of configuring the Amazon Authentication node and have encountered a challenge that I hope someone here can help me with. I’ve managed to successfully set up the node for use on the KNIME server by enabling the “Switch Role” option and using the “Default Credentials Provider Chain” specifically for AWS Athena access. The account and role details have been set as per the requirements.

However, I’m having trouble when it comes to setting up the Default Credential Provider Chain for local use with SSO connections. The same configuration I used above is not valid when I run the workflow locally (after authenticating with SSO). This has led to the need for two separate configurations for AWS access - one for local use and another for server use.

I would greatly appreciate any guidance on how to configure the Amazon Authentication node to work both locally and on the KNIME executor. Any advice on how to effectively use the Default Credential Provider Chain in our setup would also be very helpful.

Thank you in advance.

Best,

Gio

[schulz_d]

Hi Gio,

I don’t have a quick fix how to configure the Amazon Authentication node itself that its works for both but what you could do as a workaround is to create a component / metanode with two configurated nodes, one for local one and one for server use, that automatically detects with the context properties node if a workflow is run on a server or not.

We normally use the context.job.id variable to detect this. In case it is run on a server, the value is filled if the workflows runs on a server and not filled if run locally.

Hope I could help!

[gcincilla]

Hi Schulz_d,
Thanks for your quick reply! I’ve thought to something like that but, being not familiar with the Extract Context Properties node, I didn’t know which variable I should use. Using the context.job.id as you suggested, it seems to work.
Thank you very much for your help.
Gio