Hi there, I want to enable a plugin in datahub docker quick start. Can someone tell me how can i do that. Also which container should i login to install the plugin's. Does datahub cli provide that functionality.

Hi Shiva! Check out this part our Quickstart guide for how you can customize your installation

Hey there, I'm looking for the same thing. Unfortunately, I didn't find any meaningful information regarding installing plugins like 'acryl-datahub[s3]'. I'm cooking a kubernetes deployment via Helm Chart, so ideally there would be YAML value that can be overwritten?

If you looking to install Datahub ingest integration plugins like ‘acryl-datahub[s3]’, you have to install these plugins on the client side, which means install where you are running your ingest recipes or code. these plugins not not required installed on service containers

I don't understand. Assuming that I have a SAP Hana instance, were I have tables, views, etc. and a Helm Chart deployment (with all those containers) of Datahub. I want to connect my Datahub to my SAP Hana database to get all the data lineage information. In theory, Datahub can have SAP Hana as source, but it not offered on the web frontend. Therefore, I assumed, that it has to be installed as additional plugin. Where would I now need to install the source plugin? Certainly not on the database machine, right?

if you are running Datahub quickstart on your laptop or PC, you install these ingest plugins to your laptop or PC(better with Python virtual environment) or build Python docker container, install these plugins and run ingest recipes

@gray-ocean-32209, thank you for your answer. But, like I said, I am not running the quickstart setup, but the production ready Helm Chart: https://github.com/acryldata/datahub-helm

It is better to have a server(jump host) that has a connection allowed to SAP Hana database for ingest purposes. Another approach is to build a docker image with an ingest plugin installed and deploy it as Kubernetes jobs

Hello @gray-ocean-32209, don't get me wrong, I'm very thankful that you are trying to help, but I have the impression, that I can't seem to make my point. I'll try again: I'm using the official Helm chart as provided by Acryl (https://github.com/acryldata/datahub-helm), which deploys a couple of containers, and which I expect to be functional. They all spawn successfully. So, no problem here. Then I want to establish SAP Hana as a source ingested by Datahub. The Datahub documentation states, that I ought to "pip install" the respective package (https://datahubproject.io/docs/generated/ingestion/sources/hana/) without mentioning, in which container. Additionally, this kind of manual intervention, i.e. going into the container and manually installing something, is an anti pattern in the context of containerized productive solutions. Therefore, I'm looking for how to adjust the aforementioned Helm chart so that the missing packages are installed upon provisioning of the respective containers. @little-megabyte-1074, can you maybe help here or have one of your developer colleagues help out? Again, I'm much obliged for any help I can get.