Hey, is there any documentation or guide on how to deploy the meltano project to kubernetes (besides the Deployment in Production which did not help much)? Also, how would such a deployment react regarding new versions of the tap/targets? For example: I push my latest tap to gitlab. To use is I would need to install it first in meltano. How is this handled in k8s? Currently I do notuse meltano in prod. I simply have singer taps which have a git ci/cd to be built as docker image. Then I use fleet in k8s to get the latest changed in my gitlab repo and redeploy the pods on every job run (using CronJobs).

Hey @michel_ebner 👋 I am working on deploying Meltano (for Meltano) on EKS at the moment, with the intent to provide it as a reference to the community as well as to help inform future integrations with tools like helm and terraform in Meltano 🙂 It should be finished in the next week or so, but the WIP branch is here. The approach I plan to take (at least initially) is to build images for Meltano and Airflow during CI/CD tagged with a commit ID, and then update both Meltano and Airflow helm chart values to point to the new tag.

meltano install

will happen as part of building the Docker image (check out the example Docker image here), so any changes made locally and then pushed/merged to a CI/CD-enabled remote branch will cause a new Docker build with the updated

meltano.yaml

config (in your case pulling the latest tap version) 👍 Hope this helps, and I will demoing what I have as soon as it is ready, so keep an eye on the Demo Day schedule 🙂

Hi @ken_payne 👋 I saw some threads about it and an open issue but did not know where it stands. Thanks for the update and your work 💪 !! Thrilled to test it 😄 I tried some quick approach (same idea as yours), but it is not clean and I had "issues" with the secrets/.envs and git credentials. I will clear my schedule as soon as I see the demo schedule 😉