Agenta-AI
Create a HELM chart for kubernetes deployment
Is your feature request related to a problem? Please describe. A distributed and horizontally scalable version of this application brings a lot of interest into the project and running it mainstream. Creating a HELM chart as first step and then make it horizontally scalable could proceed after. This should also give a way to export all configurations / mount as volume and use it for next deployment because these all should be / act as stateless services.
Describe the solution you'd like A bare bone version HELM chart that can deploy these services as is to a kubernetes cluster should be good.
Describe alternatives you've considered Creating a own helm chart or use docker compose but this wouldn't give us consistency on restarts.
Additional context Features requested in HELM chart are
- Ingress Settings
- SSL options with ingress
- Supporting K8s volumes
- Horizontally scalable api services
- Reader Nodes for Postgresql
- Integration or posting extra metrics for Prometheus about prompts endpoints
Hi @inventivepotter,
We'd love to have helm charts and k8 deployment. We just don't have the resources to work on it at the moment, but we'd welcome any contributions toward this and can provide support from our side. Let me know if you're interested and we can sync up.
@adarshsingh2 I've created a draft PR a while ago I am sharing here:
https://github.com/Agenta-AI/agenta/pull/2775
I tested it in my local machine and it worked. But it's far from ready.
You're welcome to test it, contribute to improve it.
If you do, would love some feedback.
Thanks for sharing the draft PR @mmabrouk 🙌 I’ll pull it down and test it in my Kubernetes setup. Will share feedback on any gaps or improvements I notice, and can also contribute changes if I get things working smoothly.