bobbai00 opened a new pull request, #5272:
URL: https://github.com/apache/texera/pull/5272

   ### What changes were proposed in this PR?
   
   The agent-service image is built/shipped and runs under single-node compose, 
but the Helm chart (`bin/k8s/`) had no deployment for it. This adds:
   
   - `bin/k8s/templates/agent-service-deployment.yaml` — deployment mirroring 
the other Texera services. Backend endpoints are wired to in-cluster service 
DNS (dashboard `webserver-svc:8080`, compiling 
`workflow-compiling-service-svc:9090`, LLM gateway 
`<release>-access-control-service-svc:9096`, execution via the computing-unit 
pool service). Adds readiness/liveness probes on `/api/healthcheck`, since the 
service fetches operator metadata from the dashboard service at startup and 
should not receive gateway traffic until ready.
   - `bin/k8s/templates/agent-service-service.yaml` — `agent-service-svc` 
(ClusterIP, port 3001).
   - `agentService` values block in `values.yaml` and `values-development.yaml`.
   - An `/api/agents` rule in `gateway-routes.yaml` (covers REST and the 
`/api/agents/:id/react` WebSocket), placed ahead of the `/api` catch-all.
   
   Full in-cluster LLM proxy (LiteLLM) wiring is out of scope and tracked 
separately by #4108; `LLM_ENDPOINT` points at the access-control-service, 
mirroring the single-node nginx routing of `/api/chat` and `/api/models`.
   
   ### Any related issues, documentation, discussions?
   
   Closes #5269
   
   ### How was this PR tested?
   
   `helm lint` and `helm template` against the chart (subchart `dependencies:` 
stripped locally so the render needs no remote charts):
   
   ```
   helm lint .                       # 1 chart(s) linted, 0 chart(s) failed
   helm template texera . -f values-development.yaml   # RC=0, 44 documents, no 
errors
   ```
   
   Verified the rendered deployment (image, port 3001, env wiring, probes), 
service (`agent-service-svc:3001`), and the `/api/agents -> 
agent-service-svc:3001` gateway rule ordered before the `/api` catch-all.
   
   ### Was this PR authored or co-authored using generative AI tooling?
   
   Generated-by: Claude Opus 4.8 (1M context)


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to