Docs: added migration guide #1558
Conversation
k8s-ci-robot
added
kind/documentation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: capri-xiyue The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
✅ Deploy Preview for gateway-api-inference-extension ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
k8s-ci-robot
added
the
size/L
Signed-off-by: Xiyue Yu <[email protected]>
|
/assign @kfswain |
| If you are not using Helm, you will need to manually delete all resources associated with your `v1alpha2` deployment. The key is to remove the `HTTPRoute`'s reference to the old `InferencePool` and then delete the `v1alpha2` resources themselves. | ||
|
|
||
| 1. **Update or Delete the `HTTPRoute`**: Modify the `HTTPRoute` to remove the `backendRef` that points to the `v1alpha2` `InferencePool`. | ||
| 2. **Delete the `InferencePool` and associated resources**: You must delete the `v1alpha2` `InferencePool`, any `InferenceModel` resources that point to it, and the corresponding Endpoint Picker (EPP) Deployment and Service. |
There was a problem hiding this comment.
| 2. **Delete the `InferencePool` and associated resources**: You must delete the `v1alpha2` `InferencePool`, any `InferenceModel` resources that point to it, and the corresponding Endpoint Picker (EPP) Deployment and Service. | |
| 2. **Delete the `InferencePool` and associated resources**: You must delete the `v1alpha2` `InferencePool`, any `InferenceModel` (or 'InferenceObjective') resources that point to it, and the corresponding Endpoint Picker (EPP) Deployment and Service. |
| 2. **Delete the `InferencePool` and associated resources**: You must delete the `v1alpha2` `InferencePool`, any `InferenceModel` resources that point to it, and the corresponding Endpoint Picker (EPP) Deployment and Service. | ||
| 3. **Delete the `v1alpha2` CRDs**: Once all `v1alpha2` custom resources are deleted, you can remove the CRD definitions from your cluster. | ||
| ```bash | ||
| kubectl delete -f https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/download/v0.3.0/manifests.yaml |
There was a problem hiding this comment.
Consider making the version portion of the path configurable
| Curl the endpoint to make sure you are getting a successful response with a **200** response code. | ||
|
|
||
| ```bash | ||
| IP=$(kubectl get gateway/inference-gateway -o jsonpath='{.status.addresses[0].value}') |
There was a problem hiding this comment.
I think leaving the GW name as inference-gateway is fine in this case, but I would make mention that you need to put your GW name here
| After cleaning up the old resources, you can proceed with a fresh installation of the `v1` Inference Gateway. This involves installing the new `v1` CRDs, creating a new `v1` `InferencePool` and corresponding `InferenceObjective` resources, and creating a new `HTTPRoute` that directs traffic to your new `v1` `InferencePool`. | ||
|
|
||
|
|
||
| ### 3. Verify the Deployment |
There was a problem hiding this comment.
This should probably include mention of the fact that you need to deploy a new EPP image that is compatible with the v1 API
|
Added comments, the split seems fine, but we need to add that deploying a new EPP is likely needed. But this is a great addition, thank you! |
What type of PR is this?
/kind documentation
What this PR does / why we need it:
Added migration guide for Inference Pool v1 apo
Which issue(s) this PR fixes:
Fixes #
Does this PR introduce a user-facing change?: