Hi, I'd like to deploy my model via Yatai, however, I faced an error which may be due to the different version of serving API (ai.yatai.serving.v2alpha1). Do you happen to know any idea to find out available versions or fix this error? Versions • Yatai: 1.1.10 • BentoML: 1.0.18 • minikube: 1.30.1 • helm: 3.11.3

$ kubectl apply -f my_deployment.yaml 
<http://bentorequest.resources.yatai.ai/iris-classifier|bentorequest.resources.yatai.ai/iris-classifier> unchanged
error: error validating "my_deployment.yaml": error validating data: ValidationError(BentoDeployment.spec.runners[0].resources): unknown field "autoscaling" in ai.yatai.serving.v2alpha1.BentoDeployment.spec.runners.resources; if you choose to ignore these errors, turn validation off with --validate=false


$ kubectl get apiservices | grep autoscaling
v1.autoscaling                         Local                        True        46m
v2.autoscaling                         Local                        True        46m

Is it possible to upgrade yatai and yatai-deployment in your cluster?

Thank you for answering. I followed the upgrade guideline, and my yatai and yatai-deployment versions are the latest ones...

Oh ic. Let me take a look on the doc. I remember it should be hpa_config rather than autoscaling.

Thanks a lot !!! Now, I'm following this doc verifying each step, but the same error occurs again at the end when running

kubectl -n yatai-deployment logs -f deploy/yatai-deployment

as below

$ kubectl -n yatai-deployment logs -f job/yatai-deployment-default-domain
time="2023-04-14T06:34:15Z" level=info msg="Creating ingress default-domain- to get a ingress IP automatically"
time="2023-04-14T06:34:15Z" level=info msg="Waiting for ingress default-domain-2xp4m to be ready"
time="2023-04-14T06:34:45Z" level=info msg="Ingress default-domain-2xp4m is ready"                                                                                                                                           
time="2023-04-14T06:34:45Z" level=info msg="you have not set the domain-suffix in the network config, so use magic DNS to generate a domain suffix automatically: `***.***.***.**.<http://sslip.io|sslip.io>`, and set it to the network config"


$ kubectl -n yatai-deployment logs -f deploy/yatai-deployment                                                                                                          
Found 2 pods, using pod/yatai-deployment-84b5584ff9-mhwb7
1.6814581142693202e+09  INFO    setup   starting manager        {"bento deployment namespaces": ["yatai"]}                                                                                                                   
1.6814581145232673e+09  INFO    controller-runtime.metrics      Metrics server is starting to listen    {"addr": "127.0.0.1:8080"}
...
...
...
1.6814581299666548e+09  ERROR   controller-runtime.source       if kind is a CRD, it should be installed before calling Start   {"kind": "HorizontalPodAutoscaler.autoscaling", "error": "no matches for kind \"HorizontalPod
Autoscaler\" in version \"autoscaling/v2beta2\""}
<http://sigs.k8s.io/controller-runtime/pkg/source.(*Kind).Start.func1.1|sigs.k8s.io/controller-runtime/pkg/source.(*Kind).Start.func1.1>
        /go/pkg/mod/sigs.k8s.io/controller-runtime@v0.13.0/pkg/source/source.go:139
<http://k8s.io/apimachinery/pkg/util/wait.runConditionWithCrashProtectionWithContext|k8s.io/apimachinery/pkg/util/wait.runConditionWithCrashProtectionWithContext>
...
...
...

It IS autoscaling. The issue is

runners:
        - name: iris_clf
          resources:
              limits:
                  cpu: "1000m"
                  memory: "1Gi"
              requests:
                  cpu: "500m"
                  memory: "512m"
              autoscaling:
                  maxReplicas: 4
                  minReplicas: 1

Oh, I got it. Then the example *yaml file* in the README may be wrong

Ah you are correct.

I'm using this version

$ kubectl version
Client Version: <http://version.Info|version.Info>{Major:"1", Minor:"25", GitVersion:"v1.25.0", GitCommit:"a866cbe2e5bbaa01cfd5e969aa3e033f3282a8a2", GitTreeState:"clean", BuildDate:"2022-08-23T17:44:59Z", GoVersion:"go1.19", Compiler:"gc", Platform:"linux/amd64"}
Kustomize Version: v4.5.7
Server Version: <http://version.Info|version.Info>{Major:"1", Minor:"26", GitVersion:"v1.26.3", GitCommit:"9e644106593f3f4aa98f8a84b23db5fa378900bd", GitTreeState:"clean", BuildDate:"2023-03-15T13:33:12Z", GoVersion:"go1.19.7", Compiler:"gc", Platform:"linux/amd64"}

Sorry I'm AFK for a while. I noticed that you are using 1.26, and yatai doesn't support k8s 1.26 now due to the API changes.

Oh, it's okay! Then, which version should I use?

From 1.20 to 1.24 is verified. 1.25 should also work.

I'm gonna try it right now! Thank you for answering 🙂

That's great. I also opened two PR for the yatai doc.