English

Upgrade from AI 1.3

Before Upgrading

Keep Inference Service Runtimes

This step ensures that older ClusterServingRuntime resources are not inadvertently removed during the upgrade process. The script adds the "helm.sh/resource-policy": "keep" annotation to all ClusterServingRuntime resources, which instructs Helm to preserve them during the upgrade.

Execute the following script in the destination cluster:

migrate-clusterservingruntime-resources.sh

#!/bin/bash
set -e

# Check if kubectl is installed
if ! command -v kubectl &> /dev/null; then
    echo "Error: kubectl not found. Please install kubectl and ensure it is in your PATH."
    exit 1
fi

echo "Fetching all ClusterServingRuntime resources and checking for annotations..."

kubectl get clusterservingruntime --no-headers -o name | while read -r name; do

    if kubectl get "$name" -o yaml | grep -q 'helm.sh/resource-policy: keep'; then
        echo "ClusterServingRuntime $name already has the 'helm.sh/resource-policy: keep' annotation. Skipping."
    else
        echo "ClusterServingRuntime $name does not have the 'helm.sh/resource-policy: keep' annotation. Patching..."

        kubectl patch "$name" \
            --type=merge \
            -p '{"metadata":{"annotations":{"helm.sh/resource-policy": "keep"}}}'

        echo "Successfully patched $name."
    fi
done

echo "migrate-clusterservingruntime-resources completed."

Migrate InferenceService resources

Before Alauda AI 1.4, InferenceService resources did not explicitly specify their deployment mode. You must now explicitly define the deployment mode for all your inference services before upgrading.

WARNING

This migration updates your inference services. In Knative Serverless mode it triggers a rolling update: a new revision is created and the old one is deleted only after the new one is Ready.

If the service may be resource‑constrained (GPU/CPU/memory), schedule a maintenance window or temporarily stop the service before running the script.

Execute following script in destination cluster:

migrate-inferenceservice-resources.sh

#!/bin/bash

set -e

kubectl get inferenceservice --all-namespaces -o custom-columns=NAMESPACE:.metadata.namespace,NAME:.metadata.name --no-headers | while read -r namespace name; do

    if kubectl get inferenceservice "$name" -n "$namespace" -o yaml | grep -q 'serving.kserve.io/deploymentMode:' ; then
        echo "InferenceService ${namespace}/${name} already has 'serving.kserve.io/deploymentMode' annotation. Skipping."
    else
        echo "InferenceService ${namespace}/${name} does not have 'serving.kserve.io/deploymentMode' annotation. Patching..."

        kubectl patch inferenceservice "$name" -n "$namespace" --type=merge -p '{"metadata":{"annotations":{"serving.kserve.io/deploymentMode": "Serverless"}}}'

        echo "Successfully patched ${namespace}/${name}."
    fi
done

echo "migrate-inferenceservice-resources completed."

Install Alauda AI Cluster Components

Please visit Alauda AI Cluster for:

WARNING

Please ignore Creating Alauda AI Cluster Instance since we are upgrading Alauda AI from a previously managed version.

Downloading operator bundle packages for Alauda AI Cluster and KServeless.
Uploading operator bundle packages to the destination cluster.
To upgrade, follow the process described below.

Upgrading

The following procedure describes how to upgrade from AI 1.3 to Alauda AI 1.4.

Upgrade Alauda AI Essentials

Alauda AI Essentials will be upgraded together with the global cluster's functional components.

In the Administrator view, navigate to Clusters / Cluster in the left navigation, find the global cluster in the right panel, and click the Upgrade button on the Functional Components tab page to trigger the upgrade.

INFO

All components that require upgrading, including Alauda AI Essentials, will be upgraded together and cannot be upgraded individually.

After the upgrade is complete, please confirm that the status of Alauda AI Essentials in the list is Running and the version is v1.4.x.

Upgrading Alauda AI Operators

The procedure for upgrading both operators is nearly identical, with only the target component being different.

Step	Alauda AI Operator	Alauda AI Model Serving Operator
1. Navigate	Log into the Web Console, then go to Marketplace > OperatorHub in the Administrator view.	Log into the Web Console, then go to Marketplace > OperatorHub in the Administrator view.
2. Select	Select your target cluster.	Select your target cluster.
3. Click	Click the Alauda AI card.	Click the Alauda AI Model Serving card.
4. Confirm	Click Confirm on the upgrade prompt.	Click Confirm on the upgrade prompt.

INFO

Once the new version is uploaded and recognized by the platform, an upgrade prompt will appear at the top of the operator's page.

Verification

Alauda AI

Check the status field from the AmlCluster resource which named default:

kubectl get amlcluster default

Should returns Ready:

NAME      READY   REASON
default   True    Succeeded

Alauda AI Model Serving

Check the status field from the KnativeServing resource which named default-knative-serving:

kubectl get KnativeServing.components.aml.dev default-knative-serving

Should returns InstallSuccessful:

NAME                      DEPLOYED   REASON
default-knative-serving   True       UpgradeSuccessful

Upgrade from AI 1.3

Before Upgrading

Keep Inference Service Runtimes

Execute the following script in the destination cluster:

migrate-clusterservingruntime-resources.sh

#!/bin/bash
set -e

# Check if kubectl is installed
if ! command -v kubectl &> /dev/null; then
    echo "Error: kubectl not found. Please install kubectl and ensure it is in your PATH."
    exit 1
fi

echo "Fetching all ClusterServingRuntime resources and checking for annotations..."

kubectl get clusterservingruntime --no-headers -o name | while read -r name; do

    if kubectl get "$name" -o yaml | grep -q 'helm.sh/resource-policy: keep'; then
        echo "ClusterServingRuntime $name already has the 'helm.sh/resource-policy: keep' annotation. Skipping."
    else
        echo "ClusterServingRuntime $name does not have the 'helm.sh/resource-policy: keep' annotation. Patching..."

        kubectl patch "$name" \
            --type=merge \
            -p '{"metadata":{"annotations":{"helm.sh/resource-policy": "keep"}}}'

        echo "Successfully patched $name."
    fi
done

echo "migrate-clusterservingruntime-resources completed."

Migrate InferenceService resources

WARNING

This migration updates your inference services. In Knative Serverless mode it triggers a rolling update: a new revision is created and the old one is deleted only after the new one is Ready.

If the service may be resource‑constrained (GPU/CPU/memory), schedule a maintenance window or temporarily stop the service before running the script.

Execute following script in destination cluster:

migrate-inferenceservice-resources.sh

#!/bin/bash

set -e

kubectl get inferenceservice --all-namespaces -o custom-columns=NAMESPACE:.metadata.namespace,NAME:.metadata.name --no-headers | while read -r namespace name; do

    if kubectl get inferenceservice "$name" -n "$namespace" -o yaml | grep -q 'serving.kserve.io/deploymentMode:' ; then
        echo "InferenceService ${namespace}/${name} already has 'serving.kserve.io/deploymentMode' annotation. Skipping."
    else
        echo "InferenceService ${namespace}/${name} does not have 'serving.kserve.io/deploymentMode' annotation. Patching..."

        kubectl patch inferenceservice "$name" -n "$namespace" --type=merge -p '{"metadata":{"annotations":{"serving.kserve.io/deploymentMode": "Serverless"}}}'

        echo "Successfully patched ${namespace}/${name}."
    fi
done

echo "migrate-inferenceservice-resources completed."

Install Alauda AI Cluster Components

Please visit Alauda AI Cluster for:

WARNING

Please ignore Creating Alauda AI Cluster Instance since we are upgrading Alauda AI from a previously managed version.

Downloading operator bundle packages for Alauda AI Cluster and KServeless.
Uploading operator bundle packages to the destination cluster.
To upgrade, follow the process described below.

Upgrading

The following procedure describes how to upgrade from AI 1.3 to Alauda AI 1.4.

Upgrade Alauda AI Essentials

Alauda AI Essentials will be upgraded together with the global cluster's functional components.

INFO

All components that require upgrading, including Alauda AI Essentials, will be upgraded together and cannot be upgraded individually.

After the upgrade is complete, please confirm that the status of Alauda AI Essentials in the list is Running and the version is v1.4.x.

Upgrading Alauda AI Operators

The procedure for upgrading both operators is nearly identical, with only the target component being different.

Step	Alauda AI Operator	Alauda AI Model Serving Operator
1. Navigate	Log into the Web Console, then go to Marketplace > OperatorHub in the Administrator view.	Log into the Web Console, then go to Marketplace > OperatorHub in the Administrator view.
2. Select	Select your target cluster.	Select your target cluster.
3. Click	Click the Alauda AI card.	Click the Alauda AI Model Serving card.
4. Confirm	Click Confirm on the upgrade prompt.	Click Confirm on the upgrade prompt.

INFO

Once the new version is uploaded and recognized by the platform, an upgrade prompt will appear at the top of the operator's page.

Verification

Alauda AI

Check the status field from the AmlCluster resource which named default:

kubectl get amlcluster default

Should returns Ready:

NAME      READY   REASON
default   True    Succeeded

Alauda AI Model Serving

Check the status field from the KnativeServing resource which named default-knative-serving:

kubectl get KnativeServing.components.aml.dev default-knative-serving

Should returns InstallSuccessful:

NAME                      DEPLOYED   REASON
default-knative-serving   True       UpgradeSuccessful

Guides

Guides

How To

Troubleshooting

Guides

Guides

Guides

Inference Service APIs

Workbench APIs

Manage APIs

Operator APIs

Upgrade from AI 1.3

TOC

Before Upgrading

Keep Inference Service Runtimes

Migrate InferenceService resources

Install Alauda AI Cluster Components

Upgrading

Upgrade Alauda AI Essentials

Upgrading Alauda AI Operators

Verification

Alauda AI

Alauda AI Model Serving

Upgrade from AI 1.3

TOC

Before Upgrading

Keep Inference Service Runtimes

Migrate InferenceService resources

Install Alauda AI Cluster Components

Upgrading

Upgrade Alauda AI Essentials

Upgrading Alauda AI Operators

Verification

Alauda AI

Alauda AI Model Serving

Guides

Guides

How To

Troubleshooting

Guides

Guides

Guides

Inference Service APIs

Workbench APIs

Manage APIs

Operator APIs

#Upgrade from AI 1.3

#TOC

#Before Upgrading

#Keep Inference Service Runtimes

#Migrate InferenceService resources

#Install Alauda AI Cluster Components

#Upgrading

#Upgrade Alauda AI Essentials

#Upgrading Alauda AI Operators

#Verification

#Alauda AI

#Alauda AI Model Serving

#Upgrade from AI 1.3

#TOC

#Before Upgrading

#Keep Inference Service Runtimes

#Migrate InferenceService resources

#Install Alauda AI Cluster Components

#Upgrading

#Upgrade Alauda AI Essentials

#Upgrading Alauda AI Operators

#Verification

#Alauda AI

#Alauda AI Model Serving

Upgrade from AI 1.3

TOC

Before Upgrading

Keep Inference Service Runtimes

Migrate InferenceService resources

Install Alauda AI Cluster Components

Upgrading

Upgrade Alauda AI Essentials

Upgrading Alauda AI Operators

Verification

Alauda AI

Alauda AI Model Serving

Upgrade from AI 1.3

TOC

Before Upgrading

Keep Inference Service Runtimes

Migrate InferenceService resources

Install Alauda AI Cluster Components

Upgrading

Upgrade Alauda AI Essentials

Upgrading Alauda AI Operators

Verification

Alauda AI

Alauda AI Model Serving