I have followed all steps mentioned in https://github.com/Azure/feast-azure/tree/main/provider/tutorial and had even changes roleDefinitionID parameter as per previous defect but still I'm getting same error.

I've followed the steps at https://github.com/Azure/feast-azure/tree/main/provider/tutorial but couldnt get table for customer data to load. Did a printf and made sure keyvault secret has a value. I am using the one button deploy method (not k8 cluster).

Screenshot: Error: OperationalError: (pyodbc.OperationalError) ('HYT00', '[HYT00] [Microsoft][ODBC Driver 17 for SQL Server]Login timeout expired (0) (SQLDriverConnect)') (Background on this error at: https://sqlalche.me/e/14/e3q8)

Hey all, Glad to see this initiative. Earlier this year, I tried Feast 0.9 on GCP/GKE : https://paravatha.medium.com/feast-setup-your-own-ml-feature-store-on-kubernetes-5b3193c2b62c

Unfortunately, I don't have access to create Redis. So, If you have any sample notebooks for just the offline store part, I'd like to give it a try.

Adding helm chart and changing script, so feast can be installed from local helm chart that contains the updates rbac apis thats compatible with latest versions of Kubernetes.

This PR solves the issue where the older rbac v1beta apis are not compatible with latest Kubernetes version. This fixes that problem but the chart needs to be installed from local feast-0.9.5-helmchart folder.

Updating to 0.17 with pip install --upgrade feast fails with error message:

ERROR: feast-azure-provider 0.2.2 has requirement feast[redis]==0.15.1, but you'll have feast 0.17.0 which is incompatible.

Tried modifying setup.py in my own fork with deployment template pointing to my own fork that has ==0.17 specified but still error out. own fork: https://github.com/cbtham/feast-azure/blob/main/provider/sdk/setup.py

It seems like the source of feast-azure-provider(URL below) needs to be updated to support 0.17 because the current version 0.2.2 hardcoded ==0.15.1

setup install_requires=[ "feast[redis]==0.15.1"

https://pypi.org/project/feast-azure-provider/

Following the instructions listed in cluster installation on AKS cluster v1.22.4: https://github.com/Azure/feast-azure/tree/main/cluster/setup

Initial installation steps pass, but api versions are not recognized in newer version of AKS (stable API version rbac.authorization.k8s.io/v1 should work instead of rbac.authorization.k8s.io/v1beta1).

Any change to specify a newer version of helm chart in the installation shell script?

Here is the issue which I encounter:

sudo ./installfeast.sh (parameters passed here) INFO: Install feast error: failed to create secret secrets "feast-postgresql" already exists Error: unable to build kubernetes objects from release manifest: [unable to recognize "": no matches for kind "ClusterRole" in version "rbac.authorization.k8s.io/v1beta1", unable to recognize "": no matches for kind "ClusterRoleBinding" in version "rbac.authorization.k8s.io/v1beta1", unable to recognize "": no matches for kind "Role" in version "rbac.authorization.k8s.io/v1beta1", unable to recognize "": no matches for kind "RoleBinding" in version "rbac.authorization.k8s.io/v1beta1"]

In the tutorial - step 2 in notebook 2 "Registre..." fails with missing module. The error is triggered by the command "fs = FeatureStore("./feature_repo")" The Feast evrsion is 0.15.1 and the Azure feast provider is 0.2.1.

Error "snippet" first lines and last line. ModuleNotFoundError Traceback (most recent call last) /anaconda/envs/azureml_py38/lib/python3.8/site-packages/feast/infra/online_stores/redis.py in 43 from redis import Redis ---> 44 from rediscluster import RedisCluster ......... FeastModuleImportError: Could not import RedisOnlineStoreConfig module 'feast.infra.online_stores.redis'

When trying to clone the feast-azure repo, I'm running into a race condition - where the Users directory hasn't yet been mounted to the CI when the cd /home/azureuser/cloudfiles/code/Users portion of the script is run.

Here's the error in the logs...

Successfully installed pymssql-2.2.2
/mnt/batch/tasks/startup/wd/scripts/creation.sh: line 1: cd: /home/azureuser/cloudfiles/code/Users: No such file or directory
Cloning into 'feast-azure'...

Summary: When attempting to create the conda environment "feast-env" as found in the training and deployment notebook, the environment creation stalls out after an hour during "installing pip dependencies" and never completes.

Details: First occurrence of this issue was during attempts to deploy a scoring container, which timed out after an hour. Looking at the logs, it simply stops at the line "Installing pip dependencies: ... working ...", with no further detail.

Attempting to build the conda environment from the "model_service_env.yml" in a local terminal fail in the same spot. Manually removing the pip packages from the yml file and manually installing those packages in the conda environment does not fail. Also, removing just the "feast-azure-provider" package causes to build to succeed, and it can then be manually installed. However, since Azure ML does not allow the pip installs to be done in separate steps, this is not a viable workaround for the container deployment.

Hello , In using the feature online_store with the sample code form your provider example here.

With the current redis client version for Python 3.8 ( using code version as of date ) 0.3.0 https://pypi.org/project/feast-azure-provider/?msclkid=6debb6a2b9b511ec876e161bfda67531

Looks like the connection string does not exclude the default values for the Redis Cache connection string .elements Sample : myfsonlinesotre.redis.cache.windows.net:6380,password=zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz=,ssl=True,abortConnect=False

If you provide the connection string as is , Python Redis client throw exception about the abortConnect element

The only way the connection string work is by removing the ,abortConnect=False portion

Working Connection String myfsonlinesotre.redis.cache.windows.net:6380,password=zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz=,ssl=True

-George Gergues. Best of luck

I got following error when deploying service based on tutorial in:

Error: ... ImportError: cannot import name 'json' from itsdangerous ...

I have fixed that by pinning version of itsdangerous library in inference.dockerfile to 2.0.1 (newer version doesn't work):

RUN pip install 'azureml-defaults==1.35.0' \
                'feast-azure-provider==0.2.2' \
                'scikit-learn==0.22.2.post1' \
		'joblib===1.1.0'\
                'itsdangerous==2.0.1'

I followed this SO: https://stackoverflow.com/questions/71189819/python-docker-importerror-cannot-import-name-json-from-itsdangerous

Hi, I'm new to feast and was trying to reproduce the tutorial in azure portal. Faced the below error.

ProgrammingError: (pyodbc.ProgrammingError) ('42000', '[42000] [Microsoft][ODBC Driver 17 for SQL Server][SQL Server]Not able to validate external location because The remote server returned an error: (409) Conflict. (105215) (SQLExecDirectW)')
[SQL: COPY INTO dbo.driver_hourly
FROM 'https://feastonazuredatasamples.blob.core.windows.net/feastdatasamples/driver_hourly.csv'
WITH
(
	FILE_TYPE = 'CSV'
	,FIRSTROW = 2
	,MAXERRORS = 0
)
]

Thanks in advance!

Bumps pyspark from 3.1.3 to 3.2.2.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

We have been using Feast with a SQL db as an offline store and used JDBC to append features from a Spark dataframe directly to a table in SQL. Now for a recommender we'd like to build a historical dataset to train models on which will use a couple hundred-millions rows. Each is a customer with a timestamp. Feast's get_historical_features only takes a pandas dataframe as entity or a SQL query, so a workaround has been to store the entity df in the SQL db and use the query to fetch the features like so:

sql_job = fs.get_historical_features(
    entity_df="SELECT * FROM test_entitity_df",
    features=[
        'feature_view1:feature1',
        'feature_view1:feature2',
    ]
)

However, the sql_job only has to_df, to_arrow, or persist functionality. My question is, how to load features efficiently into a Spark DF for training? One solution would be to store the result of the Feast query in a sql table and use JDBC again to load that into Spark, however, I cannot seem to get the persist functionality to work as the docs on SavedDatasetStorage is very limited. Please advice.

Resources: https://docs.feast.dev/reference/offline-stores/overview#functionality https://docs.feast.dev/getting-started/concepts/dataset#creating-a-saved-dataset-from-historical-retrieval

Dear All,

I am quite used to work with terraform and Azure and I am happy to help adding the terraform script for creation of the two current infra setup.

Any suggestion or mention to add ?

Thanks Rob

train and deploy with Feast notebook fails (Part 3). When the following block is executed:

import uuid
from azureml.core.model import InferenceConfig
from azureml.core.environment import Environment
from azureml.core.model import Model

# get the registered model
model = Model(ws, "order_model")

# create an inference config i.e. the scoring script and environment
inference_config = InferenceConfig(
    entry_script="score.py", 
    environment=env, 
    source_directory="src"
)

# deploy the service
service_name = "orders-service" + str(uuid.uuid4())[:4]
service = Model.deploy(
    workspace=ws,
    name=service_name,
    models=[model],
    inference_config=inference_config,
    deployment_config=aciconfig,
)

service.wait_for_deployment(show_output=True)

The error is: ImportError: cannot import name 'case_insensitive_dict' from 'azure.core.utils'

Longer trace: Error:

{
  "code": "AciDeploymentFailed",
  "statusCode": 400,
  "message": "Aci Deployment failed with exception: Error in entry script, ImportError: cannot import name 'case_insensitive_dict' from 'azure.core.utils' (/azureml-envs/feast/lib/python3.8/site-packages/azure/core/utils/__init__.py), please run print(service.get_logs()) to get details.",
  "details": [
    {
      "code": "CrashLoopBackOff",
      "message": "Error in entry script, ImportError: cannot import name 'case_insensitive_dict' from 'azure.core.utils' (/azureml-envs/feast/lib/python3.8/site-packages/azure/core/utils/__init__.py), please run print(service.get_logs()) to get details."
    }
  ]
}

Hi, thankyou for creating this extension on feast, I am just curious if there will be any update on version on feast that will be used with feast-azure-provider, right now the default feast version its downloading is 0.18.x, but we are using 0.19+ version, so is there any way I can use feast azure provider with updated version of feast.