Security exception while using Feature Store. How can I get this whitelisted?

[VM]

I was following the Databricks Academy "New Capabilities Overview: Feature Store" module. However when I try to run the code in the example notebook I get a security exception as explained below.

When I try to run the example notebook "01-Populate a feature store" I get an error while creating a feature table, I get a

py4j.security.Py4JSecurityException: Method public java.lang.String org.apache.spark.api.java.JavaSparkContext.getLocalProperty(java.lang.String) is not whitelisted on class class org.apache.spark.api.java.JavaSparkContext

This error is on the first line of the code

fs.create_feature_table(
 
   name=f"fxs_eu_ntwkdaor_efsx_001.tmp_{username}_trip_pickup_features",
 
   keys=["zip", "ts"],
 
   features_df=pickup_features,
 
   partition_columns="yyyy_mm",
 
   description="Taxi Fares. Pickup Features",
 
)
 
fs.create_feature_table(
 
   name=f"fxs_eu_ntwkdaor_efsx_001.tmp_{username}_trip_dropoff_features",
 
   keys=["zip", "ts"],
 
   features_df=dropoff_features,
 
   partition_columns="yyyy_mm",
 
   description="Taxi Fares. Dropoff Features",
 
)

:

How can I get this whitelisted?

[VM]

Hi @Kaniz Fatma​ 

Yes. It is a high concurrency cluster. Screenshot attached.

[VM · Accepted solution]

The issue was due to the use of High Concurrency cluster. It was resolved when a standard cluster ML cluster was used. See screenshot.

[998925]

@Kaniz Fatma​ - this is happening again with the following configuration.

 

What configuration shall we use now that Cluster mode Standard is no longer a configuration option?

[998925]

@Kaniz Fatma​ any thoughts on the above? thanks so much

[OMG]

Hi @Daniel Barrundia​ - please select "No isolation shared" Access mode, it should resolve this problem.