edit apache article to be up-to-date, clean sfi violations and bring acrolinx scorte above 80

Author: markingmyname

articles/synapse-analytics/spark/apache-spark-handle-livy-error.md

@@ -3,73 +3,69 @@ title: Handle Livy Errors on Apache Spark in Synapse
 description: Learn how to handle and interpret job failures on Apache Spark in Synapse Analytics.
 author: midesa
 ms.author: midesa
-ms.date: 02/27/2025
+ms.date: 03/30/2026
 ms.service: azure-synapse-analytics
 ms.subservice: spark
 ms.topic: error-reference
 ---
 
-# Interpret error codes in Synapse Analytics
+# Interpret error codes in Azure Synapse Analytics
 
-There are many factors that can play into why a spark application fails in Azure Synapse Analytics today. For instance, it can be due to a system error or even a user related error. Previously, all errors corresponding to failing jobs on Synapse Analytics were surfaced with a generic error code displaying *LIVY_JOB_STATE_DEAD*. This error code gave no further insight into why the job failed. It requires significant effort to identify the root cause by digging into the driver, executor, Spark Event, Livy logs, and find a resolution.
+Many factors can cause a Spark application to fail in Synapse Analytics. For example, the failure can stem from a system error or a user error. Previously, all errors that correspond to failing jobs on Synapse Analytics surfaced a generic error code that displayed *LIVY_JOB_STATE_DEAD*. This error code gave no further insight into why the job failed. You had to put in significant effort to identify the root cause by digging into the driver, executor, Spark Event, and Livy logs to find a resolution.
 
-:::image type="content" source="media/apache-spark-handle-livy-error/apache-spark-old-error-view.png" border="true" alt-text="Screenshot of Apache Spark error code without detailed message." lightbox="media/apache-spark-handle-livy-error/apache-spark-old-error-view.png":::
-
-We have introduced a more precise list of error codes that replaces the previous generic message. The new message describes the cause of failure. Whenever a job fails on Azure Synapse Analytics, the error handling feature parses and checks the logs on the backend to identify the root cause. It then displays a message to the user on the monitoring pane along with the steps to resolve the issue.
-
-:::image type="content" source="media/apache-spark-handle-livy-error/apache-spark-new-error-view.png" border="true" alt-text="Screenshot of Apache Spark error code with detailed message." lightbox="media/apache-spark-handle-livy-error/apache-spark-new-error-view.png":::
+The new error codes provide a more precise list that replaces the previous generic message. The new message describes the cause of failure. Whenever a job fails on Synapse Analytics, the error handling feature parses and checks the logs on the backend to identify the root cause. It then displays a message to the user on the monitoring pane along with the steps to resolve the issue.
 
 ## Enable error classification in Synapse
 
-The error classification feature can be enabled or disabled by setting the following Spark configuration to `true` or `false` at the job or pool level:
+Set the following Spark configuration to `true` or `false` at the job or pool level to enable or disable the error classification feature:
 
 `livy.rsc.synapse.error-classification.enabled`
 
-The following section lists some error types that are currently supported. We're continuously refining and adding more to these error codes by improving our model.
+The following section lists some error types that are currently supported. The product team is continuously refining and adding more error codes by improving the model.
 
 ## Error code categories
 
-Each error code falls under one of the following four buckets:
+Each error code falls under one of the following four categories:
 
-- **User** - Indicating a user error
-- **System** - Indicating a system error
+- **User** - Indicates a user error
+- **System** - Indicates a system error
 - **Ambiguous** - Could be either user or system error
 - **Unknown** - No classification yet, most probably because the error type isn't included in the model
 
 ## Error code examples for each classification type
 
 ### Spark_User_TypeError_TypeNotIterable
 
-In Python, the error `TypeError: argument of type 'insert type' is not iterable` occurs when the membership operator (in, not in) is used to validate the membership of a value in non iterable objects such as list, tuple, dictionary. This is usually due to the search of value in a non-iterable object. Possible solutions:
+In Python, the error `TypeError: argument of type 'insert type' is not iterable` occurs when you use the membership operator (`in`, `not in`) to check if a value is in non-iterable objects such as list, tuple, or dictionary. This error usually happens because you try to search for a value in a non-iterable object. To fix this error, try the following solutions:
 
 - Check if the value is present in the iterable object.
-- If you want to check one value to another, use logical operator instead of Membership Operator.
-- If the membership operator contains "None" value, it won't be able to iterate, and a null check or assigned default must be done.
-- Check if the type of the value used can actually be checked and the typing is correct.
+- If you want to check one value against another, use a logical operator instead of the membership operator.
+- If the membership operator contains a `None` value, it can't iterate. Add a null check or assign a default value.
+- Check if the type of the value you're using can actually be checked and if the typing is correct.
 
 ### Spark_System_ABFS_OperationFailed
 
-An operation with Azure Data Lake Storage (ADLS) Gen2 has failed.
+An operation with Azure Data Lake Storage (ADLS) Gen2 failed.
 
-This error occurs typically due to a permissions issue.
+This error typically happens because of a permissions problem.
 
-Ensure that for all ADLS Gen2 resources referenced in the Spark job, has "Storage Blob Data Contributor" RBAC role on the storage accounts the job is expected to read and write from.
-Check the logs for this Spark application. Navigate to your Synapse Studio, select the **Monitor** tab from the left pane. From the **Activities** section, select **Apache Spark Applications** and find your Spark job from the list. For the ADLS Gen2 storage account name that is experiencing this issue, inspect the logs available in the **Logs** tab at the bottom part of this page.
+Make sure that all ADLS Gen2 resources referenced in the Spark job have the **Storage Blob Data Contributor** RBAC role on the storage accounts that the job needs to read and write from.
+Check the logs for this Spark application. Go to your Synapse Studio, select the **Monitor** tab from the left pane. From the **Activities** section, select **Apache Spark Applications** and find your Spark job from the list. For the ADLS Gen2 storage account name that is experiencing this problem, check the logs available in the **Logs** tab at the bottom part of this page.
 
 ### Spark_Ambiguous_ClassLoader_NoClassDefFound
 
-A class required by the code couldn't be found when the script was run. For more information, see:
+A class required by the code couldn't be found when the script ran. For more information, see:
 
-- For Notebook scenarios: [Apache Spark manage packages for interactive jobs](./apache-spark-manage-scala-packages.md) 
+- For Notebook scenarios: [Apache Spark manage packages for interactive jobs](./apache-spark-manage-scala-packages.md)
 - For Spark batch scenarios (see section 6): [Apache Spark manage packages for batch jobs](./apache-spark-job-definitions.md#create-an-apache-spark-job-definition-for-apache-sparkscala)
 
-Ensure that all the code dependencies are included in the JARs Synapse runs. If you do not or cannot include third party JARs with your own code, ensure that all dependencies are included in the workspace packages for the Spark pool you are executing code on, or they are included in the "Reference files" listing for the Spark batch submission. 
+Make sure that all the code dependencies are included in the JARs that Synapse runs. If you don't or can't include third party JARs with your own code, make sure that all dependencies are included in the workspace packages for the Spark pool you're executing code on, or they're included in the **Reference files** listing for the Spark batch submission.
 
 ### Spark_Unknown_Unknown_java.lang.Exception
 
-An unknown failure, the model wasn't able to classify.
+An unknown failure. The model can't classify the error.
 
-The error codes (including and beyond the previous list) along with the troubleshooting instructions on how to resolve the issue will show up on the Synapse Studio application error pane if this feature is enabled.
+If you enable this feature, Synapse Studio shows the error codes (including and beyond the previous list) along with troubleshooting instructions in the application error pane.
 
-> [!NOTE]
-> If you built any tooling around the Synapse monitoring job that checks for a failing job by filtering the `LIVY_JOB_STATE_DEAD` error code, your app would no longer work. Because the returned error codes are different. Modify any scripts accordingly in order to utilize this feature or disable the feature if it's not needed.
+> [!NOTE]  
+> If you built any tooling around the Synapse monitoring job that checks for a failing job by filtering the `LIVY_JOB_STATE_DEAD` error code, your app no longer works because the returned error codes are different. Modify any scripts accordingly to use this feature or disable the feature if it's not needed.