Building an Effective Strategy
Mastering Evaluation Techniques
How-To Guides

Building an Effective Strategy

Identifying What to Evaluate

The user interacts with the frontend UI to pose a question.
The frontend service forwards the user’s question to the Orchestrator.
The Orchestrator retrieves the user’s conversation history from the database.
The Orchestrator accesses the AI Search key stored in the Key Vault.
The Orchestrator retrieves relevant documents from the AI Search index.
The Orchestrator uses Azure OpenAI to generate a user response.

The connection from the App Service to Storage Account indicates the scenario when the user wants to view the document that grounds the provided answer.
The connection from the App Service to Speech Services indicates the cases when the user wishes to interact with the application through audio.

Test Scenario

RPM = (u * p * s * i) / n / 60

Key Metrics to Monitor During Load Tests

   Request Rate: Monitor the request rate during load testing. Ensure that the LLM application can handle the expected number of requests per second.
   Response Time: Analyze response times under different loads. Identify bottlenecks and optimize slow components.
   Throughput: Measure the number of successful requests per unit of time. Optimize for higher throughput.
   Resource Utilization: Monitor CPU, memory, and disk usage. Ensure efficient resource utilization.

Best Practices for Executing Load Tests

   Test Scenarios: Create realistic test scenarios that mimic actual user behavior
   Ramp-Up Strategy: Gradually increase the load to simulate real-world traffic patterns. The warm-up period typically lasts between 20 to 60 seconds. After the warm-up, the actual load test begins
   Think Time: Include think time between requests to simulate user interactions.
   Geographical Distribution: Test from different Azure regions to assess global performance.

Performance Tuning Strategies for LLM Apps

Application Design

Optimize Application Code: Examine and refine the algorithms and backend systems of your LLM application to increase efficiency. Utilize asynchronous processing methods, such as Python’s async/await, to elevate application performance. This method allows data processing without interrupting other tasks.
Batch Processing: Batch LLM requests whenever possible to reduce overhead. Grouping multiple requests for simultaneous processing improves throughput and efficiency by allowing the model to better leverage parallel processing capabilities, thereby optimizing overall performance.
Implement Caching: Use caching for repetitive queries to reduce the application’s load and speed up response times. This is especially beneficial in LLM applications where similar questions are frequently asked. Caching answers to common questions minimizes the need to run the model repeatedly for the same inputs, saving both time and computational resources. Some examples of how you can implement this include using Redis as a semantic cache or Azure APIM policies.
Revisit your Retry Logic: LLM model deployments might start to operate at their capacity, which can lead to 429 errors. A well-designed retry mechanism can help maintain application responsiveness. With the OpenAI Python SDK, you can opt for an exponential backoff algorithm. This algorithm gradually increases the wait time between retries, helping to prevent service overload. Additionally, consider the option of falling back on another model deployment. For more information, refer to the load balance item in the Solution Architecture section.

Prompt Design

​Microsoft Tech Community – Latest Blogs –Read More 

Today, I worked on a service request where our customer detected a high number of execution plans consuming resources in the plan cache for a single query. I would like to share my lessons learned and experience to prevent this type of issue.

We have the following table definition:

CREATE Table TestTable(ID INT IDENTITY(1,1), string_column NVARCHAR(500))

–We added dummy data in the table running the following script.
DECLARE @Total INT = 40000;
DECLARE @I int =0
DECLARE @Fill INT;
DECLARE @Letter INT;
WHILE @i <= @Total
BEGIN
SET @I=@I+1
SET @Letter = CAST((RAND(CHECKSUM(NEWID())) * (90 – 65 + 1)) + 65 AS INT)
set @Fill = CAST((RAND(CHECKSUM(NEWID())) * 500) + 1 AS INT)
INSERT INTO TestTable (string_column) values(REPLICATE(CHAR(@Letter),@Fill))
end

— Finally, we created a new index for this column.
create index TestTable_Ix1 on TestTable (String_column)

Our customer identified that the application is generating this query: 

SELECT TOP 1 * FROM TestTable WHERE string_column = N’AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA’

To reproduce the issue and understand the impact about how many execution plan our customer reported, we started running the demo function called StartAdhocNoParam: This function executes a non-parameterized query. Running the following DMV to identify the number of plans we could see around 13K cached plans.

— dbcc freeproccache –Only to clear the cache.

WITH XMLNAMESPACES (DEFAULT ‘http://schemas.microsoft.com/sqlserver/2004/07/showplan’)
SELECT
qs.sql_handle,
qs.execution_count,
qs.total_elapsed_time,
qs.total_logical_reads,
qs.total_logical_writes,
qs.total_worker_time,
qs.creation_time,
qs.last_execution_time,
st.text AS sql_text,
qp.query_plan
FROM
sys.dm_exec_query_stats AS qs
CROSS APPLY
sys.dm_exec_sql_text(qs.sql_handle) AS st
CROSS APPLY
sys.dm_exec_query_plan(qs.plan_handle) AS qp
WHERE
st.text LIKE ‘%SELECT TOP 1 * FROM TestTable%’

In this situation, we changed the property of the database called Parameterization to Force, to  This resulted in only one execution plan with a parameter. That’s is great but our customer wants to modify the source code and avoiding using Parameterization to Force.

Additionally:

OPTIMIZE_FOR_AD_HOC_WORKLOADS might reduce the memory usage, altohough it may not promote the plan reuse – Database scoped optimizing for ad hoc workloads – Microsoft Community Hub 

Also, review the option called plan guides that might help on that – Create a New Plan Guide – SQL Server | Microsoft Learn

When our customer finished the modification of their code, we noticed that their application is not specifing the size of parameter or specifing the length of the text that the application is searching, like we could see in the function demo StartAdhocWithParam.

This function is going to run a parametrized query using different length for the parameter, because, for example, if the application is not specifying the length of the parameter or the text that is looking for. In this situation, running the DMV to identify the number of plans we could see around 500 cached plans.

In this situation, we suggested using the function StartParametrize, specifying the max length of the column (500), we could have only an action plan. This reduced the cached plan usage. 

This exercise highlights the importance of specifying the length of the parameter, 

Finally, I would like to share two new functions: 

ImprovedVersionStartParametrize that helps us to reduce the roundtrips of the text sent to the database, only sending values.
GetColumnLength that connects to the database to determine the total size of the column base on the internal table INFORMATION_SCHEMA.columns to perform this more dynamic.

sing System;
using System.Data;
using Microsoft.Data.SqlClient;

class Program
{
static void Main()
{
// Parámetros de conexión
string connectionString = “Server=tcp:servername.database.windows.net,1433;User Id=username;Password=pwd!;Initial Catalog=dbname;Persist Security Info=False;MultipleActiveResultSets=False;Encrypt=True;TrustServerCertificate=False;Connection Timeout=30;Pooling=true;Max Pool size=100;Min Pool Size=1;ConnectRetryCount=3;ConnectRetryInterval=10;Application Name=ConnTest”;

//ImprovedVersionStartParametrize(connectionString);
for (int j = 65; j <= 90; j = j + 1)
{
Console.WriteLine(“Letter:” + (char)j);
for (int i = 1; i <= 500; i = i + 1)
{
if (i % 10 == 0)
{
Console.Write(” {0} ,”, i);
}
//StartAdhocWithParam(connectionString, (char)j, i);
//StartAdhocWithGuide(connectionString, (char)j, i);
StartAdhocNoParam(connectionString, (char)j,i);
//StartParametrize(connectionString, (char)j, i);
}
}
}

static void StartAdhocWithParam(string connectionString, char Letter, int Length)
{
string query = “SELECT TOP 1 * FROM TestTable WHERE string_column = @stringParam –Adhoc with Param”;
using (SqlConnection conn = new SqlConnection(connectionString))
{
using (SqlCommand cmd = new SqlCommand(query, conn))
{
string stringParam = new string(Letter, Length);
cmd.Parameters.Add(new SqlParameter(“@stringParam”, SqlDbType.NVarChar, Length) { Value = stringParam });
conn.Open();
SqlDataReader reader = cmd.ExecuteReader();
}
}
}

static void StartAdhocNoParam(string connectionString, char Letter, int Length)
{
string stringParam = new string(Letter, Length);
string query = “SELECT TOP 1 * FROM TestTable WHERE string_column = N'” + stringParam + “‘ –Adhoc without Param”;
using (SqlConnection conn = new SqlConnection(connectionString))
{
using (SqlCommand cmd = new SqlCommand(query, conn))
{
conn.Open();
SqlDataReader reader = cmd.ExecuteReader();
}
}
}

static void StartParametrize(string connectionString, char Letter, int Length)
{
string query = “SELECT TOP 1 * FROM TestTable WHERE string_column = @stringParam –Adhoc with Max Length”;
using (SqlConnection conn = new SqlConnection(connectionString))
{
using (SqlCommand cmd = new SqlCommand(query, conn))
{
string stringParam = new string(Letter, Length);
cmd.Parameters.Add(new SqlParameter(“@stringParam”, SqlDbType.NVarChar, 500) { Value = stringParam });
conn.Open();
SqlDataReader reader = cmd.ExecuteReader();
}
}
}

static void ImprovedVersionStartParametrize(string connectionString)
{
string query = “SELECT TOP 1 * FROM TestTable WHERE string_column = @stringParam –Adhoc with Max Length”;
using (SqlConnection conn = new SqlConnection(connectionString))
{
using (SqlCommand cmd = new SqlCommand(query, conn))
{
cmd.Parameters.Add(new SqlParameter(“@stringParam”, SqlDbType.NVarChar, GetColumnLength(connectionString, “dbo”, “TestTable”, “string_column”)));
conn.Open();
cmd.Prepare();
for (int j = 65; j <= 90; j = j + 1)
{
Console.WriteLine(“Letter:” + (char)j);
for (int i = 1; i <= 500; i = i + 1)
{
if (i % 10 == 0)
{
Console.Write(” {0} ,”, i);
}
cmd.Parameters[0].Value = new string((char)j, i);
SqlDataReader reader = cmd.ExecuteReader();
reader.Close();
}
}
}
}
}

static void StartAdhocWithGuide(string connectionString, char Letter, int Length)
{
string query = @”
DECLARE @sqlQuery NVARCHAR(MAX) = N’SELECT TOP 1 * FROM TestTable WHERE string_column = @stringColumn’;
EXEC sp_executesql @sqlQuery, N’@stringColumn NVARCHAR(500)’, @stringColumn = @stringParam”;

using (SqlConnection conn = new SqlConnection(connectionString))
{
using (SqlCommand cmd = new SqlCommand(query, conn))
{
string stringParam = new string(Letter, Length);
cmd.Parameters.Add(new SqlParameter(“@stringParam”, SqlDbType.NVarChar, Length) { Value = stringParam });
conn.Open();
SqlDataReader reader = cmd.ExecuteReader();
}
}
}
static int GetColumnLength(string connectionString, string schemaName, string tableName, string columnName)
{
using (SqlConnection connection = new SqlConnection(connectionString))
{

using (SqlCommand cmd = new SqlCommand(@”
SELECT CHARACTER_MAXIMUM_LENGTH
FROM INFORMATION_SCHEMA.COLUMNS
WHERE TABLE_SCHEMA = @SchemaName AND TABLE_NAME = @NameT AND COLUMN_NAME = @ColumnName”, connection))
{
cmd.Parameters.Add(“@SchemaName”, SqlDbType.NVarChar, 128);
cmd.Parameters.Add(“@NameT”, SqlDbType.NVarChar, 128);
cmd.Parameters.Add(“@ColumnName”, SqlDbType.NVarChar, 128);

cmd.Parameters[“@SchemaName”].Value=schemaName;
cmd.Parameters[“@NameT”].Value = tableName;
cmd.Parameters[“@ColumnName”].Value = columnName;

connection.Open();
var result = cmd.ExecuteScalar();
if (result != null)
{
return Convert.ToInt32(result);
}
else
{
return 0;
}
}
}
}
}

Disclaimer

The use of this application and the provided scripts is intended for educational and informational purposes only. The scripts and methods demonstrated in this guide are provided “as is” without any warranties or guarantees. It is the user’s responsibility to ensure the accuracy, reliability, and suitability of these tools for their specific needs.

​Microsoft Tech Community – Latest Blogs –Read More 

Introduction

When it comes to building a robust foundation for ETL (Extract, Transform, Load) pipelines, the trio of Azure Data Factory or Azure Synapse Analytics, Azure Batch, and Azure Storage is indispensable. These tools enable efficient data movement, transformation, and processing across diverse data sources, thereby helping us achieve our strategic goals.

This document provides a comprehensive guide on how to authenticate Azure Batch with SAMI and Azure Storage with Synapse SAMI. This enables user-driven connectivity to storage, facilitating data extraction. Furthermore, it allows the use of custom activities, such as High-Performance Computing (HPC), to process the extracted data.

The key enabler of these functionalities is the Synapse Pipeline. Serving as the primary orchestrator, the Synapse Pipeline is adept at integrating various Azure resources in a secure manner. Its capabilities can be extended to Azure Data Factory (ADF), providing a broader scope of data management and transformation.

Through this guide, you will gain insights into leveraging these powerful Azure services to optimize your data processing workflows.

Services Overview

During this procedure we will use different services, below you have more details about each of them.

Azure Synapse Analytics / Data Factory

Azure Synapse Analytics is an enterprise analytics service that accelerates time to insight across data warehouses and big data systems. Azure Synapse brings together the best of SQL technologies used in enterprise data warehousing, Spark technologies used for big data, Data Explorer for log and time series analytics, Pipelines for data integration and ETL/ELT, and deep integration with other Azure services such as Power BI, CosmosDB, and AzureML.
Documentation:

What is Azure Synapse Analytics? – Azure Synapse Analytics | Microsoft Learn
Introduction to Azure Data Factory – Azure Data Factory | Microsoft Learn
 

Azure Batch

Azure Batch is a powerful platform service designed for running large-scale parallel and high-performance computing (HPC) applications in the cloud.
Documentation: Azure Batch runs large parallel jobs in the cloud – Azure Batch | Microsoft Learn
 

Azure Storage

Azure Storage provides scalable and secure storage services for various data types, including services like Azure Blob storage, Azure Table storage, and Azure Queue storage.
Documentation: Introduction to Azure Storage – Cloud storage on Azure | Microsoft Learn

Managed Identities

Azure Managed Identities are a feature of Azure Active Directory that automatically manages credentials for applications to use when connecting to resources that support Azure AD authentication. They eliminate the need for developers to manage secrets, credentials, certificates, and keys.
There are two types of managed identities:

System-assigned: Tied to your application.
User-assigned: A standalone Azure resource that can be assigned to your app

Documentation: Managed identities for Azure resources – Managed identities for Azure resources | Microsoft Learn

Scenario

Run an ADF / Synapse Pipeline that pulls a script located in a Storage Account and execute it into the Batch nodes using User Assigned Managed Identities (UAMI) for Authentication to Storage and System Assigned Managed Identity (SAMI) to authenticate with Batch.

Prerequisites

ADF / Synapse Workspace

Documentation: Quickstart: create a Synapse workspace – Azure Synapse Analytics | Microsoft Learn

UA Managed Identity

Documentation: Manage user-assigned managed identities – Managed identities for Azure resources | Microsoft Learn
Blog Documentation: https://techcommunity.microsoft.com/t5/azure-data-factory-blog/support-for-user-assigned-managed-identity-in-azure-data-factory/ba-p/2841013

Storage Account

Documentation: Create a storage account – Azure Storage | Microsoft Learn

Procedure Overview

During this procedure we will walk through step by step to complete the following actions:

Create UAMI Credentials
Create Linked Services for Storage and Batch Accounts
Add UAMI and SAMI to Storage and Batch Accounts
Create, Configure and Execute an ADF / Synapse Pipeline

We will refer to ADF (Portal, Workspace, Pipelines, Jobs, Linked Services) as Synapse during all the exercise and examples to avoid redundancy.

Debugging

Procedure

Create UAMI Credentials

1. In your Synapse Portal, go to Manage -> Credentials -> New and fill in the details and click Create.

Create Linked Services Connections for Storage and Batch

2. In your Synapse Portal, go to Manage – Linked Services -> New -> Azure Blob Storage -> Continue and complete the form

a. Authentication Type: UAMI

b. Azure Subscription: Choose your one

c. Storage Account name: Choose your one where the script to be used is allocated

d. Credentials: choose the created into the Step #1

e. Click on Create

3. In Azure Portal go to your Batch Account -> Keys and Copy the Batch Account name & Account Endpoint to be used in next step, also copy the Pool Name to be used for this example.

4. In your Synapse Portal, go to Manage -> Linked Services -> New -> Azure Batch -> Continue and fill in the information

a. Authentication Method: SAMI (Copy the Managed Identity Name to be used later)

b. Account Name, Batch URL and Pool Name: Paste on here the values copied from Step#3

c. Storage linked service Name: Choose the one created from Step#2

5. Publish all your changes

Adding UAMI RBAC Roles to Storage Account

6. In the Azure Portal, go to your Storage Account -> Access Control (IAM)

a. Click on Add Option and then on Add role assignment and search for “Storage Blob Data Contributor”, then click on Next.

b. Choose Managed Identity and select your UAMI click on Select and then click Next, Next and Review + assign.

Adding SAMI RBAC Roles to Batch Account

7. In the Azure Portal, go to your Batch Account -> Access Control (IAM)

a. Click on Add Option and then on Add role assignment

b. Click on “Privileged administrator roles” tab and then choose the Contributor role and click Next.

c. Choose Managed Identity and under Managed Identity lookup for “Synapse workspace” and then choose the SAMI same as it is added into the step 4a., then click on Select and Next, Next and Review and Assign.

Adding UAMI to Batch Pool

If you need to create a new Batch Pool, you can follow the following procedure:

Documentation: Configure managed identities in Batch pools – Azure Batch | Microsoft Learn
Make sure to select the UAMI configured into the Step 1

8. If you already have a Batch Pool created follow the next steps:

a. Into the Azure Portal go to your Batch Account -> Pools -> Choose your Pool -> Go to Identity

b. Click on Add then choose the necessary UAMI (on this example it was selected the one used by the Synapse Linked Services for Storage and another one used for other integrations) and click on Add.
 

Important: In case your Batch Pool use multiples UAMI’s (as example to connect with Key Vault or other services), you have first to remove the existing one and then add all of them together.

c. Then, it is required to Scale in and Scale out the Pool to apply the changes.

Setting up the Pipeline

9. In your Synapse Portal, go to Integrate -> Add New Resource -> Pipeline

10. Into the right panel Activities -> Batch Services -> Drag and drop the Custom activities

11. In the Azure Batch tab details for the Custom Activities, click on the Azure Batch linked service and click the one created in Step 4 and test the connection (if you receive a connection error, please go to the Troubleshooting scenario 1)

12. Then go to Settings tab and add your script. Ffor this example, we will use a Powershell script previously uploaded to a Storage Blob Container and send the output to txt file.

a. Command: your script details

b. Resource linked Service: The Storage Service Linked connection configured previously on Step#2

c. Browse Storage: lookup for the Container where your script was uploaded

       d. Publish your Changes and perform a Debug

Debugging

12. Check the Synapse Jobs Logs and outputs

       a. Copy the Activity Run ID
 

       b. Then, in the Azure Portal Go to your Storage Account –> Containers -> adfjobs -> select the folder with the activityID -> output.

       c. On here you will find two files, “stderr.txt” and “stdout.txt” both of them contains information about the errors or the outputs of the commands executed during the task execution

13. Check the Batch Logs and outputs. To get the Batch logs you have different ways:
 

       a. Over Nodes: In Azure Portal go to your Batch Account -> Pools -> Choose your Pool -> Nodes -> then into the Folders details go to the folder for this Synapse execution -> job-x -> lookup for the activityID
 

       b. Over Jobs: In Azure Portal go to your Batch Account -> Jobs -> Select a pool with a name of adfv2-yourPoolName -> click on the Task with the ID same as it was the ActivityID of the Synapse Pipeline from step 12a.

What we have learned

During this walkthrough procedure we have learned and implemented about

Authentication: Utilizing User Assigned Managed Identities (UAMI) and System Assigned Managed Identity (SAMI) for secure connections.
Linked Services: Creation and configuration of linked services for Azure Storage and Azure Batch accounts.
Pipeline Execution: Steps to create, configure, and execute an ADF/Synapse Pipeline, emphasizing the use of Synapse as a unified term to avoid redundancy.
Debugging: Detailed instructions for creating credentials, adding RBAC roles, and setting up pipelines, along with troubleshooting tips.
Logs Analysis: How to access and analyze Synapse Jobs logs and Azure Batch logs for troubleshooting.
Error Handling: Understanding the significance of ‘stderr.txt’ and ‘stdout.txt’ files in identifying and resolving errors during task execution.

If you have any questions or feedback, please leave a comment below!

​Microsoft Tech Community – Latest Blogs –Read More 

Hi experts,

I’m trying to deploy an hybrid cluster with Azure Stack HCI 23H2 servers, I follow the steps in the documentation:

https://learn.microsoft.com/en-us/azure-stack/hci/deploy/deployment-introduction

I’m deploying the cluster from Azure portal and I get this error message:

I reviewed the C:MASLogsAzStackHciEnvironmentChecker.log log and this is the error:

[5/25/2024 2:52:12 PM] [INFORMATIONAL] [Add-AzStackHciEnvJob] Adding current job to progress: System.Collections.Hashtable
[5/25/2024 2:52:12 PM] [INFORMATIONAL] [Test-OrganizationalUnit] Executing Test-OrganizationalUnit
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test on LAB-HCI1
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing tests with parameters:
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] ClusterName : mscluster
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] UsersADOUPath : OU=Users,OU=ms309,DC=mycompany,DC=com
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] AdServer : mycompany.com
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] NamingPrefix : HCI01
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] PhysicalMachineNames : LAB-HCI1 LAB-HCI2
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] AdCredentialsUserName : msdeployuser
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] ADOUPath : OU=ms309,DC=mycompany,DC=com
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] DomainFQDN : mycompany.com
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] ComputersADOUPath : OU=Computers,OU=ms309,DC=mycompany,DC=com
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] AdCredentials : System.Management.Automation.PSCredential
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test RequiredOrgUnitsExist
[5/25/2024 2:52:12 PM] [INFO] [RequiredOrgUnitsExist] Checking for the existance of OU: OU=ms309,DC=mycompany,DC=com
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Test RequiredOrgUnitsExist completed with: System.Collections.Hashtable
[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test LogPhysicalMachineObjectsIfExist
[5/25/2024 2:52:12 PM] [INFO] [PhysicalMachineObjectsExist] Validating seednode : LAB-HCI1 is part of a domain or not
[5/25/2024 2:52:13 PM] [ERROR] [PhysicalMachineObjectsExist] Seed node LAB-HCI1 joined to the domain. Disconnect the seed node from the domain and proceed with the deployment
[5/25/2024 2:52:13 PM] [INFO] [Test-OrganizationalUnitOnSession] Test LogPhysicalMachineObjectsIfExist completed with: System.Collections.Hashtable
[5/25/2024 2:52:13 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test GpoInheritanceIsBlocked
[5/25/2024 2:52:17 PM] [INFO] [Test-OrganizationalUnitOnSession] Test GpoInheritanceIsBlocked completed with:
[5/25/2024 2:52:17 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test ExecutingAsDeploymentUser
[5/25/2024 2:52:17 PM] [WARNING] [ExecutingAsDeploymentUser] User ‘msdeployuser not found in ‘ hence skipping the rights permission check. This may cause deployment failure during domain join phase if the user doesn’t have the permissions to create or delete computer objects
[5/25/2024 2:52:17 PM] [INFO] [Test-OrganizationalUnitOnSession] Test ExecutingAsDeploymentUser completed with: System.Collections.Hashtable
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Close-AzStackHciEnvJob] Updating current job to progress with endTime: 2024/05/25 14:52:17 and duration 5
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciEnvProgress] AzStackHCI progress written: MASLogsAzStackHciEnvironmentReport.xml
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciEnvReport] JSON report written to MASLogsAzStackHciEnvironmentReport.json
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Log location: MASLogsAzStackHciEnvironmentChecker.log
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Report location: MASLogsAzStackHciEnvironmentReport.json
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Use -Passthru parameter to return results as a PSObject.
[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Invoke-AzStackHciExternalActiveDirectoryValidation completed. Id:ArcInitializationExternalActiveDirectoryc04daeb4

I assigned all admin permissions in the AD (like Administrators and Domain Admins Groups) and Delegate Control of the OU for msdeployuser 

Regards.

​Hi experts, I’m trying to deploy an hybrid cluster with Azure Stack HCI 23H2 servers, I follow the steps in the documentation: https://learn.microsoft.com/en-us/azure-stack/hci/deploy/deployment-introduction I’m deploying the cluster from Azure portal and I get this error message:                                I reviewed the C:MASLogsAzStackHciEnvironmentChecker.log log and this is the error: [5/25/2024 2:52:12 PM] [INFORMATIONAL] [Add-AzStackHciEnvJob] Adding current job to progress: System.Collections.Hashtable[5/25/2024 2:52:12 PM] [INFORMATIONAL] [Test-OrganizationalUnit] Executing Test-OrganizationalUnit[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test on LAB-HCI1[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing tests with parameters:[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] ClusterName : mscluster[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] UsersADOUPath : OU=Users,OU=ms309,DC=mycompany,DC=com[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] AdServer : mycompany.com[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] NamingPrefix : HCI01[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] PhysicalMachineNames : LAB-HCI1 LAB-HCI2[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] AdCredentialsUserName : msdeployuser[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] ADOUPath : OU=ms309,DC=mycompany,DC=com[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] DomainFQDN : mycompany.com[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] ComputersADOUPath : OU=Computers,OU=ms309,DC=mycompany,DC=com[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] AdCredentials : System.Management.Automation.PSCredential[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test RequiredOrgUnitsExist[5/25/2024 2:52:12 PM] [INFO] [RequiredOrgUnitsExist] Checking for the existance of OU: OU=ms309,DC=mycompany,DC=com[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Test RequiredOrgUnitsExist completed with: System.Collections.Hashtable[5/25/2024 2:52:12 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test LogPhysicalMachineObjectsIfExist[5/25/2024 2:52:12 PM] [INFO] [PhysicalMachineObjectsExist] Validating seednode : LAB-HCI1 is part of a domain or not[5/25/2024 2:52:13 PM] [ERROR] [PhysicalMachineObjectsExist] Seed node LAB-HCI1 joined to the domain. Disconnect the seed node from the domain and proceed with the deployment[5/25/2024 2:52:13 PM] [INFO] [Test-OrganizationalUnitOnSession] Test LogPhysicalMachineObjectsIfExist completed with: System.Collections.Hashtable[5/25/2024 2:52:13 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test GpoInheritanceIsBlocked[5/25/2024 2:52:17 PM] [INFO] [Test-OrganizationalUnitOnSession] Test GpoInheritanceIsBlocked completed with:[5/25/2024 2:52:17 PM] [INFO] [Test-OrganizationalUnitOnSession] Executing test ExecutingAsDeploymentUser[5/25/2024 2:52:17 PM] [WARNING] [ExecutingAsDeploymentUser] User ‘msdeployuser not found in ‘ hence skipping the rights permission check. This may cause deployment failure during domain join phase if the user doesn’t have the permissions to create or delete computer objects[5/25/2024 2:52:17 PM] [INFO] [Test-OrganizationalUnitOnSession] Test ExecutingAsDeploymentUser completed with: System.Collections.Hashtable[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Close-AzStackHciEnvJob] Updating current job to progress with endTime: 2024/05/25 14:52:17 and duration 5[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciEnvProgress] AzStackHCI progress written: MASLogsAzStackHciEnvironmentReport.xml[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciEnvReport] JSON report written to MASLogsAzStackHciEnvironmentReport.json[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Log location: MASLogsAzStackHciEnvironmentChecker.log[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Report location: MASLogsAzStackHciEnvironmentReport.json[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Use -Passthru parameter to return results as a PSObject.[5/25/2024 2:52:17 PM] [INFORMATIONAL] [Write-AzStackHciFooter] Invoke-AzStackHciExternalActiveDirectoryValidation completed. Id:ArcInitializationExternalActiveDirectoryc04daeb4 I assigned all admin permissions in the AD (like Administrators and Domain Admins Groups) and Delegate Control of the OU for msdeployuser  Regards.  Read More

​