Windows Custom Updates Solution

Windows Custom Updates Solution

Overview

This project is light weight patch management tool for Windows based OS. It helps to monitor and control patch management of OS. Originally written to help with patch management for Windows Virtual Desktop (WVD) based solutions but works also very well for other - non WVD based solutions or setups. (E.g. patch management for family computers)

When to use this project?

For patch management for Windows based workloads you have typically many enterprise level options. For example System Center Configuration Manager, Windows Update for Business, Windows Update, in case of WVD patch management also injection into base WVD image.

But sometimes, maybe you do not have setup which allows you to use tools highlighted above. Maybe you have many AD domains and forests, maybe you might be missing some capabilities like reporting, dashboards, telemetry to name few.

This is when you might benefit from this project which is trying to address such needs.

Generally, our recommendation is always to look first on professional solutions described above and only if you see some gaps with those professional solutions you might want to check this open source project.

Features

• Robust NT Service based tool, designed for Auto Recovery
• Automated updates installation, no manual interventions required
• Flexibility with updates configuration (E.g. you can specify which category of updates should be installed, maybe security patches only)
• Centralized place where telemetry data and states about monitored hosts and patch states is stored
• Powerful query language to analyze telemetry data in very fast way
• Rich Reporting capability based on powerful query language above
• Rich Alerting mechanism based on powerful query language above (E.g. send email when automatic update installation fails)
• This solution benefits from the Azure Cloud native components, providing serverless experience with high availability and redundancy, providing minimalist need for any maintenance (Configure once and forget)
• Configuration options how frequently should be online patches checked for availability and which category of patches should be downloaded and installed
• Windows Events Logs support
• Built on top of Windows Update Agent API
• Open source tool released under permissive free software MIT license. Anybody can benefit, anybody can contribute, no commercial restrictions.

Changes

Version 1.0.1

Telemetry data with Azure VM metadata

Telemetry data now supports information coming from Azure platform (If the VM runs inside Azure Platform). Information about VM metadata like Azure VM name (VMResourceName), resource group name (ResourceGroupName) and subscription ID (SubscriptionId). Basically information where your Azure VM is defined.

This allows to support richer filtering capabilities and identifying Azure VMs and where such VMs are defined. In cases when this hosting environment is NOT running inside Azure Platform, such information is not available in telemetry header.

Prerequisites

• Components of this solution require .NET Framework runtime (not .NET Core) to be installed on monitored hosts. For the time being it is version 4.8.
  • You can download the .NET Framework runtime here: https://dotnet.microsoft.com/download/dotnet-framework. Select Download .NET Framework Runtime. (not Developer Pack).
  • Information about .NET Framework runtime installation is here: https://docs.microsoft.com/en-us/dotnet/framework/deployment/deployment-guide-for-developers.
• You will need to have access to Azure Cloud and provision Log Analytics Workspace. You can verify the access in Azure Portal https://portal.azure.com and try to search and create the service Application Insights and Log Analytics Workspace. You can also create free Azure account if needed: https://azure.microsoft.com/en-us/free

Installation

• Make sure all project prerequisites are met and installed
• Download the latest version of this project from the releases web page: https://github.com/jbinko/Windows-Custom-Updates-Solution/releases. Following files are available in assets section:
  • WVDCUS.zip - contains binary files which needs to be deployed to monitored hosts
  • Source code.zip - Source code of this project
  • Source code.tar.gz - Source code of this project in different tar/zip format
• Download the file WVDCUS.zip. In the downloads directory right click Properties on the file and mark the checkbox Unblock. Click OK.
• Extract the content of the WVDCUS.zip file on all monitored hosts into the WVDCUS directory. For example into the directory: C:Program FilesWVDCUS

Provisioning of required Azure Cloud Services

• Go to the Azure Portal https://portal.azure.com. If you do not have Azure Cloud enabled yet, you can create free Azure account: https://azure.microsoft.com/en-us/free
• Create or use existing Resource Group in preferred region: https://docs.microsoft.com/en-us/azure/azure-resource-manager/management/manage-resource-groups-portal#create-resource-groups
• Into Resource Group create resource Log Analytics Workspace in preferred region, with unique name and pay-as-you-go plan: https://docs.microsoft.com/en-us/azure/azure-monitor/learn/quick-create-workspace#create-a-workspace
• Into Resource Group create workspace-based resource Application Insights into the same region as previous resource, with unique name, with workspace-based option and select the workspace you created before. https://docs.microsoft.com/en-us/azure/azure-monitor/app/create-workspace-resource#create-workspace-based-resource
• On the just created Application Insights resource click Overview and copy Instrumentation Key alias Telemetry Key which will be used later during configuration.

Configuration

• Go to the directory where you extracted the binary files of this project. E.g. directory: C:Program FilesWVDCUS
• Open the file WVDCUS.Service.exe.config in the notepad (or any other text editor)
• Search for the configuration value TelemetryKey and replace the content 00000000-0000-0000-0000-000000000000 with the value you obtained from the Application Insights resource from Instrumentation Key item.
• When needed, change the value of configuration key CheckForUpdatesIntervalHours to more suitable value. This value specifies how frequently check for new updates should be done. Default is 4 hours and valid value can be between 1 - 590 hours.
• When needed, change the value of the configuration key UpdatesSearchQuery to more suitable value. This value specifies which type and categories of patches should be installed automatically. Description of the query and parameters are here: https://docs.microsoft.com/en-us/previous-versions/windows/desktop/ff357803(v=vs.85) and here: https://docs.microsoft.com/en-us/windows/win32/api/wuapi/nf-wuapi-iupdatesearcher-search
• Save the file
• Repeat the same configuration for the file WVDCUS.Console.exe.config
• Save the file

Test a Configuration with Console Application

• If you configured and provisioned mandatory settings, you can execute the Console Application WVDCUS.Console.exe now to see if the application is configured correctly. Watch for any errors in the application output. If you see some errors verify that you have enough permissions. Application might to check, download and install patches if they are available.
• Check that application works correctly (including patches installations) and press any key to stop the application

Register Windows Service

Once you verified a configuration with the console application you can finalize configuration with registration of application to run as Windows service. To register application to run as Windows service, you must have administrator credentials on the computer where it's installed. You need to use installutil.exe tool, which is part of .NET runtime. This tool is installed with the .NET Framework to the folder %windir%Microsoft.NETFramework[64]<framework version>. For example, the default path for the 64-bit version is %windir%Microsoft.NETFramework64v4.0.30319InstallUtil.exe.

• Open Command Prompt with administrative credentials - Run as Administrator
• Change directory into directory where you extracted binary files. E.g. cd C:Program FilesWVDCUS
• Execute registration process with command: C:WindowsMicrosoft.NETFramework64v4.0.30319InstallUtil.exe WVDCUS.Service.exe
• If the service installs successfully, the command reports success. If the installutil.exe process fails, check the install log to find out why. By default, the log is in the same folder as the service executable.
• If you run console application WVDCUS.Console.exe before - then the installutil.exe will fail. The error is that Event Log is registered already by console application. If fails, just uninstall the Windows service with command bellow (removes Event Log) and install Windows service again.
• Uninstall command (parameter /u): C:WindowsMicrosoft.NETFramework64v4.0.30319InstallUtil.exe /u WVDCUS.Service.exe
• Install command again: C:WindowsMicrosoft.NETFramework64v4.0.30319InstallUtil.exe WVDCUS.Service.exe

Verify Windows Service

• Run command: services.msc where you can find just registered Windows Service under the display name WVDCUS Service. The name of the service itself is WVDCUS.
• Open the service WVDCUS Service. On Recovery tab specify for First and Second failure Restart the Service action.
• Click Apply
• On General tab click button Start
• Service should remain in Started state.

Verify health of Windows Service

You can see detailed information about health of service, failures, informational messages in an event log.

• Run command: eventvwr.msc where you can find detailed information in log files.
• In Event Viewer, expand Applications and Services Logs, search for WVDCUSLog log.
• Check information inside the log file to identify any potential issues

Query Telemetry Data

Once the Windows Service is up and in running state it will start to generate Telemetry data and store them into provisioned Application Insights resource through TelemetryKey identifier. Because the Application Insights resource was connected with Log Analytics Workspace during provisioning process you might benefit from all features which are available in Log Analytics Workspace resource. For more details see: https://docs.microsoft.com/en-us/azure/azure-monitor/log-query/log-query-overview

Telemetry data are stored as custom events and you might find them in the customEvents table. This allows you to view and analyze data from all hosts and see current state, health and informational messages about patch management from many perspectives and specified time range. For more details see: https://docs.microsoft.com/en-us/azure/azure-monitor/log-query/get-started-queries

Reporting

Analyzing data through queries and multiple perspectives is very powerful feature. However, for daily monitoring you typically need to quickly see overall state of your hosts through some kind of reports. Fortunately, Log Analytics Workspace resource has such capability as well.

You can create custom workbook with custom queries over telemetry data and render results through relevant visual representation.
For more details see: https://docs.microsoft.com/en-us/azure/azure-monitor/platform/workbooks-overview

• Go to the Application Insights resource, section Workbooks
• Create New Workbook
• Copy/Paste query from below
• Select Time Range for Telemetry Data. This value needs to be aligned with CheckForUpdatesIntervalHours settings. When Time Range is lower you might not see any data.
• Click Run Query button to render chart from the latest data
• In the Toolbar at the top click the Done Editing button
• In the Toolbar at the top click the Save button and save the Workbook under some preferred name.

Next time you can go directly into the Workbooks sections and click on just created Workbook to see the report with latest data.

let WVDCUS = customEvents 
| where itemType == 'customEvent' and operation_Name == 'WVDCUS.Event';

let WVDCUS_TelemetryStart = WVDCUS 
| where name == 'TelemetryStart'
| project MachineName = tostring(customDimensions.MachineName), operation_Id, timestamp, VMResourceName = tostring(customDimensions.VMResourceName), ResourceGroupName = tostring(customDimensions.ResourceGroupName), SubscriptionId = tostring(customDimensions.SubscriptionId)
| summarize arg_max(timestamp, *) by MachineName;

let WVDCUS_Result = WVDCUS_TelemetryStart
| join kind = innerunique customEvents on operation_Id
| where name in ('NoUpdatesAvailable', 'UpdatesInstalledNoUpdates', 'UpdatesNotInstalledFailed', 'UpdatesInstalledRebootRequired', 'UpdatesInstalledAllOK', 'Error')
| extend State=replace(@'NoUpdatesAvailable', @'Up to date', name)
| extend State=replace(@'UpdatesInstalledNoUpdates', @'Up to date', State)
| extend State=replace(@'UpdatesNotInstalledFailed', @'Failed', State)
| extend State=replace(@'UpdatesInstalledRebootRequired', @'Reboot Required', State)
| extend State=replace(@'UpdatesInstalledAllOK', @'Up to date', State)
| extend State=replace(@'Error', @'Error', State)
| project MachineName, Timestamp = timestamp1, State, VMResourceName, ResourceGroupName, SubscriptionId;

WVDCUS_Result
| project MachineName, State, VMResourceName, ResourceGroupName, SubscriptionId
| summarize Count=count() by State
| render piechart;

You might want to also see report showing current state of each host. You can extend just created Workbook with new report part.

• Select just create Workbook
• In the Toolbar at the top click the Edit button
• In lower part you will see Add button. Click it and select Add Query.
• Copy/Paste query from below
• Select Time Range for Telemetry Data. This value needs to be aligned with CheckForUpdatesIntervalHours settings. When Time Range is lower you might not see any data.
• Click Run Query button to see results from the latest data
• In the Toolbar at the top click the Done Editing button
• In the Toolbar at the top click the Save button

let WVDCUS = customEvents 
| where itemType == 'customEvent' and operation_Name == 'WVDCUS.Event';

let WVDCUS_TelemetryStart = WVDCUS 
| where name == 'TelemetryStart'
| project MachineName = tostring(customDimensions.MachineName), operation_Id, timestamp, VMResourceName = tostring(customDimensions.VMResourceName), ResourceGroupName = tostring(customDimensions.ResourceGroupName), SubscriptionId = tostring(customDimensions.SubscriptionId)
| summarize arg_max(timestamp, *) by MachineName;

let WVDCUS_Result = WVDCUS_TelemetryStart
| join kind = innerunique customEvents on operation_Id
| where name in ('NoUpdatesAvailable', 'UpdatesInstalledNoUpdates', 'UpdatesNotInstalledFailed', 'UpdatesInstalledRebootRequired', 'UpdatesInstalledAllOK', 'Error')
| extend State=replace(@'NoUpdatesAvailable', @'Up to date', name)
| extend State=replace(@'UpdatesInstalledNoUpdates', @'Up to date', State)
| extend State=replace(@'UpdatesNotInstalledFailed', @'Failed', State)
| extend State=replace(@'UpdatesInstalledRebootRequired', @'Reboot Required', State)
| extend State=replace(@'UpdatesInstalledAllOK', @'Up to date', State)
| extend State=replace(@'Error', @'Error', State)
| project MachineName, Timestamp = timestamp1, State, VMResourceName, ResourceGroupName, SubscriptionId;

WVDCUS_Result
| project MachineName, Timestamp, State, VMResourceName, ResourceGroupName, SubscriptionId

You can see combined example report - containing two parts on picture below.

Alerting

Finally, you might not have interest in checking reports daily. Maybe you want to do it only if this is really needed. In this case you might benefit from alerting mechanism which will notify you that some host needs your attention.

Again, Log Analytics Workspace has such built-in capability. For more details see: https://docs.microsoft.com/en-us/azure/azure-monitor/learn/tutorial-response

Here is how you can create Alert based on set of conditions.

• Go to the Application Insights resource, section Logs, close window with Example queries
• Select little Eye icon next to the customEvents table.
• You should see sample of data from this table, click See in query editor
• Click New alert rule button in the toolbar
• In the window keep Scope as it is - should be pointing to correct Application Insights instance.
• Focus now on Condition section, specify Time range value - This value needs to be aligned with CheckForUpdatesIntervalHours settings. When Time Range is lower you might not see any data.
• Copy/Paste query from below into Search query text box
• Specify Alert logic, based on Number of results, Greater than and threshold value 0 - basically the alert will be triggered if some new record which is in state Failed, Reboot Required, Error was returned by query.
• Provide value for Period in minutes - time span over which to execute the query.
• Provide Frequency value - how frequently you would like to scan for new changes.
• You need to create or select existing Action Group. Click the button Select action group. If needed, click to Create action group. See https://docs.microsoft.com/en-us/azure/azure-monitor/platform/action-groups?WT.mc_id=Portal-Microsoft_Azure_Monitoring.
• You can customize email subject name
• Provide Alert rule name and Description
• Finalize Alert by clicking button Create alert rule

let WVDCUS = customEvents 
| where itemType == 'customEvent' and operation_Name == 'WVDCUS.Event';

let WVDCUS_TelemetryStart = WVDCUS 
| where name == 'TelemetryStart'
| project MachineName = tostring(customDimensions.MachineName), operation_Id, timestamp, VMResourceName = tostring(customDimensions.VMResourceName), ResourceGroupName = tostring(customDimensions.ResourceGroupName), SubscriptionId = tostring(customDimensions.SubscriptionId)
| summarize arg_max(timestamp, *) by MachineName;

let WVDCUS_Result = WVDCUS_TelemetryStart
| join kind = innerunique customEvents on operation_Id
| where name in ('NoUpdatesAvailable', 'UpdatesInstalledNoUpdates', 'UpdatesNotInstalledFailed', 'UpdatesInstalledRebootRequired', 'UpdatesInstalledAllOK', 'Error')
| extend State=replace(@'NoUpdatesAvailable', @'Up to date', name)
| extend State=replace(@'UpdatesInstalledNoUpdates', @'Up to date', State)
| extend State=replace(@'UpdatesNotInstalledFailed', @'Failed', State)
| extend State=replace(@'UpdatesInstalledRebootRequired', @'Reboot Required', State)
| extend State=replace(@'UpdatesInstalledAllOK', @'Up to date', State)
| extend State=replace(@'Error', @'Error', State)
| project MachineName, Timestamp = timestamp1, State, VMResourceName, ResourceGroupName, SubscriptionId;

WVDCUS_Result
| where State in ('Failed','Reboot Required', 'Error')
| project MachineName, Timestamp, State, VMResourceName, ResourceGroupName, SubscriptionId

This should trigger action and attention when some host will fall into the one of the states Failed, Reboot Required, Error.

Alert email which you can receive to attract your attention.