Collect events and performance counters from virtual machines with Azure Monitor Agent

This article describes how to collect events and performance counters from virtual machines by using Azure Monitor Agent.

Prerequisites

To complete this procedure, you need:

• Log Analytics workspace where you have at least contributor rights.
• Permissions to create Data Collection Rule objects in the workspace.
• Associate the data collection rule to specific virtual machines.

Create a data collection rule

You can define a data collection rule to send data from multiple machines to multiple Log Analytics workspaces, including workspaces in a different region or tenant. Create the data collection rule in the same region as your Log Analytics workspace. You can send Windows event and Syslog data to Azure Monitor Logs only. You can send performance counters to both Azure Monitor Metrics and Azure Monitor Logs.

Note

At this time, Microsoft.HybridCompute (Azure Arc-enabled servers) resources can't be viewed in Metrics Explorer (the Azure portal UX), but they can be acquired via the Metrics REST API (Metric Namespaces - List, Metric Definitions - List, and Metrics - List).

Note

To send data across tenants, you must first enable Azure Lighthouse.

Portal

1. On the Monitor menu, select Data Collection Rules.

2. Select Create to create a new data collection rule and associations.

   

3. Enter a Rule name and specify a Subscription, Resource Group, Region, and Platform Type:

   • Region specifies where the DCR will be created. The virtual machines and their associations can be in any subscription or resource group in the tenant.
   • Platform Type specifies the type of resources this rule can apply to. The Custom option allows for both Windows and Linux types.

   

4. On the Resources tab:

5. 1. Select + Add resources and associate resources to the data collection rule. Resources can be virtual machines, Virtual Machine Scale Sets, and Azure Arc for servers. The Azure portal installs Azure Monitor Agent on resources that don't already have it installed.

      Important

      The portal enables system-assigned managed identity on the target resources, along with existing user-assigned identities, if there are any. For existing applications, unless you specify the user-assigned identity in the request, the machine defaults to using system-assigned identity instead.

      If you need network isolation using private links, select existing endpoints from the same region for the respective resources or create a new endpoint.

   2. Select Enable Data Collection Endpoints.

   3. Select a data collection endpoint for each of the resources associate to the data collection rule.

   

6. On the Collect and deliver tab, select Add data source to add a data source and set a destination.

7. Select a Data source type.

8. Select which data you want to collect. For performance counters, you can select from a predefined set of objects and their sampling rate. For events, you can select from a set of logs and severity levels.

   

9. Select Custom to collect logs and performance counters that aren't currently supported data sources or to filter events by using XPath queries. You can then specify an XPath to collect any specific values. For an example, see Sample DCR.

   

10. On the Destination tab, add one or more destinations for the data source. You can select multiple destinations of the same or different types. For instance, you can select multiple Log Analytics workspaces, which is also known as multihoming.

    You can send Windows event and Syslog data sources to Azure Monitor Logs only. You can send performance counters to both Azure Monitor Metrics and Azure Monitor Logs. At this time, hybrid compute (Arc for Server) resources do not support the Azure Monitor Metrics (Preview) destination.

    

11. Select Add data source and then select Review + create to review the details of the data collection rule and association with the set of virtual machines.

12. Select Create to create the data collection rule.

API

1. Create a DCR file by using the JSON format shown in Sample DCR.

2. Create the rule by using the REST API.

3. Create an association for each virtual machine to the data collection rule by using the REST API.

PowerShell

Data collection rules

Action	Command
Get rules	Get-AzDataCollectionRule
Create a rule	New-AzDataCollectionRule
Update a rule	Set-AzDataCollectionRule
Delete a rule	Remove-AzDataCollectionRule
Update "Tags" for a rule	Update-AzDataCollectionRule

Data collection rule associations

Action	Command
Get associations	Get-AzDataCollectionRuleAssociation
Create an association	New-AzDataCollectionRuleAssociation
Delete an association	Remove-AzDataCollectionRuleAssociation

Azure CLI

This capability is enabled as part of the Azure CLI monitor-control-service extension. View all commands.

Resource Manager template

For sample templates, see Azure Resource Manager template samples for data collection rules in Azure Monitor.

Note

It can take up to 5 minutes for data to be sent to the destinations after you create the data collection rule.

Filter events using XPath queries

You're charged for any data you collect in a Log Analytics workspace. Therefore, you should only collect the event data you need. The basic configuration in the Azure portal provides you with a limited ability to filter out events.

Tip

For strategies to reduce your Azure Monitor costs, see Cost optimization and Azure Monitor.

To specify more filters, use custom configuration and specify an XPath that filters out the events you don't need. XPath entries are written in the form LogName!XPathQuery. For example, you might want to return only events from the Application event log with an event ID of 1035. The XPathQuery for these events would be *[System[EventID=1035]]. Because you want to retrieve the events from the Application event log, the XPath is Application!*[System[EventID=1035]]

Extract XPath queries from Windows Event Viewer

In Windows, you can use Event Viewer to extract XPath queries as shown in the screenshots.

When you paste the XPath query into the field on the Add data source screen, as shown in step 5, you must append the log type category followed by an exclamation point (!).

Tip

You can use the PowerShell cmdlet Get-WinEvent with the FilterXPath parameter to test the validity of an XPath query locally on your machine first. The following script shows an example:

$XPath = '*[System[EventID=1035]]'
Get-WinEvent -LogName 'Application' -FilterXPath $XPath

• In the preceding cmdlet, the value of the -LogName parameter is the initial part of the XPath query until the exclamation point (!). The rest of the XPath query goes into the $XPath parameter.
• If the script returns events, the query is valid.
• If you receive the message "No events were found that match the specified selection criteria," the query might be valid but there are no matching events on the local machine.
• If you receive the message "The specified query is invalid," the query syntax is invalid.

Examples of using a custom XPath to filter events:

Description	XPath
Collect only System events with Event ID = 4648	System!*[System[EventID=4648]]
Collect Security Log events with Event ID = 4648 and a process name of consent.exe	Security!*[System[(EventID=4648)]] and *[EventData[Data[@Name='ProcessName']='C:\Windows\System32\consent.exe']]
Collect all Critical, Error, Warning, and Information events from the System event log except for Event ID = 6 (Driver loaded)	System!*[System[(Level=1 or Level=2 or Level=3) and (EventID != 6)]]
Collect all success and failure Security events except for Event ID 4624 (Successful logon)	Security!*[System[(band(Keywords,13510798882111488)) and (EventID != 4624)]]

Note

For a list of limitations in the XPath supported by Windows event log, see XPath 1.0 limitations.
For instance, you can use the "position", "Band", and "timediff" functions within the query but other functions like "starts-with" and "contains" are not currently supported.

Next steps

• Collect text logs by using Azure Monitor Agent.
• Learn more about Azure Monitor Agent.
• Learn more about data collection rules.