Azure Monitor for VMs Generally Available (GA) Frequently Asked Questions

This General Availability FAQ covers changes that are happening in Azure Monitor for VMs as we prepare for our GA release.

Updates for Azure Monitor for VMs

We are planning to release a new version of Azure Monitor for VMs in January 2020. Customers enabling Azure Monitors for VMs after this release automatically receive the new version, but existing customers already using Azure Monitor for VMs will be prompted to upgrade. This FAQ and our documentation offers guidance to perform an upgrade at scale if you have large deployments across multiple workspaces.

With this upgrade, Azure Monitor for VMs performance data are stored in the same InsightsMetrics table as Azure Monitor for containers, and makes it easier for you to query the two data sets. Also, you are able to store more diverse data sets that we could not store in the table previously used. Our performance views will also be updated to use this new table.

We are moving to new data types for our connection data sets. This change will be happening in December 2019 and will be announced in an Azure update blog. Data presently stored in ServiceMapComputer_CL and ServiceMapProcess_CL, which are custom log tables, will move to dedicated data types named VMComputer and VMProcess. By moving to dedicated data types, they receive priority for data ingestion and the table schema will be standardized across all customers.

We realize that asking existing customers to upgrade causes disruption to their workflow, which is why we have chosen to do this now while in Public Preview rather than later after GA.

What is changing?

Currently when you complete the onboarding process for Azure Monitor for VMs, you enable the Service Map solution in the workspace you selected to store your monitoring data, and then configure performance counters for the data we collect from your VMs. We will be releasing a new solution, named VMInsights, that includes additional capabilities for data collection along with a new location for storing this data in your Log Analytics workspace.

Our current process of using performance counters in your Log Analytics workspace sends the data to the Perf table. This new solution sends the data to a table named InsightsMetrics that is also used by Azure Monitor for containers. This table schema allows us to store additional metrics and service data sets that are not compatible with the Perf table format.

What should I do about the Performance counters in my workspace if I install the VMInsights solution?

The current method of enabling Azure Monitor for VMs uses performance counters in your workspace. The new method stores this data in a new table, named InsightsMetrics.

Once we update our user interface to use the data in the InsightsMetrics table, we will update our documentation, and communicate this announcement via multiple channels, including displaying a banner in the Azure portal. At that point, you may choose to disable these performance counters in your workspace if you no longer need to use them.

Note

If you have Alert Rules that reference these counters in the Perf table, you need to update them to reference new data stored in the InsightsMetrics table. Refer to our documentation for example log queries that you can use that refer to this table.

If you decide to keep the performance counters enabled, you will be billed for the data ingested and stored in the Perf table based on [Log Analytics pricing[(https://azure.microsoft.com/pricing/details/monitor/).

How will this change affect my alert rules?

If you have created Log alerts that query the Perf table targeting performance counters that were enabled in the workspace, you should update these rules to refer to the InsightsMetrics table instead. This guidance also applies to any log search rules using ServiceMapComputer_CL and ServiceMapProcess_CL, because those data sets are moving to VMComputer and VMProcess tables.

We will update this FAQ and our documentation to include example log search alert rules for the data sets we collect.

How will this affect my bill?

Billing is still based on data ingested and retained in your Log Analytics workspace.

The machine level performance data that we collect is the same, is of a similar size to the data we stored in the Perf table, and will cost approximately the same amount.

What if I only want to use Service Map?

That is fine. You will see prompts in the Azure portal when viewing Azure Monitor for VMs about the upcoming update. Once released, you will receive a prompt requesting that you update to the new version. If you prefer to only use the Maps feature, you can choose not to upgrade and continue to use the Maps feature in Azure Monitor for VMs and the Service Map solution accessed from your workspace or dashboard tile.

If you chose to manually enable the performance counters in your workspace, then you may be able to see data in some of our performance charts viewed from Azure Monitor. Once the new solution is released we will update our performance charts to query the data stored in the InsightsMetrics table. If you would like to see data from that table in these charts, you will need to upgrade to the new version of Azure Monitor for VMs.

The changes to move data from ServiceMapComputer_CL and ServiceMapProcess_CL will affect both Service Map and Azure Monitor for VMs, so you still need to plan for this update.

If you chose to not upgrade to the VMInsights solution, we will continue to provide legacy versions of our performance workbooks that refer to data in the Perf table.

Will the Service Map data sets also be stored in InsightsMetrics?

The data sets will not be duplicated if you use both solutions. Both offerings share the data sets that will be stored in VMComputer (formerly ServiceMapComputer_CL), VMProcess (formerly ServiceMapProcess_CL), VMConnection, and VMBoundPort tables to store the map data sets that we collect.

The InsightsMetrics table will store VM, process, and service data sets that we collect and will only be populated if you are using Azure Monitor for VMs and the VM Insights solution. The Service Map solution will not collect or or store data in the InsightsMetrics table.

Will I be double charged if I have the Service Map and VMInsights solutions in my workspace?

No, the two solutions share the map data sets that we store in VMComputer (formerly ServiceMapComputer_CL), VMProcess (formerly ServiceMapProcess_CL), VMConnection, and VMBoundPort. You will not be double charged for this data if you have both solutions in your workspace.

If I remove either the Service Map or VMInsights solution will it remove my data?

No, the two solutions share the map data sets that we store in VMComputer (formerly ServiceMapComputer_CL), VMProcess (formerly ServiceMapProcess_CL), VMConnection, and VMBoundPort. If you remove one of the solutions, these data sets notice that there is still a solution in place that uses the data and it remains in the Log Analytics workspace. You need to remove both solutions from your workspace in order for the data to be removed from it.

When will this update be released?

We expect to release the update for Azure Monitor for VMs in early January 2020. As we get closer to the release date in January, we'll post updates here and present notifications in the Azure portal when you open Azure Monitor.

Health feature is in limited public preview

We have received a lot of great feedback from customers about our VM Health feature set. There is a lot of interest around this feature and excitement over its potential for supporting monitoring workflows. We are planning to make a series of changes to add functionality and address the feedback we have received.

To minimize impact of these changes to new customers, we have moved this feature into a limited public preview. This update happened in October 2019.

We plan to re-launch this Health feature in 2020, after Azure Monitor for VMs is in GA.

How do existing customers access the Health feature?

Existing customers that are using the Health feature will continue to have access to it, but it will not be offered to new customers.

To access the feature, you can add the following feature flag feature.vmhealth=true to the Azure portal URL https://portal.azure.com. Example https://portal.azure.com/?feature.vmhealth=true.

You can also use this short url, which sets the feature flag automatically: https://aka.ms/vmhealthpreview.

As an existing customer, you can continue to use the Health feature on VMs that are connected to an existing workspace setup with the health functionality.

I use VM Health now with one environment and would like to deploy it to a new one

If you are an existing customer that is using the Health feature and want to use it for a new roll-out, please contact us at vminsights@microsoft.com to request instructions.

Next steps

To understand the requirements and methods that help you monitor your virtual machines, review Deploy Azure Monitor for VMs.