We have created alert rule using Insight metrics for Free space . The detection logic works well. However big problem I face is neither the email, nor the alert in portal shows what logical drive on that particular machine caused alert. Q:\Can you please help me guide how to add drive to the alert and email? (LikeC:, D:,...) Alert rule with build in signal name Email notification Fired:Sev2 Azure Monitor Alert VM Free Space - vm-winserver-azuremonitortest02 on vm-winserver-azuremonitortest02 ( microsoft.compute/virtualmachines ) at 6/7/2023 5:55:48 PM View the alert in Azure Monitor > Summary Alert name VM Free Space - vm-winserver-azuremonitortest02 Severity Sev2 Monitor condition Fired Affected resource vm-winserver-azuremonitortest02 Resource type microsoft.compute/virtualmachines Resource group rg_lab_stelcik Subscription Microsoft Azure Sponsorship Monitoring service Platform Signal type Metric Fired time June 7, 2023 17:55 UTC Alert ID f816d974-8dcf-469e-8588-265e86a6e799 Alert rule ID https://portal.azure.com/xxxxxxxx Metric alert condition type SingleResourceMultipleMetricCriteria Time aggregation Minimum Metric name LogicalDisk\% Free Space Metric namespace Azure.VM.Windows.GuestMetrics Metric value (when alert fired) 90 Operator LessThan Threshold 30

Hi, The way the alert is defined it monitors the free space across all drives on the server. I believe there is dimension with name DiskMountId. When you add that dimension to the alert rule you will receive alert per disk and the threshold will be applied seperately for each disk. When the alert is fired the value in DiskMountId will be there to see which disk has reached the threshold. Please " Accept the answer " if the information helped you. This will help us and others in the community as well.

yep that the tricky part, ama is running and config looked good to me also....weird... anyhow thanks for all your help Stanislav. Will try to dig more and then provide root cause if found for future references.

Azure Alert rule Free space metrics email notification does not show Drive

Accepted answer

Stanislav Zhelyazkov 21,411 Reputation points MVP

2023-06-08T12:03:23.6933333+00:00

Hi,

The way the alert is defined it monitors the free space across all drives on the server. I believe there is dimension with name DiskMountId. When you add that dimension to the alert rule you will receive alert per disk and the threshold will be applied seperately for each disk. When the alert is fired the value in DiskMountId will be there to see which disk has reached the threshold.

Please "Accept the answer" if the information helped you. This will help us and others in the community as well.
Please sign in to rate this answer.
Marek Stelcik 20 Reputation points

2023-06-08T14:31:00.4566667+00:00

Hi,

in the dimension I only see Instance and cannot get to DiskMountID dimension.

Stanislav Zhelyazkov 21,411 Reputation points MVP

2023-06-09T09:44:14.3333333+00:00

Hi,

May be you have not configured correctly. I would assume that you have not configured to send VM Insights metrics to Azure Metrics store or you are not selecting the correct metric. The data collection rule you hare using for VM Insights should contain sending the performance metrics to Azure Metrics storage:

Once you have that you will see in the Metrics window for that VM the metrics and you can see that there is DiskMountId. Note the Metrics namespace.

From there you can create new alert rule with that dimension added:

Marek Stelcik 20 Reputation points

2023-06-09T12:41:43.1466667+00:00

Thank you,

The first screenshot is from what resource in your case please?

I created one new resource and onboarded it old way (i.e.

on VM ->Insights I select connect via Log Analytics agent

then in Monitor->VM->overview configured DCR

i see the metrics CORRECTLY there and is also available for the previously onboarded machines via Azure Monitor Agent.

However when I have it onboarded new way via AMA which is what I was doing until now.

1 VM->insight-Azure Monitor Agent

2 DCR

its does not show there, and the guest level metric cannot be splitted further mounteddiskid

is it possible this metric is available only in older LA agent?

Stanislav Zhelyazkov 21,411 Reputation points MVP

2023-06-09T12:54:42.42+00:00

as mentioned the first screenshot is for Data collection rule. Data collection rules are resources that defined what data to collect by using Azure Monitor agent. When you use the UI experience to onboard VM to VM Insights such rule is created but the rule created by default does not sends the data to Azure Metrics store, it only sends it to Azure Log Analytics workspace. You need to modify your data collection rule with the changes needed for sending the metrics to Azure Metrics store and after that you will see those appear under the metrics of the VM. That is of course if you want to create metric alerts type for these metrics. If you want to Log Analytics, you can create Log Alert v2 and achieve the same thing. Here is a guide on editing data collection rules

Marek Stelcik 20 Reputation points

2023-06-09T13:58:05.0066667+00:00

Thanks Stanislav, I think I am getting somewhere. So after analysis of my case, it seems I need to change the destination from mine

destinations": [ "VMInsightsPerf-Logs-Dest" ]

to

azuremonitormetrics-default,correct?

However it is strange to me, as I have the same DCR used for 3 machines and only for one machine the correct insights are populated (the one with LA agent).

Stanislav Zhelyazkov 21,411 Reputation points MVP

2023-06-09T14:05:13.2233333+00:00

In dataFlows you already have VM Insights metrics being send to Azure Monitor metrics - the first entry. The configuration above looks good and does not need any more modification. Translated the configuration says - VM Insights metrics will be send to both the Log Analytics workspace and the Azure Metrics store. Service Map data will be send only to Log Analytics workspace as that is the only option for that data. Make sure that you have monitor agent installed on the machine and that this Data collection rule is assigned to the VM.

Marek Stelcik 20 Reputation points

2023-06-09T17:19:39.5133333+00:00

Once i had Steps done as @Stanislav Zhelyazkov described + installed Performance diagnostic extension it started working. But the installing Perf extension seemed to be mandatory, not just AMA.

Thank you @Stanislav Zhelyazkov you are star.

Stanislav Zhelyazkov 21,411 Reputation points MVP

2023-06-09T17:30:09.13+00:00

Glad that it worked. Overall you only need Azure Monitor agent, dcr with VM insights metrics send to Metric store in order to see the metrics. No other extension is needed. Prior taking the screen shots above I did it on brand new VM and workspace in order to show you things visually.

Marek Stelcik 20 Reputation points

2023-06-13T11:31:10.1+00:00

Stanislav, do you please know if it is supported to use this metrics we discuss over multiple VM at once (scope section in alert rules). If I am creating individually for VM it works, if I select two at once (even though individually work) the metrics become unavailable. Thank you
Sign in to comment