Monitor multiple Azure Stack HCI clusters with Insights

Article
12/23/2024

Applies to: Azure Stack HCI, version 22H2

This article explains how to use Insights to monitor multiple Azure Stack HCI clusters. For a single Azure Stack HCI cluster, see Monitor Azure Stack HCI with Insights.

Important

If you registered your Azure Stack HCI cluster and configured Insights before November 2023, certain features that use Azure Monitor Agent (AMA), such as Arc for Servers, VM Insights, Defender for Cloud, or Sentinel might not collect logs and event data correctly. For troubleshooting guidance, see the Troubleshoot clusters registered before November 2023 section.

For information about the benefits, prerequisites, and how to enable Insights on each cluster, see Benefits, Prerequisites, and Enable Insights.

View health, performance, and usage insights

Insights stores its data in a Log Analytics workspace, which allows it to deliver powerful aggregation and filtering and analyze data trends over time. There is no direct cost for Insights. Users are billed based on the amount of data ingested and the data retention settings of their Log Analytics workspace.

You can access Insights from Azure Monitor > Insights hub > Azure Stack HCI. You will see the following tabs to toggle between views: Add to monitoring, Cluster health, Servers, Virtual machines, Storage.

Filtering results

The visualization can be filtered across subscriptions. You can filter the results based on the following drop-down menus:

Time range: This filter allows you to select a range for trend view. The default value is Last 24 hours.
Subscriptions: Shows the subscriptions that have registered Azure Stack HCI clusters. You can select multiple subscriptions in this filter.
HCI clusters: Lists the registered Azure Stack HCI clusters that have Logs and Monitoring capabilities enabled in the selected time range. You can select multiple clusters from this filter.
Resource groups: This filter allows you to select all the clusters within a resource group.

Add to monitoring

This feature provides details of clusters that are not monitored by the user. To start monitoring a cluster, select it to open that cluster, and then select Capabilities > Insights. If you don't see your cluster, make sure it has recently connected to Azure.

Column	Description	Example
Cluster	The name of the cluster.	27cls1
Azure connection status	The HCI resource status.	Connected
OS version	The operating system build on the server.	10.0.20348.10131

By default, the grid view shows the first 250 rows. You can set the value by editing the grid rows as shown in the following image:

You can export the details in Excel by selecting Export to Excel as shown in the following image:

Excel will provide Azure connection status as follows:

0: Not Registered
1: Disconnected
2: Not Recently
3: Connected

Cluster health

This view provides an overview of the health of clusters.

Column	Description	Example
Cluster	The name of the cluster.	27cls1
Last updated	The timestamp of when server was last updated.	4/9/2022, 12:15:42 PM
Status	Provides health of server resources in the cluster. It can be healthy, warning, critical, or other.	Healthy
Faulting resource	Description of which resource caused the fault.	Server, StoragePool, Subsystem
Total servers	The number of servers within a cluster.	4

If your cluster is missing or showing the status Other, go to the Log Analytics workspace used for the cluster and make sure that the Agent configuration is capturing data from the microsoft-windows-health/operational log. Also make sure the clusters have connected recently to Azure, and check that the clusters aren't filtered out in this workbook.

Server

This view provides an overview of server health and performance, and usage of selected clusters. This view is built using the server event ID 3000 of the Microsoft-Windows-SDDC-Management/Operational Windows Event Log Channel. Each row can be further expanded to see the node health status. You can interact with the cluster and server resource to navigate to the respective resource page.

Virtual machines

This view provides the state of all the VMs in the selected cluster. The view is built using the virtual machine event ID 3003 of the Microsoft-Windows-SDDC-Management/Operational Windows Event Log Channel. Each row can be further expanded to view the distribution of VMs across servers in the cluster. You can interact with the cluster and node resource to navigate to the respective resource page.

Metric	Description	Example
Cluster > Server	The name of the cluster. On expansion, it shows the servers within the cluster.	Sample-VM-1
Last Updated	The datetimestamp of when the server was last updated.	4/9/2022, 12:24:02 PM
Total VMs	The number of VMs in a server node within a cluster.	1 of 2 running
Running	The number of VMs running in a server node within a cluster.	2
Stopped	The number of VMs stopped in a server node within a cluster.	3
Failed	The number of VMs failed in a server node within a cluster.	2
Other	If VM is in one of the following states (Unknown, Starting, Snapshotting, Saving, Stopping, Pausing, Resuming, Paused, Suspended), it is considered as "Other."	2

Storage

This view shows the health of volumes, usage, and performance across monitored clusters. Expand a cluster to see the state of individual volumes. This view is built using the volume event ID 3002 of the Microsoft-Windows-SDDC-Management/Operational Windows Event Log Channel. The tiles on the top provide an overview of the health of storage.

Metric	Description	Example
Cluster > Volume	The name of the cluster. On expansion, it shows the volumes within a cluster.	AltaylCluster1 > ClusterPerformanceHistory
Last updated	The datetimestamp of when the storage was last updated.	4/14/2022, 2:58:55 PM
Volume health	The status of the volume. It can be healthy, warning, critical, or other.	Healthy
Size	The total capacity of the device in bytes during the reporting period.	25B
Usage	The percentage of available capacity during the reporting period.	23.54%
Iops	Input/output operations per second.	45/s
Trend	The IOPS trend.
Throughput	Number of bytes per second the Application Gateway has served.	5B/s
Trend (B/s)	The throughput trend.
Average Latency	Latency is the average time it takes for the I/O request to be completed.	334 μs

Customize Insights

Because the user experience is built on top of Azure Monitor workbook templates, users can edit the visualizations and queries and save them as a customized workbook.

If you are using the visualization from Azure Monitor > Insights hub > Azure Stack HCI, select Customize > Edit > Save As to save a copy of your modified version to a custom workbook.

Workbooks are saved within a resource group. Everyone with access to the resource group can access the customized workbook.

Most queries are written using Kusto Query Language (KQL). Some queries are written using the Resource Graph Query. For more information, see the following articles:

Support

To open a support ticket for Insights, use the service type Insights for Azure Stack HCI under Monitoring & Management.

Event Log Channel

Insights and monitoring views are based on Microsoft-Windows-SDDC-Management/Operational Windows Event Log Channel. When monitoring is enabled, the data from this channel is saved to a Log Analytics workspace.

Viewing and changing the dump cache interval

The default interval to dump the cache is set to 3600 seconds (1 hour).

Use the following PowerShell cmdlets to view the cache dump interval value:

Get-ClusterResource "sddc management" | Get-ClusterParameter

Use the following cmdlets to change the frequency of cache dump. If set to 0 it will stop publishing events:

Get-ClusterResource "sddc management" | Set-ClusterParameter -Name CacheDumpIntervalInSeconds -Value <value in seconds>

Windows events in the log channel

This channel includes five events. Each event has cluster name and Azure Resource Manager ID as EventData.

Event ID	Event type
3000	Server
3001	Drive
3002	Volume
3003	Virtual machine
3004	Cluster

Server event 3000 RenderedDescription column value

{
   "m_servers":[
      {
         "m_statusCategory":"Integer",
         "m_status":[
            "Integer",
            "…"
         ],
         "m_id":"String",
         "m_name":"String",
         "m_totalPhysicalMemoryInBytes":"Integer",
         "m_usedPhysicalMemoryInBytes":"Integer",
         "m_totalProcessorsUsedPercentage":"Integer",
         "m_totalClockSpeedInMHz":"Integer",
         "m_uptimeInSeconds":"Integer",
         "m_InboundNetworkUsage":"Double (Bits/sec)",
         "m_OutboundNetworkUsage":"Double (Bits/sec)",
         "m_InboundRdmaUsage":"Double (Bits/sec)",
         "m_OutboundRdmaUsage":"Double (Bits/sec)",
         "m_site":"String",
         "m_location":"String",
         "m_vm":{
            "m_totalVmsUnknown":"Integer",
            "m_totalVmsRunning":"Integer",
            "m_totalVmsStopped":"Integer",
            "m_totalVmsFailed":"Integer",
            "m_totalVmsPaused":"Integer",
            "m_totalVmsSuspended":"Integer",
            "m_totalVmsStarting":"Integer",
            "m_totalVmsSnapshotting":"Integer",
            "m_totalVmsSaving":"Integer",
            "m_totalVmsStopping":"Integer",
            "m_totalVmsPausing":"Integer",
            "m_totalVmsResuming":"Integer"
         },
         "m_osVersion":"String",
         "m_buildNumber":"String",
         "m_totalPhysicalProcessors":"Integer",
         "m_totalLogicalProcessors":"Integer"
      },
      "…"
   ],
   "m_alerts":{
      "m_totalUnknown":"Integer",
      "m_totalHealthy":"Integer",
      "m_totalWarning":"Integer",
      "m_totalCritical":"Integer"
   }
}

Most variables are self-explanatory from this JSON information. However, the table below lists a few variables which are a bit harder to understand.

Variable	Description
m_servers	Array of server nodes.
m_statusCategory	Health status of the server.
m_status	State of the server. It is an array that can contain one or two values. The first value is mandatory (0-4). The second value is optional (5-9).

Values for the m_statusCategory variable are as follows:

Value	Meaning
0	Healthy
1	Warning
2	Unhealthy
255	Other

Values for the m_status variable are as follows:

Value	Meaning
0	Up
1	Down
2	In maintenance
3	Joining
4	Normal
5	Isolated
6	Quarantined
7	Draining
8	Drain completed
9	Drain failed
0xffff	Unknown

Drive event 3001 RenderedDescription column value

Drive event 3001

{
    "m_drives":[
        {
            "m_uniqueId":"String",
            "m_model":"String",
            "m_type":"Integer",
            "m_canPool":"Boolean",
            "m_sizeInBytes":"Integer",
            "m_sizeUsedInBytes":"Integer",
            "m_alerts":{
                "m_totalUnknown":"Integer",
                "m_totalHealthy":"Integer",
                "m_totalWarning":"Integer",
                "m_totalCritical":"Integer"
            }
        },
        "…"
    ],
    "m_correlationId":"String",
    "m_isLastElement":"Boolean"
}

Volume event 3002 RenderedDescription column value

Volume event 3002

{
   "VolumeList":[
      {
         "m_Id":"String",
         "m_Label":"String",
         "m_Path":"String",
         "m_StatusCategory":"Integer",
         "m_Status":[
            "Integer",
            "…"
         ],
         "m_Size":"Integer (Bytes)",
         "m_SizeUsed":"Integer (Bytes)",
         "m_TotalIops":"Double (Count/second)",
         "m_TotalThroughput":"Double (Bytes/Second)",
         "m_AverageLatency":"Double (Seconds)",
         "m_Resiliency":"Integer",
         "m_IsDedupEnabled":"Boolean",
         "m_FileSystem":"String"
      },
      "…"
   ],
   "m_Alerts":{
      "m_totalUnknown":"Integer",
      "m_totalHealthy":"Integer",
      "m_totalWarning":"Integer",
      "m_totalCritical":"Integer"
   }
}

Most variables are self-explanatory from the above JSON information. However, the table below lists a few variables which are a bit harder to understand.

Variable	Description
VolumeList	Array of volumes.
m_StatusCategory	Health status of volume.
m_Status	State of the volume. It is an array that can contain one or two values. The first value is mandatory (0-4). The second value is optional (5-9).