How to Check HCI Cluster Health Before Troubleshooting a VM
  

Lucky1001 Lv1Posted 2026-Aug-15 09:42

Hi Sangfor Community,

When a virtual machine has performance or connectivity issues, it is easy to focus only on the VM itself.

However, before troubleshooting the VM, I think it is useful to check the overall HCI cluster health first.

I normally check the following:

* HCI node status
* CPU and memory utilization
* Storage health and capacity
* Network connectivity
* Cluster alarms
* Hardware or disk errors


For example, if a VM suddenly becomes slow, but one of the HCI nodes is showing a storage or hardware warning, the problem may not be related to the VM itself.

Therefore, checking the HCI cluster health can help us identify potential infrastructure problems before troubleshooting the application or VM.

For experienced Sangfor HCI administrators:

What is the first thing you normally check when a VM has performance issues?

I would be interested to hear your troubleshooting workflow and recommendations.

Thanks!

Like this topic? Like it or reward the author.

Creating a topic earns you 5 coins. A featured or excellent topic earns you more coins. What is Coin?

Enter your mobile phone number and company name for better service. Go

Newbie517762 Lv5Posted 2026-Aug-17 09:39
  
HiHi,

For a VM experiencing performance issues, the first step is to check the HCI cluster health status.

Infrastructure-level problems can impact VM performance even when the VM itself appears healthy. Common causes include:

  • Storage performance or connectivity issues
  • Hardware warnings or failures on cluster nodes
  • Cluster service or resource operation problems
  • Host or storage subsystem health issues


The built-in health checks help identify these conditions by validating:

  • Cluster service health and platform resource status
  • Cluster configuration and operational readiness
  • Host health and hardware status
  • Storage health and performance


In addition, the health check provides recommended actions and specific remediation guidance for any detected issues, helping administrators quickly isolate and resolve the root cause of VM slowness.
Prosi Lv4Posted 2026-Aug-19 12:40
  
Hi,

Agree with the idea that an HCI health check should be performed before delving too deeply into VM troubleshooting. However, this health check should be a "quick check" rather than a time-consuming, in-depth inspection.

The principle I recommend is: "Check infrastructure health first, then isolate the scope of the failure."

Troubleshooting workflow:

1. Verify the impact and scope of the issue
Is only one VM affected?
Are multiple VMs on the same host/node affected?
Is the entire cluster experiencing issues?
Is the problem related to performance, network connectivity, or both?
When did the issue start, and is it intermittent?

2. Quick check of HCI cluster health
Before making changes to the VM, check the cluster status:
Status of all nodes/hosts.
CPU, memory, and disk utilization.
Storage health and latency.
Virtual network status.
HA/cluster services.
HCI alarms/events. The condition of the node running the VM.
The goal is not to immediately conclude that HCI is the cause, but rather to rule out potential issues across the entire infrastructure.

3. Isolate the VM
If the cluster is healthy and other VMs are functioning normally, focus the investigation on that specific VM:
VM status and power state.
CPU/memory allocation.
Disk I/O and latency.
Snapshot/backup activity. vNIC and virtual networking.
Guest OS.

4. Check the network
For connectivity issues:
IP address and subnet mask.
Default gateway.
DNS.
VLAN/port group.
vNIC status.
Guest OS firewall.
Ping the gateway and other VMs.
Ping the problematic VM from another VM.

5. Compare with a normal VM