HPE VM Essentials Setup – Storage and Network Configuration Issues

[abhivintech]

Dear Team,

I am writing to seek your guidance regarding two issues we are currently facing with our HPE VM Essentials setup.

---

Issue 1: iSCSI Storage Pool Configuration

We have configured a 2-host VME cluster, with all hosts connected to a shared iSCSI storage pool of 100 GB. The iSCSI initiator and target setup has been successfully completed, and the shared storage is visible from all three hosts.

However, when we attempt to configure a GFS2 shared storage pool through the HPE VM Manager dashboard under the "Cluster Storage Pools" section, we are unable to proceed due to an error message indicating a minimum requirement of 3 nodes.  the system does not allow us to create the GFS2 pool..

We would appreciate your advice on the following:

• Is there an alternative method to configure the iSCSI storage pool in this 2 Host setup?
• Are there any known workarounds or specific steps required to enable GFS2 pool creation when using iSCSI with HPE VM Essentials?

---

Issue 2: Bridge Network Configuration on Host 2

We are also experiencing a VM-network issue on Host 2. When we create a virtual machine (VM) on Host 2, the VM fails to receive an IP address from the bridge network .

In contrast, the same VM created on Host 1 receives an IP address from the mgmt bridge  and the VM full internet access.

This raises the question: Is the reason the VM on Host 1 receives the correct IP and internet access because the HPE VM Manager is running on Host 1 and has the bridge network properly configured, while Host 2 may not have the same bridge setup or configuration applied?

We would appreciate your guidance in understanding whether the VM Manager being hosted on Host 1 affects how network settings are applied to VMs created on other hosts, and how we can ensure consistent bridge network behavior across all hosts in the cluster.

[Arnout Verbeken]

Issue 1:  You need at least 3 hosts to create a GF2 pool.

Issue 2: Make sure your hosts are configured identically.  Check/compare the netplan on all hosts and reapply if needed.   Seems host 1 is configured correctly so verify that the netplan on host2 is identical (with different IP's of course).

netplan get
ll /etc/netplan
nano "the yml-file(s)"
netplan apply

Also check the upstream switch ports to make sure they are configured correctly.  It's general advise since you did not gave a lot of details about your network:  Seperate compute vlans?  tagged vlans?  Deploy the VM on management network?

[abhivintech]

Regarding Issue 2, I’ve reviewed the network configurations on both hosts and can confirm that they are functionally aligned. However, there are some differences in how the mgmt bridge was configured on each host, which might be worth pointing out for clarity and consistency.

Host 1 (hpe-vm1)

• mgmt bridge IP: 172.16.30.114/24

• Configuration was auto-generated by the HPEVME dashboard.

• Uses ens10f0 as the bridge interface.

• Gateway configured via a route block.

# /etc/netplan/60-mvm-mgmt.yaml network: bridges: mgmt: addresses: - 172.16.30.114/24 interfaces: - ens10f0 nameservers: search: [8.8.8.8] openvswitch: {} routes: - to: default via: 172.16.30.1 version: 2

Host 2 (hpe-vme2)

• mgmt bridge IP: 172.16.30.147/24

• Manually configured using Netplan.

• Uses ens22f0 as the bridge interface.

• Gateway is configured using gateway4 directly.

• Additional bridge parameters (e.g. stp, forward-delay) are specified.

# /etc/netplan/60-mvm-mgmt.yaml network: bridges: mgmt: addresses: - 172.16.30.147/24 gateway4: 172.16.30.1 interfaces: - ens22f0 nameservers: addresses: [8.8.8.8] openvswitch: {} parameters: forward-delay: 0 stp: false version: 2

Notes:

• No VLANs are in use on either host.

• Both servers are connected to the same physical LAN (no tagging, just plain Ethernet).

• Both have proper IP addressing on the mgmt bridge, and can communicate across the same subnet.

I have double-checked the configurations and applied the Netplan changes using netplan apply. The interfaces come up properly (state UP), and there’s no indication of misconfiguration at the OS level.

I’ll also verify the upstream switch ports to ensure there’s no port-specific issue (e.g., blocking, tagging, etc.).

Let me know if you’d like me to align the gateway configuration style or clean up any interface declarations for consistency.