Proxmox HA Cluster Setup

I need an experienced Proxmox engineer to stand up a 3-to-4-node cluster and tune it for true high availability and fast replication. All hosts will run Proxmox VE on hardware RAID 10 arrays (we are intentionally staying away from ZFS for now). The cluster must serve both Linux and Windows virtual machines and allow live migration without interruption.

Each node has two network interfaces:
• the primary 1 Gb/2.5 Gb port that will remain on the main LAN for user traffic
• a dedicated 10 Gb card reserved exclusively for inter-node traffic and for talking to a Synology NAS. That 10 Gb backplane needs to be the path for corosync traffic, VM replication and storage I/O so bandwidth is never a bottleneck.

Your work will cover the clean installation of Proxmox VE, creation of the cluster, configuration of HA groups, tuning of corosync for the separate 10 Gb network, and seamless integration of the Synology (NFS or iSCSI—whichever you judge most reliable for HA on hardware RAID). Once finished, I expect to be able to pull the power on any single node and watch all running guests automatically restart elsewhere.

Acceptance is straightforward:
1. The cluster shows healthy quorum with all nodes online.
2. Live migration succeeds for both a Linux VM and a Windows VM.
3. A simulated node failure triggers automatic fail-over within the default timeout.
4. Replication traffic and storage I/O stay on the 10 Gb interface, while management remains on the primary NIC.

If you have recent experience delivering similar Proxmox HA setups, let’s get this done.