An anonymous reader writes: Recently, an employer has had experience with Red Hat High Availability Add-On in a High Availability Cluster. Previous engineering left much to be desired with the old engineering not even being correctly licensed, CLVM missing basic setup options and more. Eventually, they called in the 'consultant army' to grasp at straws and eventually suggest that a High Availability NFS service can only be made with VMWare Fault Tolerant VMs. As it turns out, VMWare Fault Tolerant VMs have their own problems. NFS will be served from VMWare SCSI translated from Fiber Channel SAN and rely on a mash-up of technician responses or VMWare triggering a restart of the machine in the event of an issue.
Today, the current engineers are in a bit of a tough position. With a single consultant defining NFS service availability being a product of VMWare's HA VMs, the management has decided that the only reasonable direction is to push the NFS and PostgreSQL environment into VMWare.... permanently. This will eventually lead to a maintenance nightmare as the environment must be "Highly Available" — which includes forgoing any patching for uptime. Any advice from fellow slashdotters on explaining the misunderstanding of Guest HA and Service HA we seem to have?