Many outages are caused by human error. Before performing maintenance on hosts or the network and storage infrastructure on which they depend, take precautions to avoid unintentional downtime.
There are a few steps you can take to avoid unintentional downtime.
- For operations that could impact access to storage, check that you have current backups for the VMs in the vSphere cluster, and if not, take a backup before proceeding. There operations include maintenance on hosts in a vSAN cluster, datastore mount points for external storage, and storage-array LUN masks.
- For operations that may take compute capacity offline, check that there is sufficient capacity available to continue running the VMs if the host being repaired cannot be brought back online as planned.
In addition, if the maintenance involves a host in a vSphere cluster, before proceeding with the maintenance, check the vCenter Sever and NSX Managers associated with the host for any alerts that indicate a problem beyond the one you are planning to fix. If there are any alerts, address them first.
