Azure Local Storage Spaces Direct - Network Resilience

Question

I need to upgrade the network switch firmware which is servicing the storage extents on my Azure Local cluster, which will mean a reboot. Is the best practice for this to shutdown all VMs and turn the cluster shared volumes offline prior to switch reboot. Or is it OK to put one of the servers in my two node cluster into maintenance mode (will that achieve the same result?)

Answer 1

Hi Chris Davis,

Welcome to the Microsoft Q&A Platform! Thank you for asking your question here.

When you shut down all VMs and turn the cluster shared volumes offline prior to switch reboot, there is complete safety by eliminating the risk of data corruption but there will be a downtime during maintenance.
If you put one of the servers in your two node cluster into maintenance mode, it allows continued operation of VMs and workloads with minimal service interruption, but one node becomes a single point of failure during the maintenance.
According to your environment and specific requirements you can choose one. If you want to ensure safety, shutting down all VMs and taking the Cluster Shared Volumes offline is the most thorough way. If you need to maintain some level of service availability, putting one server into your two node cluster into maintenance mode can achieve a similar result with less downtime.
You can refer this https://learn.microsoft.com/en-us/azure/azure-local/manage/maintain-servers

Please let us know if you have any further queries. If this solution is helpful, please Upvote it.