Our current Nutanix infrastructure has been a bit neglected for a while, AOS, PC, NCC, etc have been kept up to date but not firmware\bios or hypervisor so I need to get these done.
Current infrastructure is nothing massive, two clusters one 4 node one 3 running on NX-8035-G6 and NX-8035-G7, both are AOS 5.20.4. We have one PC running at the latest version and its sitting on a separate Management vSphere cluster along with vCenter. Hypervisor is currently ESXi 7.0 Update 2a 17867351 on all hosts, this is Standard Edition so NO DRS ☹ My plan is to upgrade the firmware and BIOS etc via the LCM on the smaller of the two clusters. I have freed up one host so there aren’t any VM’s on it apart from the CVM. I take it I can just run the LCM to do the firmware updates on that host and let it do its thing without having to do anything else, it’ll shut down the CVM maintenance mode the host and restart it?
Once that is complete I want to upgrade ESX on that host. I have checked the compatibility matrix and I can upgrade to 7.0 U3c as that is certified for our current AOS 5.20.4 and for AOS 18.104.22.168 (which I want to upgrade to in the new year) so the question is what’s the best way to do this without having DRS! Can I use vCenter LCM for this? Download the 7.0U3c ISO and attach that as a baseline to the host, like I would do on our non-nutanix hosts? Put the host in to Maintenance mode SSH to the CVM and run ncli cluster get-domain-fault-tolerance-status type=node and then shut the cvm down with cvm_shutdown -P now once the CVM is shut down REMEDIATE the attached update baseline from vCenter. Once all is back up move VM’s back to the host and move on to the next one?
Is this the best way to do it and am I missing anything or any steps or commands?
Once the smaller cluster is fully up to date firmware and hypervisor wise I want to upgrade it to AOS 22.214.171.124 and then move on to the larger cluster
Best answer by Moustafa HindawiView original
Please do it automatically through Prism Central LCM :
it’s not really the firmware updates thats the issue, I’ll just use the LCM for that but rather the best way to do the ESX hypervisor upgrade
LCM supports ESXi upgrade. For Upgrade esxi from PE: https://portal.nutanix.com/page/documents/details?targetId=vSphere-Admin6-AOS-v6_5:upg-cluster-hypervisor-upgrade-esx-pe-t.html
If you have not enabled DRS in your environment and want to upgrade the ESXi host, you must upgrade the ESXi host manually: https://portal.nutanix.com/page/documents/details?targetId=vSphere-Admin6-AOS-v6_5:vsp-node-upgrade-manual-vsphere-c.html
ok basically that article says follow VMWare upgrade procedures once you have issued the commands to correctly shutdown the CVM on the host you’re upgrading ESX on.
So I’m going to go with what I had planned and use vCenter updates to update the host once VM’s have been migrated and the CVM shut down