As you probably know vCenter Server provides a centralized platform for managing your vSphere environments, allowing you to automate and deliver a virtual infrastructure. I’d say it’s the core of every VMware infrastructure.
Starting with vCenter 6.5 you have vCenter High Availability, where a three-node vCenter Server Cluster (Active, Passive, and Witness node) is deployed. I’m not going to cover this topic in this post.
What I’m going to talk is about a question that I often get and I saw it also on VMTN forums, what happens when vCenter is down? What happens with HA, DRS, FT, vDS?
First of all, if vCenter becomes unavailable, all of your VMs will still run, but you will not be able to make any changes to the cluster configuration. If you need to make any change to a VM, you will have to log individually on each ESXi host and do it from there.
What about High Availability (HA)? In case of ESXi host is down, HA will continue to restart VMs that were powered on before vCenter became unavailable. vCenter is responsible for the protection and unprotection of VMs, so any new powered VM won’t be protected by HA until vCenter is back online.
DRS (Distributed Resource scheduler) will not work, because moving the VMs between hosts is a vCenter function.
Fault Tolerance (FT) will continue to work for the configured VMs, but in case of a failover no new secondary will be created.
All the VMs that are connected to a Distributed Switch (vDS) will continue to have network access, but you will not be able to make any changes to vDS configuration.
What about High Availability (HA)? In case of ESXi host is down, HA will continue to restart VMs that were powered on before vCenter became unavailable. vCenter is responsible for the protection and unprotection of VMs, so any new powered VM won’t be protected by HA until vCenter is back online.
>> How HA will decide which host in the cluster has enough resources and to which host VM’s needs to migrate?
DRS will be involved for rebalancing the cluster
“DRS (Distributed Resource scheduler) will not work, because moving the VMs between hosts is a vCenter function.”
>>could you explain here relate to the above Swapnil question,
If the vCenter server goes down with a situation that it was pre-configured with vSphere HA and DRS , how DRS can rebalance the cluster where DRS will not function?
I assumed that Swapnil question was not related to when the vCenter is down. If vCenter is down, DRS is not working.
Thank you so much for a clear picture of “if vCenter goes down”.