I have an esx with 100+ vms, in case of a disaster, please suggest a quick way to recover them

[shwetachaudhury]

I have an esx with 100+ vms, in case of a disaster, please suggest a quick way to recover them.

Since in such scenarios, windows machines go to paused state, but unix machines go to read-only mode (Which will need a reboot), please suggest to recover 100+ machines

[a_p_]

Welcome to the Community,

what kind of disaster are you thinking of?

In order to be able to give you an advice, please provide some information about your physical environment (locations, hosts, network, storage, UPS) and current vSphere setup (clusters, HA, ...).

André

[shwetachaudhury]

Disaster as in - usually in case of PSOD, i have noticed HA kicks in but during pause flooding, that doesnt happen.

Using Vsphere 5.5, would like to know a way to bring up the machines as quickly as possible to prevent production impact.

Thanks in advance.

[AdilArif]

There are various options available, like HA and FT.

HA will have downtime whereas FT will not, but that in turn will require double the storage for each VM and you cannot have thin provisioning.

[shwetachaudhury]

true, in case of Pause flooding, HA doesnt kick in, correct, in that case what would be the best option?

Vmotion?

Also, say an ESx went down, now has recovered, any suggestion for a good recovery plan.

Current scenario: say there are 200 Vms under a host, checking each one individually would take around 10 mins, how can this time be reduced.

[TomHowarth]

lower your density of VMs per  host.

[cyberfed2727]

If your ESXi host dies for whatever reason you have a few options for protection.

     1. FT - using FT you will not lose your guest OS however you have to be willing to sacrifice the resources needed to run a FT VM

     2. HA - if your cluster has HA enabled when your ESXi host crashes the guest VM's will be migrated to a functional host and will be powered back up.

VMware itself isn't going to perform a health check of every VM for you. The most it can do is check for VMtools heartbeat and act accordingly. It will be up to you or some other software item to verify that the guest OS's are operational as expected, VMware will just ensure they power back up on a healthy host.