VMware ESX crash

[manishsawant]

Hi,

I am in a process of identifying potential threats in case we procure VMware ESX for virtualization of our critical servers. First treat that comes in my mind is in case ESX crash then how soon i will be able to recover it or how can i avoid ESX crash.

Also please tell me incase if there are any other threats for using virtualization.

Regards,

Manish

[oreeh]

FYI: this thread has been moved to the Enterprise Strategy and Planning forum.

Oliver Reeh[/i]

[VMware Communities User Moderator|http://communities.vmware.com/docs/DOC-2444][/i]

[christianZ]

For critical servers (vms) you should get 2 (or more) Esx hosts with a shared storage (incl. redundant controllers) - this way you can restart your vm on a second Esx host or you purchase HA (included in "standard", as I remember) and that will occur automatically when you wish

[Texiwill]

Hello,

VMware HA will do two things. If the Host crashes it will reboot a VM onto another host.....

If the VM crashes, you have enabled VM monitoring, will reboot the VM onto another host..... So basically the time it takes the VM to reboot + up to 60 or so seconds usually.

If you do not enable VM monitoring nothing will happen until the host crashes.... This a new to ESX v3.5.

If it was me I would download the Enterprise evaluation and run it through its paces for all your failure scenarios. I think you will find it is fairly robust.

Best regards,
Edward L. Haletky
VMware Communities User Moderator, VMware vExpert 2009
====
Author of the book 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', Copyright 2008 Pearson Education.
Blue Gears and SearchVMware Pro Blogs -- Top Virtualization Security Links -- Virtualization Security Round Table Podcast

[rt7500]

Manish,

You really have two questions that need to be looked at very closely.

1. Crash recovery - As others have already stated HA is your best bet. But you need review the requirements and be aware of the little things that can get you. Secondary Service Console and fault isolation are two that come to mind. There are many posts about these items.

2. Crash avoidance - This is very difficult to answer. Review the HCLs from both VMware and your hardware vendors. Set your standards and audit for drift. Just because your standards are to set your SAN switches to 4GB does not mean that they will always be deployed that way. I have found this drift to be root cause for most of my issues.

I don't feel that virtualization is a threat. In fact it is the complete opposite. I can maintain firmware/bios/patchlevels much easier in a virtual infrastructure. Replace failed hardware and perform hardware maintenance all with out impact to the VM and the services they support.

Hope this helps.