Whilst many organisations will have strict policies regarding patching, others will not and I’ve lost count of the number of times I’ve found myself troubleshooting strange errors in a virtual machine, only to find that the underlying host operating system has automatically updated itself and is waiting for a restart. Consequently, it’s worth mentioning that automatic updates and hosted virtualisation server products (e.g. Microsoft Virtual Server or VMware Server) do not mix well. Of course, those running a non-hosted virtualisation solution (like VMware ESX server) won’t have this issue; although even ESX needs patching from time to time.
2 easy ways we get around this. We set a policy on your host systems to one download the updates, but not install, then install this at a time when convienient. The other thing we have in effect on all our Virtual Hosts is shutdown scripts. If the Host for some reason goes through a shutdown proceedure that we initiate or one that some software initiates, we have a script that looks at all running virtual guest systems and puts them into saved state before the Host shutsdown. This creates a log file on the system of the saved machines. We then have the inverse of that (Startup script) too which as the machines starts up, looks at the log file and starts up only those guest systems that were previously saved during the shutdown.
HI Rob,
Could you please share the shutdown and startup script with us?
With warm regards,
Krish