Hi All,
I have an application in my environment that is very sensitive to time drifts; the environment is nearly 100% virtualized (including domain controllers) and is set-up to use windows time services (domain time source is the DCs which are sourcing time from a public NTP source). I have VMware tools installed on all the guests (time synchronization between the guest and ESX boxes is not checked). We have run into a few issues where our domain controllers have lost connection to the public NTP sources and my colleagues claim that all guests in the virtual environment drift rapidly (2 mins a day). So my question is has anyone else experienced this in their environment; I am mainly trying to get a consensus on if the guests loos access to the NTP source would they start relying on the ESX host hardware clock even though the tools are not set to sync time with the hosts.
I also would like to know how I could definitively prove or disprove that the ESX hosts are affecting the Windows guest time drift. I have done some searching on this and found the following post regarding PPM in windows boxes and time drift:
I have also discovered that there is an ntp.drift file on the ESX hosts (under /etc) if I vi this file it gives me a value but I have no idea what this represents.
PS- all my hosts are running ESXi 5.0.0 build 821926.
Thanks in advance for all your help.