Standby option server time issue

To clarify, what system are you experiencing the time change? Is the time in Windows on the production server changing? Or is the time in Windows on the standby VM and that does not have the correct time on it? Or are we talking about a different time stamp altogether.

When an export is done (which is the virtual standby) it does not touch the production machine, it pulls the RP from the repo and uses that data to update the standby VM. The backup touches the production machine/VM, not the virtual standby.

I'm just trying to wrap my head around what you're describing.

To add to what Paul said, The date/time of the standby VM is the (or should be) the date/time of the recovery point (and updated AFTER you power on the VM). On the other hand the updated Date/Time timestamp on the files reflects whatever changes have been performed during the export process and possibly when they were last accessed by the Host.

I'm having issues in the original VM the prodution, not the backuped one.
I'm not complete sure the issue is the rapidrecovey, but is the only think running from hour to hour to that machine...

I see so the source VM changes time -- what are you using for NTP? Are you sync-ing with the domain controller or with a public NTP server? Could it be that your VM is set for a time zone that is 30 min away from the nearest top of the hour one? The list of those time zones is here: www.timeanddate.com/.../time-zones-interesting.html

So the time on the production machine is changing when you take a backup? If so, check your time server settings for the ESXi hosts and for the domain, one or both are probably off. Which is why pointing to something like this for both domain and VMware is a good idea: 

 

http://www.pool.ntp.org/en/

Yes i have the windows VM sync with domain controller, is because of that the after some time the time back to normal.
I confirmed and the w32tm have domain controller as peers and eveything appers to be ok... i really have no clue what is the problem... i have more windows VM in the same esxi host and the only difference is that all the others are configured in rapidrecovery as export one time a day, and that VM that is showing problems is the only with standby option and the backup is every hour

Maybe the problem occurs with others VMs, but the backup is schedule to 02:00AM i really cannot tell if that happens... because if it happens after some time it will get right .. and in the morning everything is ok...

Ok i can know confirm the problem is with rapidrecovery, i force a snapshot in other Vm in that moment the new VM change the time ...

ESXI host time -> 08:05
VM Machine time -> 08:05
RapidRecover time -> 08:05

As soon the backup starts

ESXI host time -> 08:05
VM Machine time -> 09:05
RapidRecover time -> 08:05

If we are talking about agent-less backups for VMware then actually RR is not touching the VM itself during the backup process. The api call is made to either the VC or the esxi host to take a quiesced snapshot of the VM, which VMware will then engage. Thus, if you are experiencing this time delay on the local OS during the process of taking and closing a snapshot you should be able to duplicate this simply through vSphere outside of RR. What happens if you go into vSphere and take a quiesced snapshot of the VM outside of RR entirely? Same before. Be absolutely sure however you are doing a quiesced snapshot, as that is what we are requesting by default from the RR console (unless you disabled it through our GUI). I'd venture to say that the same behavior presents itself, if only due to the fact that our core doesn't actually 'touch' the VMs themselves.