VMware Horizon Community
epa80
Hot Shot
Hot Shot

Instant Clones and DST Change Issues

We run into this twice a year for the time changes. We're linked clone with refresh on logoff enabled, so, we always end up having to recompose (after disabling refresh so users don't feel it). We have a few Instant Clone pools now though, and was wondering if they feel the same thing, or if it doesn't happen on them.

VMware Knowledge Base

Anyone using ICs and not feeling it? ICs are a bit more susceptible if only because there's no refresh on logoff to disable.

0 Kudos
4 Replies
Hotrod76
Enthusiast
Enthusiast

I've deployed instant clones for several customers and I've not run across this issue.  Thank you for pointing this out.  The next time I deploy an instant clone pool for a customer I will check the event logs for time synch issues.

0 Kudos
jeffrobinson85
Enthusiast
Enthusiast

How are you handling NTP on your domain, vCenter and ESXI hosts? I've been support my companies Horizon environment for 5 years. Originally with Linked clones and now using Instant clones. I've not ran into this issue. Our previous BCS engineer would forward us the same KB in case we saw provisioning issues, but we never have. I would be curious if there is a sync issue during daylight saving time with your domain controllers and the VM's that causes the issue. Happy to hear you have been able to workaround this issue when it has happened.

0 Kudos
Mach6
Contributor
Contributor

I have the same issues as well, twice a year. Computer GPOs not applying when the user logs on.

It started in 2012 with Windows XP on a older version of View.

We then moved on to Windows 7 on View 6.2.0 (still up and running to this day) and we are still experiencing issues. We managed to find a way to bypass this issue by installing the latest Windows updates and recomposing the pools a couple days before the time change.

We also have another environment on View 7.3.2 with Windows 7 and Windows 10 linked-clones pools.

We never had an issue with Windows 7 machines running on 7.3.2 for some reason, never had to recompose after DST.

But we seem to have this issue running Windows 10 (1607) on this same Horizon View environment 7.3.2. We have these pools up and running for the last 4 months.

I noticed this error in the event logs of these machines, Security-Kerberos Event id 5:

KERBEVT_KRB_AP_ERR_TKT_NYV

The Kerberos client received a KRB_AP_ERR_TKT_NYV error from the server %1. This indicates that the ticket used against that server is not yet valid (in relation to that server time). Contact your system administrator to make sure the client and server times are in sync, and that the KDC in realm %2 is in sync with the KDC in the client realm.

By the time the user logs in to the machine, the time has already been updated. But from what I can see, the Computer GPOs are trying to apply before the service has time to synchronize with the DC in the domain.

I wish I knew what made our Windows 7 running on Horizon 7.3.2, not have this issue at all, whether we update it or not after DST. All our templates are configured in a workgroup. These VMs are then joined to the domain with Quickprep (linked-clones). We used Fling to optimize our templates.

For Instant clones, we had too many issues at the time with AppVolumes and our antivirus solution, that we reverted back to Linked-Clones. We never had the chance to test it out after DST.

0 Kudos
Claudio_R
Enthusiast
Enthusiast

Hi, is there any news on this issue?
We with Pool Linked Clones even with Horizon 7.13 and Windows 10 have always encountered the problem and need to do a recompose at each time change.
Do you know if with Instant Clone the problem is solved or not?

Thank You

VCP DCV - DTM since 2009
0 Kudos