I have a relatively small environment of linked clone instances -- when users log into a desktop there is usually about a 20% chance that their persona/profile will fail to load and they will be given a blank local profile. Usually logging out and acquiring a new instance will resolve the issue, but sometimes it can take as many as 5-6 tries before it works.
In the Persona logs (VMWVvp.txt) on a failed machine, shortly after the logon sequence begins, I see the following errors:
[04/15/2014 09:36:36:0715][RTOGetStringSetting ] Failed Retrieving CentralProfileStore From Default Key: 0x2, Using Default [04/15/2014 09:36:36:0715][RTOGetViewSettingsFromRegistry] Failed Reading CentralProfileStore Value: 0x80070002 [04/15/2014 09:36:36:0715][RTOGetULONGSetting ] Failed Retrieving CentralProfileOverride From Default Key: 0x2, Using Default [04/15/2014 09:36:36:0715][RTOGetViewSettingsFromRegistry] Failed Reading CentralProfileOverride Value: 0x80070002 [04/15/2014 09:36:36:0715][RTOGetVMwareSettings ] Retrieved View Settings From Registry [04/15/2014 09:36:36:0715][RTOProcessLogon ] Retrieved VMware Settings: User: user, Remote Path: , Enabled: 1, UseAltPath: 0 [04/15/2014 09:36:36:0715][RTOGetMachineProfilePath ] Failed Reading MachineProfilePath, 0x80070002 [04/15/2014 09:36:36:0824][RTOGetUserADProfilePath ] DC Name: \\DC02 [04/15/2014 09:36:36:0824][RTOGetUserADProfilePath ] No Remote Path: User: user [04/15/2014 09:36:36:0824][RTOProcessLogon ] No Valid Profile Path Found, Processing As Local User: User: user
I have been digging around and haven't found any useful information about these errors, or what could be causing them, and VMware Support keeps telling me to check my NTP/time settings, however all machines are in sync when this occurs.
Any ideas of what to check next would be greatly appreciated since this has been going on for months with no resolution.
EDIT: Added clarification, i have checked http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2013263
These settings are all correct, and have been verified.
EDIT 2: So there are definitely signs that something is wrongwith NTP, however NTP is set up correctly on the master image, it seems almost like during composition of fresh clones it sometimes doesn't sync time correctly on final boot. I am not sure where to look next -- composer logs don't seem to point anything out, there aren't any significant events in vcenter or vcenter logs, VMware tools are out of date, but this has been happening since implementation (even when vmware tools were up to date).