Hi Peter

Sorry, I'm Japanese and I don't speak well.

I want to confirm user has logon credencial for linkclone/float and it is occured login storme timing, right?

if is it correct, the problem is occured delayed authentication of Horizon environment to ActiveDirectory.

As temporary solution, I suggest you might want to restart every Domain Controller and set restarting to task.

If the problem will not be occured  by temp solution, you wiil add few domain controller as parmanent Solution.

Do not mind your english at all.... glad with your reply :smileygrin:

Will investigate that, but I don't think it's a problem with the response time of the domain controller.

From what I can understand from the reports is as that a user gets a desktop assigned from which an other user has disconnected. It's as if the broker is no longer aware that a user is logged on. That should not have anything to do with domain controller response time.

As a workaround for the problem we are thinking of logging of a user as soon he/she logs off. But being able to reconnect to a disconnected desktop has been a big advantage for users. We would rather not loose this advantage.

Understand your situation.

The problem is not depend on DomainController.

I have one doubt for your environment, please confirm your DB server's OS and check ESX system time/NTP configurations.

I know every Horizon server environment shuld keep same clock.

Broker is Windows and join ActiveDirectory which meaning is force set Broker system time,if the time between Broker and DomainController over 5min.

Linux/ESX does not force set system time by DomainController.

Once I faced User select pool on Horizon client and face connection failure sooner.

the root cause is forgetting ESX NTP configuration and ESX system time goes 5 minutes / unmatch connection server DB log.

Opps

Below is correct

Broker is Windows and join ActiveDirectory which meaning is force set Broker system time by DomainController.

How many connection brokers do you have? 

What version of Horizon are you running and what does your architecture look like? It sounds like your connection servers are out of sync. Try following the resolution in Restart order of the View environment to clear ADLDS (ADAM) synchronization in Horizon View (2068381....

Everything everyone has stated should be investigated as they could be the root cause. Out of curiosity what does your environment look like? Are you using zero clients or are these windows clients made into kiosks with the Horizon Client auto connecting at boot? There is a bug with the Horizon client that causes the behavior you are experiencing but not sure if applicable here without knowing more information.