Windows 2016 -RDS user get temp profile - Error Code 0x4005.135 then 0x20.135

Question

Hello,

I have a strange and random problem with only one different user per day on RDS over 100 active users.

We use Windows Server 2016.

To explain :

• User connect on our RDS farm through a broker
• For the user that have the problem, we see that his profile is loading but at the end it finished with a temp profile
• It happens only once a day, only for one user and with different server from the farm.
• udp uses vhdx
• problem since august friday 27th

Searching on event viewer, I've found that at the first openning session from the user who got the problem that there is an error with Microsoft-Windows-TerminalServices-RemoteConnectionManager/Admin - Error code 0x4005.135

Then, on next try, errors are Microsoft-Windows-TerminalServices-RemoteConnectionManager/Admin - Error code 0x20.135

I've seen over Microsoft Community that I had to delete "***.bak" registry on ProfileList on Regedit, even the ProfilGuid registry but it doesn't work.

I've tried to move or rename the udp disk but I can't because it says that it's actually locked by System.

I've search on all RDS server and the udp disk is not mounted.

It seems that rebooting the server broker temporary correct the problem.... but only temporary.

Thanks for your help.

Answer 1

After having the same issue, however we do not use the same Panda AV software. Even removing the profile from the registry and directories on our 2016 RDS server, we would get a temporary profile every time. I have found the following if you're using User Profile Disks, as apart of your RDS Setup. Make sure if you're including files or directories as apart of your RDS User Disks, that such locations or files listed/specified are valid. You can test by stripping everything out of the settings or turning off user profile disks. Referring to the following settings:

  One of my above ntuser.dat.log1 files was set as a folder path. When the user profile is being created, it would fail during the process and give a temporary profile every time. Also showing access denied errors in the event viewer (ID Numbers, 1509,1511,1515). After removing the invalid entry from the user profile disk configuration page, it now is writing the user disk to the directory and more importantly, the user profile to the specified location via the RDS Profile Location GPO. Whereas before it would create a temporary folder under C:\Users
Hope that helps someone in the future!    

Answer 2

Hi all,

We are currently experiencing the exact same issue, also terminal server 2016 with user profile disks configuration and Watchguard (Panda) EPP.

Has any of you found a solution yet?

Answer 3

Hello,

Does any one of you have found a solution ?
We didn't use the low level hotfix.

I've build a new RDS Collection et start to migrate users on it.
Il worked great for 2 month and now it does the same error...

TY

Answer 4

Hi,
we maybe got a litte further.
There was a hotfix provided by panda. (hotfix_lowleveld_gui_v1.6.exe)
We will see how it worked out.

Answer 5

Hello!

It is clear then that it is Panda AV.

In our case, the affected servers have:
Panda Adaptive Defense 360 ​​8.0.19
Panda Endpoint Agent 1.18.00

In several servers we found a warning in application viewer ID: 201 related to Aether Agent right at the beginning of the problems.

Some days ago, we tried disabling the scanning of files with the vhd and vhdx extensions without success.

One of the affected servers has been running smoothly and without requiring a reboot for 15 days after switching from Panda to Sophos.

Another server with Eset has not reported errors either.