Problem with RDS farm and user profile disk

Question

Hello, I've already searched the internet, but unfortunately didn't find a successful solution to my problem.

Our customer uses a Windows Server 2022 terminal server farm with broker. On the broker I have activated the user profile disks. After logging on to the terminal server, the data medium is connected successfully and people can work normally.

Now, however, the phenomenon always occurs that after logging off the data medium is opened, but profile remnants remain locally. As long as the profile remnants remain, the profile data carrier will always be provided with the extension .000 the next time you log on.

When this happens, however, data is missing and the phone software does not work properly. Only after the local profile remnants have been deleted and the user has registered again does everything work as usual.

Presumably, the user profile disk is sometimes ejected too early, even though processes are still running in the background. How can I fix this? Does somebody has any idea? We wanted to try a logoff script that pushes a timeout first, but unfortunately it doesn't work on the farm, only if you connect directly to it and don't go through the broker.

Best regards

Tina Grundman

pictures.pdf

ich habe jetzt schon das Internet durchforstet, nur leider keine erfolgreiche Lösung zu meinem Problem gefunden.

Bei unserem Kunden ist eine Windows Server 2022 Terminalserverfarm mit Broker im Einsatz.

Auf dem Broker habe ich die Benutzerprofildatenträger aktiviert.

Nach Anmeldung auf dem Terminalserver wird der Datenträger erfolgreich verbunden und die Leute können ganz normal arbeiten.

Jetzt tritt allerdings immer mal das Phänomen auf, dass nach dem Abmelden der Datenträger zwar aufgeworfen wird, aber Profilreste lokal verbleiben.

Solange die Profilreste verbleiben, wird bei der nächsten Anmeldung der Profildatenträger immer mit dem Zusatz .00 versehen.

Wenn dies geschieht, fehlen aber Daten und die Telefonsoftware arbeitet nicht richtig.

Erst nachdem man die lokalen Profilreste gelöscht hat und der User sich neu angemeldet hat, geht alles wie gewohnt.

Vermutlich wird der Benutzerprofildatenträger manchmal schon zu zeitig ausgeworfen, obwohl im Hintergrund noch Prozesse laufen.

Wie kann ich diesen Fehler beheben? Hat jemand eine Idee?

Mit einem Abmeldeskript der einen Timeout pusht wollten wir es erst versuchen, aber leider greift es nicht auf der Farm, sondern nur, wenn man sich direkt drauf verbindet und nicht über den Broker geht.

Answer 1

Hello there,

For multiple and often unknown reasons, users are connected to their session on a temporary profile.

The user does not find his settings including the Outlook profile. All that is stored in the folder C:\Users\TEMP is not saved when they log off. If you apply a folder redirection GPO (desktop/document), the files and folders modified in it will be saved instead.

The solution is to delete.BAK registry keys in the following location: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\ProfileList

Profiles are listed using objectSIDs and you will need to identify and delete any objectSIDs that have a .bak suffix. Each objectSID profile has an expandable string value called ProfileImagePath. Use this string value to easily identify individual users, for example:

Value Name: ProfileImagePath

Value Data: C:\Users\BGates

Finally, manually delete any temporary profiles located in c:\users\

You can avoid the issue by creating an automatic script using a scheduled task. Or set GPO policy to log off disconnected sessions after X minutes. This will reduce the likelihood that UPDs will remain connected if the server is restarted or loses power.

Additionally, ensure all user sessions are logged off and do not allow new connections to hosts before performing maintenance and/or restarts.

Hope this resolves your Query !!

--If the reply is helpful, please Upvote and Accept it as an answer–