Random disconnects terminalserver (RDP)

Question

Hi everyone,

we have some errors regarding our terminalserver. Maybe someone could help me with that.
Our users get randomly kicked out, but can reconnect after a few seconds. It happens 1-4 times per day and not every user gets kicked out at the same time.
There are about 15-20 users active and 3-4 random users get a disconnect at exact the same second. All other users are not affected.

Our environment:
2 x Microsoft Windows Datacenter 2022 (virtual)
2 x Kemp Loadmaster (Loadbalancing the incoming traffic of port 3389)
Each server has 20 cores (Intel(R) Xeon(R) Gold 6234 CPU @ 3.30GHz) and 48 GB ram. They never get above 50%.

We never had any problems with RDP, loadmaster or something like this, but round about a year ago the users started telling us about the problems. We don't know what happened.
A few month ago I even reinstalled both terminalservers (before it was Server 2019, now 2022), but the problem is still the same.

There are a few event logs, but I cannot find anything interesting that helps.

ProviderName: Microsoft-Windows-TerminalServices-LocalSessionManager
TimeCreated Id LevelDisplayName Message

---

22.07.2024 06:53:38 40 Informationen Sitzung "6" wurde getrennt. Ursachencode: 2147942521
22.07.2024 06:53:38 40 Informationen Sitzung "4" wurde getrennt. Ursachencode: 2147942521
22.07.2024 06:53:38 40 Informationen Sitzung "2" wurde getrennt. Ursachencode: 2147942521

"ID 40" does not mean licensing error. This would not be located at "Microsoft-Windows-TerminalServices-LocalSessionManager".
Sometimes I see the code 2147942464, instead of 2147942521.
2147942521 in hex is 0x80070079
2147942464 in hex is 0x80070040

Has anybody the same issues?

Thank you.
Best regards,
Chris

Answer 1

Hello,

Intermittent disconnections where user sessions are able to reconnect in a short amount of time can be caused by a variety of issues, including network instability, configuration issues with the load balancer, or even problems with the RDP server itself.

Analyze the error code you provided:

0x80070079: Semaphore timeout period has expired. It usually indicates a network timeout problem.

0x80070040: The specified network name is no longer available. This error indicates that a network resource or server may be temporarily unavailable.

You can try the following steps to troubleshoot the problem:

1. Make sure the RDP server and load balancer are up-to-date.
2. Ensure that the network between users and the load balancer and between the load balancer and the RDP server is stable. Temporary network failures or instability can cause these timeouts. We can use network monitoring tools to check for any anomalies or interruptions associated with disconnection.
3. Further review the event logs on the client and server for any patterns associated with disconnections, not only from the TerminalServices-LocalSessionManager provider, but also system and application logs. Please enable the options below before viewing. 

Also check the logs from network devices (such as switches or firewalls) between the user and the server, especially the logs from Kemp LoadMaster devices.

4. If possible, try to temporarily bypass the load balancer and connect some users directly to an RDP server. This helps narrow down whether the problem is on the load balancer or the RDP server.

I hope this helps. Please let me know if you require further assistance or have any questions.

Best regards

Jacen