RDS with HA Connection Broker stops working

Question

I've been fighting this issue for so long and someone else must have this problem!

Currently we have a highly available connection broker on two servers. Then we have multiple session collections with session servers. We are not doing session collection load balancing, it's single session servers. I have this setup in two different domains one domain has server 2016 and the other domain is server 2019. On both I have an issue where users cannot connect to RDS. If I restart the Connection Broker service on both servers then they can connect again. I don't want to just create a scrip that restarts this service all the time - I want to fix the issue permanently.

The error message they receive is sometimes different. Usually they get "There are no available computers in the pool. Try connecting again or contact your network administrator". In the event log I receive the following:

Event ID: 824 "RD Connection Broker could not find a server for user USERNAME in collection COLLECTIONNAME (Error = 0x80070005), please ensure user is assigned to a server, and server is avaiable and running"

Event ID: 802 "RD Connection Broker failed to process the connection request for user USERNAME
Load Balancing failed OR Specified endpoint could not be found.
Error: Access is denied."

Other times I get the following Event ID's:

Event 1296 "Remote Desktop Connection Broker Client failed while getting redirection packet from Connection Broker.
User : USERNAME
Error: Element not found."

Event ID: 1306 "Remote Desktop Connection Broker Client failed to redirect the user USERNAME.
Error: NULL"

I feel like I have read every article online that exists with no answer. I have used logging to check the firewall isn't blocking anything on either side. I'm not receiving any errors connecting to the database. One domain has the databases hosted on a sql server in the domain the other one is using Azure and they both have the same issue so I'm pretty sure it's not the DB.

Any tips would be greatly appreciated.

Answer 1

Hi,

If I restart the Connection Broker service on both servers then they can connect again.
1.What was the service's status when users failed to remote to the session host server?

When issue occurred, kindly logged into the CB, check the task manager and verify if any resource or service got stuck without responding.

2.How did the users start the remote session? Via mstsc or Rdweb? Was the RDP failure occurred with all remoting methods?

Looking for your update.

Best Regards,
Jenny

Answer 2

Hi Bailey,

It looks like we are having the same problem on one of our connection broker servers.
Did you manage to solve te problem?

Beste Regards,

Hilco

Answer 3

Hi Bailey,

today morning I noticed the same problem.
Did you manage to solve the issue ?

Best regards
Tomasz Rosikoń

Answer 4

Hi,

I have the same issue CB based on server 2019

The second CB works properly but the first server is facing the issue

Best Regards,

Ahmed

Answer 5

Hi All, I have the same issue, and I think it's because CB not write data to SQL databases but to Windows Internal Database. I've try to find out why this is happening but without success.