This appears to be random. I do a backup which involves a shutdown, snapshot,
Reboot from Xen Orchestra. The backups are great. When a user tries to login using
That are configd with AD and RDP 5 user. The systems that I have that don’t have multi-user RDP (Remote Desktop) package installed work just fine. Seems to happen with less cpus and Installed on disk. I have others on NVME that work most of the time. This is on Windows server 2016,2019,2022. This is a very common occurrence and it very very annoying. I use the Xen Center and also Xen Orchestra compiled version. I have seen this for about two years. I am using the Citrix mgt agent. Other than that – XCP-ng is amazing. I use mostly R620 dell and the whole thing is rock solid for the most part. I really need to solve this problem – I am getting too many complaints. I am looking for some next steps to try or what log files I can look at. Where to start with this problem. At some point, I would hire someone to solve this problem. I am not a Windows internals guy.
I don’t really understand the issue, does the VM not start or the RDP service fails to start?
The vm is up and I can login to the console. I have tried restarting the RDP service and it still wont let anyone connect. I have to reboot the vm and then it works. So when I do nightly backups, I have to check to see that I can RDP in otherwise I get lots of grief first thing in the morning. It seems like some sort of race condition or configuration problem. I have not been able to find any rhyme or reason to it. I have multiple servers with multiple vms doing this. It is only on systems where the RDP package is setup. Most are set for 5 users. The default RDP is not ever a problem. I will get a screen capture next time it happens so I can document the error message.
I am in thru the console and Xen Center
I hit the big green button for reboot in Xen Center - RDP is now up and going after the reboot.
I guess there is some stuff to look at with Server Manager and Remote desktop services - but I
dont know what I am looking at. 17.10 seems to have this problem every night after the backup. I do the offline snapshot. I have had to resotre backups before and they work great. Xen Orchestra is awesome!
I could run the backup without rebooting - not sure how reliable that is
Try a scheduled task to reboot an hour after you do the backup. Not a proper fix, but it might be an answer.
I tried that - Could not count on this working.
I was also thinking that my VM was created with BIOS not UEFI. This should be just fine
but it seems like I get messages about “saving configuration” on the reboot. I setup a UEFI
VM with Server 2022. It boots in 90 seconds. Seems to be more reliable. I need to do more testing.
With UEFI I cant change screen res or date/time. Dont know why.
There is a UEFI “hack” and I think it was posted in these forums. There is a way to get into the EFI and change the resolution, here is the link XCP-ng VM's display resolution locked - #13 by David-DB
I can’t remember how well this worked, but it did work.
again this is windows server with AD and 10 RDP licenses. If multiple user RDP
is not setup - then it is not an issue.
Here is the service that does not start randomly. When I start it - RDP works again.
This is a script that I run on boot up(task sched) that checks to see if the service is running
if not - it starts it. Works Great!
I was able to do some testing with UEFI VM with Windows 2022 configd for AD and 10 RDP users.
It reboots faster and I have not seen any RDP failures with over 40 reboots. I am going to use UEFI
on new vm 's of Windows 2022 from now on.
This pretty much wraps up this issue