DC problem

[NetworkServices]

Good morning EduGeekers,

Did anybody experience any issues with Server 2022 domain controllers after the March update (KB5078766)? Our three domain controllers installed this update and rebooted as scheduled on Sunday. However when I examined them on Monday morning, all three of them were constantly sitting at the "spinning dots" part of booting. The servers appear to be actually working (replicating/servicing clients etc) behind the scenes and from an end user perspective there appears to be nothing wrong. If we try to RDP into any of them we can authenticate but just receive a black screen so really only can do anything from a remote command line. Rebooting hasn't made any difference. Seems very strange to me that this affected all domain controllers yet the same update installed on member servers didn't cause an issue. 

 

Any ideas on anything I could try? They exhibit the same behaviour when booting into safe mode too which is odd to say the least.

 

Many thanks.

[andy_nic]

Did one yesterday and one today, both rebooted fine no issues. Are they VMs or Physicals? 

[NetworkServices]

Two physicals and one VM. To be honest I think I may be on to a solution and have now got one of the DCs back again. It probably wasn't the update that caused it but a local security policy balls-up. The update and subsequent reboot caused the symptoms to appear so I may have unfairly blamed the update when in fact it was all my own fault!

 

Just need a window of opportunity to reboot the other two to see if they come back after the policy correction.

[NetworkServices]

Yep fixed. At ease, everyone.

[HolographicNerd]

@NetworkServices What was your fix that you had to do? We encountered a similar scenario in our environment and our tech disabled the Windows update +WSUS service+ Cryptographic service+Removed the last patch  in safe mode and finally DC server is back online 

[NetworkServices]

It was one of the Local Policies/User Rights Assignment. Unfortunately I can't remember which one it was. We'd been stripping back policies that had been in place since the domain was 2003 and many were no longer applicable or recommended. Whilst most of them reverted to local defaults when the policy was changed to not configured, one of them reverted to nothing applied at all.