Why did my restored domain controller BSOD on its first boot?

November 16, 2026 4 answers
Asked by 5u~bracketpencil_263

I restored a physical Windows Server 2022 domain controller to an isolated Hyper-V VM using Veeam Instant Recovery. It initially booted with CRITICAL_SERVICE_FAILED, but after CHKDSK and Windows repairs it starts normally, with AD DS, DNS, and Netlogon running.

The production server's C: volume is marked dirty, and CHKDSK reports corruption under Windows.old and recommends spotfix. Is the source filesystem the likely cause, and would you repair it or replace the DC?

Answers

m.407!tagfunnel

The dirty source volume is a plausible lead, especially since filesystem repair preceded a successful boot. It is not proof that the backup or restore caused the failure, or that the reported Windows.old corruption was what broke the restored VM.

Check the repair logs and any available crash details before settling on a cause. The restore is useful evidence, but it took extra recovery work to get there, so factor that into the recovery time you expect from future tests.

Vexora.4d:

I'd also check whether the backup was application-consistent or only crash-consistent. That could help narrow the investigation, though it wouldn't by itself explain the stop code.

Reply to this answer
8c!foldgrommet_516

If the other DCs are healthy, replacing this one is a reasonable alternative to repairing it in place. Build a new server, promote it, verify replication and services, then demote the questionable DC cleanly.

Check for roles and dependencies first. If a normal demotion isn't possible, the cleanup and any necessary role seizure need their own plan. This is not a zero-risk shortcut, but it avoids relying on a server with known filesystem trouble.

Reply to this answer
Drelva-5p

Repairing the source is also an option. With verified backups and other healthy DCs available, I'd schedule the recommended spotfix in a maintenance window, then check the filesystem and DC health afterward. The corruption is reported under Windows.old, but I wouldn't assume that guarantees the rest of the volume is fine.

Windows.old often points to an OS upgrade, so it's worth checking the server's upgrade history too.

Reply to this answer
8q!ferrypin_503

Don't rule out the physical-to-virtual change yet. CRITICAL_SERVICE_FAILED can have more than one cause, and a boot driver or service issue is another possibility. Look at the crash and repair information rather than assuming the dirty bit explains it.

Also, the driver injection setting mentioned for recovery media isn't necessarily relevant here. This was Instant Recovery, so check the options and logs for that workflow instead.

Reply to this answer

Add your answer

Answers are reviewed before publication.

Related Discussions