DFSR Event 2213 is the warning that quietly breaks Group Policy on a domain controller. After an unexpected shutdown, a hung VM or a hard reset during patching, the DFS Replication service finds that its database was not closed cleanly, and instead of fixing itself it stops replicating the whole volume, SYSVOL included. New GPOs and logon scripts stop reaching that DC, and nothing obvious tells you why. The event text even hands you the fix, a wmic one-liner. The catch in late 2026 is that WMIC is gone from current Windows 11 builds, so the copy-paste from your admin workstation often fails. Below is how I handle it now: confirm it really is 2213, back up SYSVOL, resume replication with PowerShell and CIM, and check that the DC is healthy again.
What DFSR Event 2213 Actually Means
DFS Replication keeps a JET database on every volume that hosts a replicated folder. If the service or the server goes down hard, that database is marked dirty. Since hotfix 2663685 for Windows Server 2008 R2, and by default on Windows Server 2012 according to Microsoft, DFSR no longer runs automatic recovery in that case. It pauses replication on the volume and logs DFSR Event 2213 in the DFS Replication event log, asking an administrator to step in. Microsoft explains the reasoning in its dirty shutdown recovery deep dive: recovery treats the local copy as non-authoritative, so local files can be moved to the PreExisting or ConflictAndDeleted folders. The pause gives you a chance to take a backup first.

On a domain controller the replicated folder that matters is SYSVOL Share. While it is paused, that DC serves stale policies and scripts, and if it was the one that took your latest GPO edit, the other DCs never get it. A quick list of the events you will see around this problem:
- 2212: DFSR detected an unexpected shutdown of its database.
- 2213: replication on the volume is stopped and waits for ResumeReplication.
- 2214: the service finished recovery on the volume. This is the one you want to see.
- 4012: a different problem. The replicated folder was disconnected longer than
MaxOfflineTimeInDays(60 by default when content freshness is on) and is now in an error state.
That last line matters more than it looks. DFSR Event 2213 and DFSR Event 4012 need different fixes, and mixing them up is the most common mistake I see in forum threads.
Why the WMIC Command in the Event No Longer Works for Everyone
The event text tells you to run wmic /namespace:\\root\microsoftdfs path dfsrVolumeConfig where volumeGuid="..." call ResumeReplication. According to Microsoft’s deprecated features list, WMIC has been deprecated since Windows 10 21H1, and it is removed from Windows 11 24H2 and later. It is no longer available as a Feature on Demand there. The removal landed with the August 2026 preview updates. Microsoft says only the WMIC utility went away. WMI itself, and the DFSR WMI provider, are still there.
On older Windows Server releases wmic.exe may still exist, but I stopped relying on it. If you paste the command into PowerShell you also need the extra single quotes Microsoft mentions in the Event 2213 article, which is an easy way to get a confusing syntax error at 7 a.m. The CIM cmdlets call the same ResumeReplication method, work remotely, and behave the same on every supported server. That is what I use below.
Step 1: Confirm DFSR Event 2213 and Find the Volume GUID
Run this on the affected domain controller from an elevated PowerShell window. It pulls the recent recovery-related events so you can see whether you have 2213, 4012 or both:
Get-WinEvent -FilterHashtable @{LogName='DFS Replication'; Id=2212,2213,2214,4012} -MaxEvents 20 |
Select-Object TimeCreated, Id, LevelDisplayNameCode language: JavaScript (javascript)If the newest relevant entry is a DFSR Event 2213 with no 2214 after it, replication is still paused. Next, get the GUID of the paused volume. You can read it straight from the event, or list the volumes DFSR knows about and match the path:
$evt = Get-WinEvent -FilterHashtable @{LogName='DFS Replication'; Id=2213} -MaxEvents 1
$guid = [regex]::Match($evt.Message, 'volumeGuid="([0-9A-Fa-f-]{36})"').Groups[1].Value
$guid
Get-CimInstance -Namespace root\microsoftdfs -ClassName DfsrVolumeConfig |
Select-Object VolumeGuid, VolumePathCode language: PHP (php)VolumeGuid is the key property of the DfsrVolumeConfig class, a 36-character GUID. If the regex returns nothing, use the second command and pick the GUID whose path matches the volume named in the event, usually C: on a DC.
Step 2: Back Up SYSVOL Before You Resume
Microsoft is clear on this one: back up the replicated folders before resuming, because recovery can lead to unexpected conflict resolution and data loss. With SYSVOL the folder is small, so there is no excuse to skip it. The default location is C:\Windows\SYSVOL\domain. Adjust the path if your SYSVOL lives elsewhere:
robocopy C:\Windows\SYSVOL\domain D:\Backup\SYSVOL-domain-20261116 /E /COPYALL /R:1 /W:1 /XD DfsrPrivateI exclude DfsrPrivate because it holds DFSR’s own staging and conflict data, not your policies. If this DC holds GPO changes that never made it to the other DCs, this copy is your only record of them. Also export a GPO backup from a healthy DC with Backup-GPO -All -Path if you have not done it recently.
Step 3: Resume Replication with PowerShell (ResumeReplication via CIM)
Now call the same ResumeReplication method the event mentions, through CIM instead of WMIC. Replace the GUID if you did not set $guid in Step 1:
$vol = Get-CimInstance -Namespace root\microsoftdfs -ClassName DfsrVolumeConfig -Filter "VolumeGuid='$guid'"
$vol | Invoke-CimMethod -MethodName ResumeReplicationCode language: JavaScript (javascript)A ReturnValue of 0 means the call was accepted. Microsoft’s ResumeReplication reference notes that the method only works when the volume is in the auto recovery state with replication stopped, so a non-zero result on a healthy volume is expected. To run it from your workstation instead of an RDP session, add -ComputerName DC02 to Get-CimInstance. The rest works the same.

The DFS Replication service now logs 2212 again and then 2214 when recovery completes, as Microsoft’s missing SYSVOL and NETLOGON troubleshooting guide describes. On a large data volume that can take a while. On a DC with only SYSVOL it is usually quick. To speed up the follow-up, ask DFSR to poll Active Directory:
dfsrdiag polladStep 4: Verify the DFSR Event 2213 Fix
I check three things before I call it done. First, Event 2214 after the resume. Second, the SYSVOL replicated folder state, where 4 means Normal and 5 means In Error. Third, the SYSVOL and NETLOGON shares, which a DC does not publish until SYSVOL is initialized:
Get-WinEvent -FilterHashtable @{LogName='DFS Replication'; Id=2214} -MaxEvents 1 | Select-Object TimeCreated, Message
Get-CimInstance -Namespace root\microsoftdfs -ClassName DfsrReplicatedFolderInfo -Filter "ReplicatedFolderName='SYSVOL Share'" |
Select-Object ReplicationGroupName, ReplicatedFolderName, State
Get-SmbShare -Name SYSVOL, NETLOGON | Select-Object Name, Path
dcdiag /test:sysvolcheck /test:advertisingCode language: JavaScript (javascript)
Finally, make a harmless change on another DC, such as editing a test GPO comment, and confirm it shows up in this DC’s SYSVOL. That end-to-end check has saved me from closing tickets too early more than once.
DFSR Event 2213 or DFSR Event 4012? Do Not Mix Them Up
If Step 1 showed DFSR Event 4012, resuming will not help. The folder was offline longer than MaxOfflineTimeInDays, and DFSR refuses to replicate content it considers stale. You can check the current value with CIM:
Get-CimInstance -Namespace root\microsoftdfs -ClassName DfsrMachineConfig |
Select-Object MaxOfflineTimeInDays, StopReplicationOnAutoRecoveryCode language: JavaScript (javascript)The tempting shortcut is to raise MaxOfflineTimeInDays above the number of days in the event so the DC starts replicating again. I do not do that for SYSVOL. Content freshness protection exists so that a DC that slept for months cannot push old GPO versions back into the domain. The clean fix is a non-authoritative (D2) sync of SYSVOL on that DC, as described in Microsoft’s authoritative and non-authoritative synchronization article. In short:
- In ADSI Edit, open
CN=SYSVOL Subscription,CN=Domain System Volume,CN=DFSR-LocalSettings,CN=<DC name>,OU=Domain Controllers,DC=<domain>and setmsDFSR-Enabledto FALSE. - Force AD replication, for example with
repadmin /syncall /AdeP, then rundfsrdiag polladon that DC and wait for Event 4114. - Set
msDFSR-Enabledback to TRUE, force AD replication again and rundfsrdiag pollad. - Wait for Events 4614 and 4604. The DC has then pulled a fresh copy of SYSVOL from a healthy partner.
Do not set msDFSR-options on this DC. That attribute is only for an authoritative (D4) restore on the one DC you trust, which is a much bigger decision.
How I Prevent SYSVOL Replication Stopped After Unexpected Shutdown
Most cases of SYSVOL replication stopped after unexpected shutdown come down to how the DC went down. Here is what I keep an eye on:
- Shut DCs down cleanly. Hard resets from the hypervisor console and storage outages are the classic cause. Make sure VM tools can do a guest shutdown and that patch windows do not force-power-off guests.
- Alert on 2213 and 4012. A scheduled task or your monitoring tool watching the DFS Replication log for these IDs turns a silent problem into a ticket.
- Think before flipping the switch.
StopReplicationOnAutoRecoverycontrols whether DFSR pauses or recovers on its own. I prefer the pause plus a backup. If you change it, document why. - Never delete the DFSR database as a shortcut. Microsoft calls it unnecessary and discouraged.
This problem taught me to read the whole event before acting. DFSR Event 2213 tells you exactly what happened and what to run. The only part that aged badly is the tool it names. If replication is acting up on more than one DC, or if the DC also lost its secure channel, fix the AD side first:
Read more: Trust Relationship Failed Between Workstation and Domain: 5 Proven Fixes
Read more: How To Windows Server 2016 Metadata Cleanup Best Way
Read more: Kerberos RC4 Failures After Windows Updates? The Proven Fix
SYSVOL Replication November 2026 Quick Checklist
If you just want the short version of SYSVOL replication November 2026 troubleshooting with current Windows builds, this is the order I follow when DFSR Event 2213 shows up:
- Check the DFS Replication log for DFSR Event 2213 versus Event 4012 before running anything.
- Back up
SYSVOL\domainwith robocopy. - Resume with
Invoke-CimMethod -MethodName ResumeReplicationon theDfsrVolumeConfiginstance, not with WMIC from a Windows 11 workstation. - Wait for Event 2214, then confirm SYSVOL state 4 and the SYSVOL and NETLOGON shares.
- For Event 4012, run a non-authoritative sync instead of raising MaxOfflineTimeInDays.
Once the DC is healthy, look at why it went down uncleanly in the first place, because a DC that throws DFSR Event 2213 after every reboot is telling you something about its storage or shutdown process.
What does DFSR Event 2213 mean on a domain controller?
It means the DFS Replication service found its database was not shut down cleanly and, because auto recovery is disabled by default, it stopped replication on that volume. On a domain controller this pauses SYSVOL replication until an administrator runs the ResumeReplication method.
How do I resume DFSR replication without WMIC?
Use PowerShell CIM: get the paused volume with Get-CimInstance -Namespace root\microsoftdfs -ClassName DfsrVolumeConfig filtered on its VolumeGuid, then pipe it to Invoke-CimMethod -MethodName ResumeReplication. A ReturnValue of 0 means the call was accepted, and Event 2214 confirms recovery finished.
Should I back up SYSVOL before running ResumeReplication?
Yes. Microsoft recommends backing up the replicated folders first, because dirty shutdown recovery treats the local copy as non-authoritative and can move files to the PreExisting or ConflictAndDeleted folders.
What is the difference between DFSR Event 2213 and Event 4012?
Event 2213 is a paused volume after an unclean database shutdown and is fixed with ResumeReplication. Event 4012 means the replicated folder was disconnected longer than MaxOfflineTimeInDays; for SYSVOL the clean fix is a non-authoritative (D2) sync on that domain controller.
How do I know the fix worked?
Look for Event 2214 in the DFS Replication log, check that the SYSVOL Share replicated folder reports State 4 (Normal) in DfsrReplicatedFolderInfo, and confirm the SYSVOL and NETLOGON shares exist with Get-SmbShare or dcdiag /test:sysvolcheck.
References
Microsoft Learn: DFSR event ID 2213 in Windows Server
Microsoft Learn: Troubleshoot missing SYSVOL and Netlogon shares
Microsoft Learn: Force authoritative and non-authoritative synchronization for DFSR-replicated SYSVOL
Microsoft Tech Community: Understanding DFSR dirty (unexpected) shutdown recovery
Microsoft Learn: ResumeReplication method of the DfsrVolumeConfig class and DfsrMachineConfig class
Microsoft Learn: Deprecated features for Windows client (WMIC)






