Skip to content

Est. 2011ยทMicrosoft Partner ID 7033487ยทDelivery under 3 minยทSupport 7 days a week

Your vault is empty.

License Error Event ID 4012

DFSR Event ID 4012: SYSVOL replication stopped after too long offline

12 min read Updated October 5, 2026 Windows Server: AD, DNS & Group Policy

Fix it now

Event ID 4012 is content freshness protection: replication on that folder failed for longer than MaxOfflineTimeInDays allows, which is 60 days by default, so DFSR stopped rather than replicate stale knowledge over good data. It will not resume on its own. A non-authoritative sync is the fix, and it changes nothing on the healthy DCs.

Run these on the broken DC and on a healthy partner, in an elevated Command Prompt

net share
dfsrdiag pollad
  1. Confirm a healthy source exists: on another DC, net share should list SYSVOL and NETLOGON. If no DC in the domain shares them, you need the authoritative procedure instead.
  2. Copy the broken DC’s SYSVOL Policies and Scripts folders aside as a safety net before you start.
  3. In ADSI Edit on the broken DC, find CN=SYSVOL Subscription,CN=Domain System Volume,CN=DFSR-LocalSettings,CN=<server name>,OU=Domain Controllers,DC=example,DC=com and set msDFSR-Enabled to FALSE.
  4. Force Active Directory replication throughout the domain, then run dfsrdiag pollad on that DC. Event ID 4114 confirms DFSR saw the change.
  5. Set msDFSR-Enabled back to TRUE, force AD replication again, run dfsrdiag pollad again, and watch for Event ID 4614 then Event ID 4604.

Nothing is lost. The local database is discarded and the content is pulled back from a partner, so the domain’s copy wins over whatever was on the broken DC.

If 4604 appears and the DC shares SYSVOL and NETLOGON, you are done. The next section explains why DFSR stops rather than catching up, and what to do when every DC is in the same state.

Why it happens

DFSR keeps a per-volume database recording a version vector for every file it replicates. Two partners compare vectors and exchange only what is missing. That works while both databases describe the same lineage of changes. Once a member’s knowledge is stale enough, DFSR can no longer distinguish a file deleted deliberately from a file that is new locally, so it refuses to guess.

MaxOfflineTimeInDays is where the line is drawn. It is enabled by default on Windows Server 2012 and later, with a default of 60 days, and setting it to 0 disables the protection. Cross it and replication for that member stops, Event ID 4012 is written, and it sits there indefinitely. That is deliberate: a domain controller that has been in a cupboard for three months should not be allowed to push its idea of Group Policy at the rest of the domain.

The remedy is a non-authoritative synchronisation, still widely called a D2 after the old File Replication Service terminology. You take the member out of replication by setting msDFSR-Enabled to FALSE, let that change replicate, poll for it, then put it back to TRUE and let DFSR rebuild its database from a healthy partner. Event ID 4114 confirms the member was disabled; 4614 says DFS Replication is waiting to do initial replication; 4604 says it has initialised SYSVOL.

The DC was off, or DFSR stopped, for longer than the offline limit

You have this one if Event 4012 names the SYSVOL path on one DC while every other DC in the domain replicates normally.

  1. Copy that DC’s SYSVOL Policies and Scripts folders aside first. It costs nothing and answers the question later.
  2. Set msDFSR-Enabled to FALSE on its SYSVOL Subscription object, force AD replication throughout the domain, then run dfsrdiag pollad. Event 4114 confirms it took.
  3. Set msDFSR-Enabled back to TRUE, force AD replication again, and run dfsrdiag pollad again.
  4. Watch for Event 4614, then Event 4604. Until 4604 appears the folder is still in initial synchronisation.

This is the common case and it is safe. The domain’s current policy replaces whatever was on the broken DC, which is exactly what you want.

Every domain controller reports 4012

You have this one if No DC shows 4604, and SYSVOL and NETLOGON are shared nowhere in the domain.

  1. Identify the DC whose SYSVOL content is most complete, and back up the Policies and Scripts folders from every DC before starting.
  2. Set the DFS Replication service Startup Type to Manual and stop the service on every domain controller in the domain.
  3. On the chosen DC, set msDFSR-Enabled to FALSE and msDFSR-Options to 1. On every other DC, set msDFSR-Enabled to FALSE.
  4. Force AD replication and validate it on all DCs, then start the DFSR service on the authoritative DC only. Event 4114 appears there.
  5. Set that DC’s msDFSR-Enabled back to TRUE, replicate again, run dfsrdiag pollad on it, and wait for Event 4602.
  6. Start DFSR on the remaining DCs, set their msDFSR-Enabled back to TRUE, run dfsrdiag pollad on each, then return the service Startup Type to Automatic everywhere.

The service stop across every DC is part of the documented procedure, not an optional precaution. Skipping it is why this goes wrong.

An authoritative synchronisation replaces the SYSVOL contents of every other domain controller. Pick the wrong source and you publish an old copy of Group Policy and logon scripts across the whole domain. Back up the Policies and Scripts folders from every DC first, and be certain which copy is the good one before you set msDFSR-Options.

An unclean shutdown left replication paused

You have this one if Event 2213 on a volume, with no replication since the last unexpected restart or power loss.

  1. Read the 2213 entry. It carries the volume GUID and the exact command to run.
  2. Resume with the documented WMI call: wmic /namespace:\\root\microsoftdfs path dfsrVolumeConfig where volumeGuid="<guid>" call ResumeReplication
  3. Confirm by checking for Event IDs 2212 and 2214 on the server you ran it on.
  4. If you want automatic recovery to run without stopping replication, set StopReplicationOnAutoRecovery to 0 under HKLM\System\CurrentControlSet\Services\DFSR\Parameters.

This is a pause, not a break. It needs one command and a check, not the whole non-authoritative procedure.

Full reference

The authoritative procedure, step by step

  1. Set the DFS Replication service Startup Type to Manual and stop the service on all domain controllers in the domain.
  2. On the DC you are making authoritative, preferably the PDC emulator, set msDFSR-Enabled=FALSE and msDFSR-options=1 on its SYSVOL Subscription object.
  3. On every other DC, set msDFSR-Enabled=FALSE on the same object.
  4. Force Active Directory replication throughout the domain and validate its success on all DCs.
  5. Start the DFSR service on the authoritative DC. Event ID 4114 appears.
  6. Set msDFSR-Enabled=TRUE on that DC.
  7. Force Active Directory replication again and validate it on all DCs.
  8. Run DFSRDIAG POLLAD on the authoritative server. Event ID 4602 confirms the D4.
  9. Start the DFSR service on the other DCs. Event ID 4114 appears on each.
  10. Set msDFSR-Enabled=TRUE on every other DC.
  11. Run DFSRDIAG POLLAD on all non-authoritative DCs.
  12. Return the DFSR service Startup Type to Automatic on all DCs.

What to set and what to watch for

Item Meaning
msDFSR-Enabled FALSE takes the member out of replication; TRUE puts it back
msDFSR-options set to 1 Marks this member as the authoritative source for SYSVOL
Event ID 4114 SYSVOL replication is no longer being replicated on this member
Event ID 4602 SYSVOL initialised as the authoritative copy: the D4
Event ID 4604 SYSVOL initialised after a non-authoritative sync: the D2
Event ID 4614 DFS Replication is waiting to do initial replication
dfsrdiag pollad Makes DFSR re-read its configuration from the directory now

Content freshness protection

Detail
Parameter MaxOfflineTimeInDays
Default 60 days
Enabled by default Windows Server 2012 and later
Disabled by setting it to 0
What crossing it produces Event ID 4012, and replication stops for that member until a sync is run

Raising the value does not resume a member that has already stopped. It changes the threshold for the future. If your estate has domain controllers that sit offline for months at a time, the number is not the problem.

Choosing the right recovery

What you see What to do
4012 on one DC, others healthy Non-authoritative sync on that member alone
4012 on every DC, SYSVOL shared nowhere The authoritative procedure, with the service stopped everywhere first
2213 after an unclean shutdown Paused, not broken. Run the ResumeReplication call from the event text
4614 with no following 4604 Initial sync started and did not finish. Chase connectivity to the partner
4012 on a DC that is also past the tombstone lifetime Deal with the Active Directory problem first; SYSVOL is not the issue

Resuming after a dirty shutdown

Detail
Command wmic /namespace:\\root\microsoftdfs path dfsrVolumeConfig where volumeGuid="<guid>" call ResumeReplication
Where the GUID comes from The text of the 2213 event itself
Confirmation Events 2212 and 2214 logged on the server you ran it on
Related registry value StopReplicationOnAutoRecovery under HKLM\System\CurrentControlSet\Services\DFSR\Parameters, DWORD, 0 to let automatic recovery run without stopping replication

Event ID 4212, and why this article does not act on it

Event ID 4212 is widely described as the DFSR database on a volume being unrecoverable, with a remedy of stopping the service and deleting the database folder under System Volume Information so it rebuilds. Microsoft does not publish a meaning for 4212 in a place this article could point you at, and that remedy forces a full initial synchronisation for every replicated folder on the volume, not only SYSVOL. On a file server that is a serious event. So: read the entry on your own server, treat the documented dirty-shutdown recovery above as the first thing to try, and if the database really is unrecoverable, plan the rebuild as a change with the full-resync cost understood rather than as a troubleshooting step.

A note on forcing Active Directory replication

Both procedures say “force Active Directory replication throughout the domain” rather than naming a command. repadmin /syncall with the appropriate switches is the usual way to do it, and this article does not quote the flag letters because it could not confirm them against a Microsoft page. Run repadmin /? on the machine in front of you, or use Sites and Services, and then validate the replication succeeded on all DCs before moving to the next step. The validation is the part the procedure actually depends on.

When a licence is the actual fix

The sync itself costs nothing: an attribute, two commands and some patience. The licensing question appears only when the DC has been offline long enough that it also runs a Windows Server release you no longer want in production, or when you decide the cleanest recovery is to demote it and build a replacement. A replacement domain controller needs a licensed Windows Server, and Arco supplies Windows Server 2025 Standard and will check whether your current agreement already lets you install the version you actually run. If the replacement is joining a host that already runs several Windows virtual machines, ask about the edition as well: Standard limits how many Windows virtual instances the licensed hardware may run and Datacenter does not, so the cheaper answer depends on your density. Client access licences are counted separately, by user or by device.

Every code this article covers

Code What it points at Source
Event ID 4012 Replication for the folder stopped because replication on it failed for longer than the MaxOfflineTimeInDays parameter allows: content freshness protection, 60 days by default Microsoft Learn
Event ID 2213 Replication on the volume is paused after a dirty shutdown was detected. The event text carries the volume GUID and the WMI command needed to resume Microsoft Learn
Event ID 2212 Logged, with 2214, on the server where the ResumeReplication command was run Microsoft Learn
Event ID 4614 DFS Replication is waiting to do initial replication. Until 4604 or 4602 follows it, the folder is still initialising Microsoft Learn
Event ID 4212 Quoted as an unrecoverable DFSR database on a volume. Microsoft publishes no meaning for it, so read the entry itself and treat any database rebuild as a planned change, because it forces a full initial sync for every replicated folder on that volume not published by the vendor

Confirm the fix worked

  1. Event ID 4604 appears in the DFS Replication log on the repaired DC, or 4602 if you ran the authoritative procedure.
  2. net share on that DC lists both SYSVOL and NETLOGON.
  3. A policy folder under SYSVOL matches the same folder on a healthy DC.
  4. The DFS Replication service Startup Type is back to Automatic on every DC you changed it on.
  5. No new 4012 appears over the following week, so you know the underlying outage is over rather than paused.

Questions people ask about this

Will a non-authoritative sync delete my Group Policy?

No. It discards the local database and pulls content back from a healthy partner, so the domain’s current policy replaces whatever was on the broken DC. Copy the Policies and Scripts folders aside first anyway; it costs nothing and answers the question later.

Can I just raise MaxOfflineTimeInDays?

It does not retroactively resume a member that has already stopped, so it will not fix the DC in front of you. The default is 60 days and setting it to 0 disables the protection entirely. Raising it to accommodate DCs that sit offline for months hides a real problem rather than solving one.

Do I have to stop the DFSR service for an authoritative sync?

Yes, on every domain controller in the domain, with the Startup Type set to Manual, before you change any attribute. That is part of Microsoft’s documented procedure and it is the step most often skipped. You start the service again on the authoritative DC only, before setting its msDFSR-Enabled back to TRUE.

How long does the sync take?

SYSVOL is normally small, so most environments see it finish quickly. Rather than guess, watch for Event ID 4604 – or 4602 for an authoritative sync. Until that appears, the folder is still in initial synchronisation.

What about Event ID 4212?

Microsoft does not publish a meaning for it, so treat the advice circulating about deleting the DFSR database folder with caution. That rebuild forces a full initial synchronisation for every replicated folder on the volume, not just SYSVOL. Try the documented dirty-shutdown recovery first, and if a rebuild really is needed, plan it as a change.

Related error codes

Was this article helpful?

Your feedback helps us improve our documentation.

Related articles

Free Fix Event ID 5313: GPOs filtered out and never applied to the target computer Free Fix Event ID 1046: the DHCP server is not authorised in Active Directory License Error Error 8418 schema mismatch: replication blocked by an out-of-date schema copy Free Fix Error 1789 trust relationship failed: rebuilding a broken secure channel
โ† Back to Knowledge Base