ERROR_DS_DRA_INCOMPATIBLE_PARTIAL_SET (0x00002110) Fix

Yeah, this one's a pain. You're mid-replication check, everything else looks fine, then boom — 0x00002110. Let's cut the noise and get it fixed.

Fix It Now

Open an elevated command prompt on the destination DC (the one getting the error). Run this:

repadmin /syncall /AdeP
repadmin /showreps | find /i "0x00002110"

The culprit is almost always a stale partial replica (also called a partial naming context or NC). Partial replicas hold only domain objects, not configuration or schema. When a DC has an outdated one — say, from a previous domain controller that was demoted improperly — replication blows up.

Here's the real fix:

1. Identify the broken NC with repadmin /showreps. Look for the line with the error code.
2. Note the NC name — it'll be something like DC=DomainDnsZones,DC=contoso,DC=com.
3. Remove the partial replica entirely:

repadmin /removeholderingring * <NC-name> <destination-DC> /partial

Then force a full sync:

repadmin /syncall /AdeP
repadmin /replicate <destination-DC> <source-DC> <NC-name> /force

Wait 5-10 minutes for replication to catch up. Check again with repadmin /showreps. If you see no errors, you're done.

Why This Works

Partial replicas are just that — partial. They don't hold the full NC. When a DC gets a partial replica from a source that doesn't have the authoritative copy (maybe because the source was demoted or the NC got deleted), the two replicas are incompatible. The error code 0x00002110 literally means "the partial replica is incompatible."

By removing the hold and forcing a fresh sync, you're telling the DC: "Forget what you had — grab the current version from the source." The DC then re-requests the partial NC, and the source sends a clean copy.

Less Common Variations

Sometimes the error shows up during a domain controller promotion — especially if you're promoting a new DC and it tries to replicate from a source that has a stale partial replica. In that case, don't bother trying to fix the promotion mid-stream. Demote the new DC, clean up the source's partial replicas (use the commands above), then promote again.

Another variation: the error appears on read-only domain controllers (RODCs). RODCs hold partial replicas by design, but if the password replication policy changes or the RODC has a cached partial replica from a now-defunct source, same error. Fix is identical — just run the removeholderingring command against the RODC.

One more: SYSVOL replication mixed with FRS vs DFSR. If you're still running FRS (Windows Server 2008 or earlier), the error can disguise itself as a FRS sync failure. Check the FRS event log. If you see 0x00002110 there, the fix is the same — clear the stale partial replica — but also consider migrating to DFSR. FRS is old and brittle. Skip the band-aid; migrate to DFSR.

Prevention

Don't let this happen again. Three things:

• Demote DCs cleanly. Use dcpromo /forceremoval only as a last resort. A clean demotion removes the partial replicas. A forced removal leaves zombie entries.
• Monitor replication health weekly. Run repadmin /replsummary in a scheduled task. Flag error codes 0x00002110 and 0x000020F2 (stale object) immediately.
• Keep your AD forest functional level consistent. Mixing 2008, 2012, 2016, and 2019 DCs can cause partial replica mismatches if older DCs don't support the newer NC attributes. Upgrade your functional level to Windows Server 2016 or higher.

That's it. This fix has saved my bacon more times than I can count. Go clean those partial replicas.