I started working my new job a few months ago, and finally gotten around to looking into an error in our DNS Event Logs. The particular event id in question is 4515(http://support.microsoft.com/kb/867464). Now, doing some research, I am finding multiple different variations to this fix, though none seem to describe my situation that I am seeing.
When I connect to our Forest and Domain DNS application partitions in adsi, I see something that a previous admin did that is quite scary. For one particular zone, I see the following:
DomainDNSZones: DC=..InProgress-<GUID>-spam.company.org twice.
ForestDNSZones: DC=..InProgress-<GUID>-spam.company.org five times.
Domain NC: Five different containers with CNF:<GUID>. None of these have come up with 4515 errors in event logs.
So, as you can see, I have a problem. My question is, what is the best way to solve it. The zone in question that is having issues, I want to blow away permanetly. It servers no purpose to us currently, as nothing is utilizing it. As far as I can tell, it was tied into the old mdaemon system here a few years ago before they migrated to Exchange.
I have seen a few different solutions on how to solve this problem. One way is to follow the KB article I linked above, which involves ADSI Edit and manually deleting the zones. Another method describes changing the zone to a Primary Zone and let replication do its thing. Another involves a combination of the two.
Here is what I am thinking I should do, but want to verify it, and clarify a few steps:
1. Stop DNS on all Domain Controllers except the one I wish to perform the change on.
2. Open ADSI Edit and connect to the Forest and Domain DNS Partitions.
3. Delete all references to the problematic zone that say DC=..InProgress, leaving the correctly loaded copy of the zone.
4. Restart DNS and Netlogon Service, and run ipconfig /flushdns.
5. Start DNS Service on remaining Domain Controllers and force replication. Monitor and verify in ADSI Edit that they are all updated.
6. Delete the AD Integrated Zone, force replication, and verify all DC's are updated.
Does the above method sound like the correct procedure to follow? Anyone have any modifications?
Thanks,
Craig Russell
[Edit] - Figured I should include our network information:
1. 2003 R2 DC
2. 2008 SP1 DC
3. 2008 R2 DC