Hello everyone!
Recently embarked on the task of getting ready to scavenge a pretty established, yet heavily stale, DNS zone. The AD integrated zone for the domain to be exact. I've poured though all the required reading regarding the do's and don'ts of scavenging. I decided that since I was not entirely sure what to expect, enable it in my lab environment - glad I did! I searched through the forum but I don't see anything exactly like this.
I followed along with "Think like a Computer" blog on how to enable scavenging, understanding the no-refresh vs refresh, common issues, etc. It was what I found to be the easiest to understand and laid out most of what I needed. Sorry,
can't post links yet - not yet verified.
I've configured the lab very straightforward and what I will go into production with: Microsoft defaults! I've set the aging on the zones I'm scavenging with 7+7 for no-refresh and refresh. I'm setting scavenging on one of the DC's only - again,
7+7 no-refresh+refresh. I've set the "Enable automatic scavenging of stale records" to run every 7 days. Most of the timestamps of records in my lab ranged from today to roughly 7 days old. According to the blog referenced
above, after setting up your aging/scavenging settings, they state:
"To actually scavenge stale resource records you right click the server and select “Scavenge stale resource records”. The server will then delete records with timestamps
older than the No-Refresh interval plus the Refresh interval. With the
default settings this is 14 days."
This is totally me! I'm using default settings too. When I executed the "Scavenge Stale Resource Records" on the AD DNS server, I was presented with a nice little 2501 event explaining what it deleted. What struck me as odd was
the amount of scavenged records. When I refreshed my zone, I was shocked to find that it deleted everything that had a timestamp older than ~4 days. For example, I ran it at 4:00pm on 5/28/2013. My oldest record is left in the zone is 5/25/2013
@1:00pm. I had dozens of other PC and server records in there that got whacked.
Given the details in the blog, and my (poor?) understanding of what to expect, a lot more got deleted than what I was expecting. There's no way I could kick this off in production since I have a couple hundred server records that are anywhere from 1 to
7 days old on their timestamp.
Has anyone else experienced this when kicking off the manual scavenge? Am I just missing a critical point somewhere?
Thank you for reading!