Isn’t this the type of thing that should be in an AD design? ☺

From: [email protected] [mailto:[email protected]] On 
Behalf Of Michael Leone
Sent: Wednesday, 17 September 2014 1:49 AM
To: [email protected]
Subject: [NTSysADM] Re: Confused about domain replication times

Turns out it was simple. As usual, I was wrong about some of my assumptions. 
Turns out, my boss *did* make changes to the replication schedule. In Sites and 
Services, any "automatically generated" links have the (default?) schedule of 
60 minutes. Any manually added servers (apparently manually added server 
connections to the NTDS settings, effectively so that every DC has a connection 
to every other DC. And these manual connections are set for replication 4x per 
hour.

That accounts for the time differentials I am seeing, and so I won't have to 
worry about replication.

I probably should change all those links to be 4x per hour. I think they may 
have been changed a long while ago, when we had slower (and less reliable) 
links between sites. Now all my links are 1G or higher, and extremely reliable, 
so 4x an hour wouldn't swamp the link (we don't generate THAT much AD traffic 
...).

So one worry down. Now to check dcdiag, to prepare for the Win2012 R2 AD 
upgrade ...


On Tue, Sep 16, 2014 at 10:18 AM, Michael Leone 
<[email protected]<mailto:[email protected]>> wrote:
We run a Win 2008 R2 domain (forest/domain functional level) - in a 
parent/child domain configuration; all users and resources in the child domain. 
We want to upgrade to Win2012 R2 AD (mostly to stay current, and also because 
we want to start using DFS for our user profiles .. but that's another post). 
Anyway, as first steps toward domain upgrade, I am checking replication (and 
later dcdiag). And I am noticing something I am not understanding ..

My replication times seem to be increasing more than what I think should be 
normal. I have 4 sites - 2 have 1 DC each, 1 has 2 DCs, the last has 6 (3 DCs 
for the root domain, 3 DCs for the child domain). Doing "repadmin /replsummary" 
is showing me the largest delta is increasing between the servers in site #4, 
up to about an hour. And not going above 15 minutes or so for the other DCs in 
the remote sites.

Mind you, there are no fails, no errors. Just oddly high times, and mostly in 
the "Source DSA" section (the times are much better in the "Destination DSA" 
section.

And then they will drop down .. Look at the difference 100 seconds make, in the 
list of deltas:
C:\Users\admin>repadmin /replsummary
Replication Summary Start Time: 2014-09-16 09:58:09

Beginning data collection for replication summary, this may take awhile:
  .............


Source DSA          largest delta    fails/total %%   error
 Site-1-001               14m:20s    0 /  12    0
 Parent-Site-4-001        14m:20s    0 /  24    0
 Parent-Site-4-002        14m:20s    0 /  24    0
 Parent-Site-4-003        59m:20s    0 /  20    0
Site-4-001                59m:20s    0 /  38    0
Site-4-002                59m:20s    0 /  40    0
Site-4-003                10m:34s    0 /  34    0
 Site-2-001               14m:20s    0 /  18    0
Site-3-001                14m:20s    0 /  24    0
Site-3-002                14m:20s    0 /  12    0


Destination DSA     largest delta    fails/total %%   error
 Site-1-001               06m:53s    0 /  18    0
 Parent-Site-4-001        11m:56s    0 /  20    0
 Parent-Site-4-002        06m:31s    0 /  16    0
 Parent-Site-4-003        01m:07s    0 /  16    0
Site-4-001                02m:00s    0 /  24    0
Site-4-002                08m:54s    0 /  38    0
Site-4-003                59m:20s    0 /  48    0
 Site-2-001                  :24s    0 /  12    0
Site-3-001                10m:34s    0 /  24    0
Site-3-002                05m:26s    0 /  30    0



C:\Users\admin>repadmin /replsummary
Replication Summary Start Time: 2014-09-16 09:59:49

Beginning data collection for replication summary, this may take awhile:
  .............


Source DSA          largest delta    fails/total %%   error
 Site-1-001               10m:33s    0 /  12    0
 Parent-Site-4-001        10m:34s    0 /  24    0
 Parent-Site-4-002        13m:36s    0 /  24    0
 Parent-Site-4-003        13m:36s    0 /  20    0
Site-4-001                13m:36s    0 /  38    0
Site-4-002                13m:36s    0 /  40    0
Site-4-003                12m:14s    0 /  34    0
 Site-2-001               10m:33s    0 /  18    0
Site-3-001                10m:33s    0 /  24    0
Site-3-002                12m:14s    0 /  12    0


Destination DSA     largest delta    fails/total %%   error
 Site-1-001               08m:33s    0 /  18    0
 Parent-Site-4-001        13m:36s    0 /  20    0
 Parent-Site-4-002        08m:11s    0 /  16    0
 Parent-Site-4-003        02m:47s    0 /  16    0
Site-4-001                03m:40s    0 /  24    0
Site-4-002                10m:34s    0 /  38    0
Site-4-003                01m:00s    0 /  48    0
 Site-2-001               02m:04s    0 /  12    0
Site-3-001                12m:14s    0 /  24    0
Site-3-002                07m:06s    0 /  30    0

Does that look normal?
Why 59 minutes delta for DCs in the same site, and less than 15 for remote 
sites? (that's configurable, I know, but I don't think we did change it. Where 
can I check the replication schedule?)

This was run on the DC Site-4-003, BTW.

I think my domain is OK, replication wise. I'm just a bit confused. Any help in 
clearing my confusion much appreciated.

Reply via email to