Some people may not agree with me but my opinion is that as long as it comes back to zero during your replication period it doesn't really matter how big it gets, you are fine. When it starts rolling over from one period into the next you have an issue. If you see it holding on between periods like that break out repadmin /showqueue or go download adqueueloop from www.joeware.net free win32 tools. Then watch to see if the stuff is making it through the queue or something is hanging the queue up.
 
Most likely you will have a DC that is hanging the queue up because it is replicating slow (needs a reboot or DC has hardware issues usually) or the network is really flaky and the replication keeps getting dorked and it has to keep restarting and it takes forever for it to decide to time out a bad DC. In cases like these I find that writing a little script that smokes the GUID DNS entry in DNS is a handy way to get the problem out of the way while you trouble shoot the DC to find out what is going on. Note that the problem goes away because it can't resolve the problem DC and pull changes from it and that is why the problem goes away, note that the DC itself can still pull changes from another DC and if you have a distributed hub and spoke replication model this is generally fine for even a day or two.
 
Note that I found a bug in one of the POST SP3 hotfixes for that counter and kicked MS pretty good for it and they fixed it. I was kind of curious how their ITG didn't find the issue if they are actually monitoring that counter since my counters were getting into the tens of thousands and not coming back down. The hot fix number is
 
 
I am not so sure if you have to have SP4 installed like it says because the buddy drop I got ran fine on SP3 though I guess they could have changed something. I haven't deployed it because I am gearing to push W2K3 instead.
 
Hope that helps.
 
  joe
 
 


From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ryan A. Conrad
Sent: Tuesday, December 09, 2003 10:03 AM
To: [EMAIL PROTECTED]
Subject: [ActiveDir] DRA Pending Synchronizations

HI!

 

We have an AD with a populated child domain and an empty root.  We are using NetIQ to monitor various components of the directory and lately have been getting the following errors for both domains:

 

Root: DRA Pending Replication Synchronizations High: The AD replication queue length is 7, over the threshold of 2.

Child: DRA Pending Replication Synchronizations High: The AD replication queue length is 8, over the threshold of 2.

 

The queue length has been growing slowly over the past few weeks, but there appears to be no problem when looking at replmon or the event logs.

 

Our domain is not very large yet (a few hundred users and 200 or so computer objects).  We had to rebuild a DC a few weeks ago and are thinking that it might be trying to replicate to an old GUID, but cannot be certain.

 

Does anyone have any ideas of an appropriate queue length, where to find out what is in the queue, or what could be potentially blocking the queue.

 

Thanks!

 

Ryan

Ryan A. Conrad                                                      
Windows 2000 MCSE
Windows Server 2003 MCSA MCSE
CCNA Network+
IMSS Platform Engineering Wintel Analyst
Bristol-Myers Squibb Company
* email: [EMAIL PROTECTED] - ( phone: 609.818.6135

 

Reply via email to