XADM: Directory Replication Takes Too Long (148381)


The information in this article applies to:

  • Microsoft Exchange Server 4.0
  • Microsoft Exchange Server 5.0
  • Microsoft Exchange Server 5.5
This article was previously published under Q148381

SUMMARY

This article explains why, depending on the Directory Replication topology, the time required to replicate directory changes to all parts of a Microsoft Exchange Organization may be several days.

MORE INFORMATION

Consider the following two Directory Replication topologies.

Topology A: Hub 
Site X  <------->  Site W   <------> Site Z
                      ^
                      |
                      |
                      V
                    Site Y

Topology B: Cascade 
Site W  <----> Site X  <---> Site Y  <-----> Site Z

In each of the above topologies, the arrows indicate a directory replication connector. Assume that each connector is scheduled to replicate at one time during the day.

In Topology A, if a new mailbox is added to Site X, the maximum delay for that directory entry to be propagated to all the other Sites would be 2 days (one day for each directory replication connector hop).

In Topology B, if a new mailbox was added to Site W, the maximum delay for that directory entry to reach all the other sites (specifically Site Z) would be 3 days.

Hence, the time required to propagate directory changes to all sites in an Exchange Organization depends on the replication schedule as well as the number of hops involved and the number of hops depends on the Directory Replication Topology.
Modification Type:MinorLast Reviewed:4/28/2005
Keywords:kbenv kbinfo KB148381
©2004 Microsoft Corporation. All rights reserved.