Monitor and Repair Critical Active Directory Replication Failures
Active Directory is one of the foundational technologies IT departments rely on. Active Directory itself is dependent on replication that ensures all controllers receive information about changes in the network.
Active Directory Replication Behavior
Depending on whether it happens within a site or between a site, AD replication takes one of two forms.
Intra-Site Replication
Within a site, replication is expected to happen very quickly, in under a minute, over high speed links. Replication occurs via a ring topology that means each domain controller just needs to communicate with two other controllers to send or receive replicated data.
Inter-Site Replication
Unlike intra-site replication, replication between sites is highly sensitive to the latency, bandwidth, and reliability of the network connections. Data is shared over links between sites and then distributed within the site via the ring topology.
Monitoring Active Directory Replication
Because keeping sites consistent is so critical, replication should be monitored using the Active Directory Replication Status Tool (ADREPLSTATUS) or REPADMIN. In repadmin, use the flag /replsummary for a brief overview of status. The /queue flag will display items remaining to be replicated.
Replication failures can arise both from problems within AD itself and from services it’s dependent on, such as network links and DNS, security, and time services.
Fixing Active Directory Replication Failures
Problems in replication are identified by associated replication error codes and event IDs. The solutions to the problem will obviously vary depending on the specifics of the issue and its root cause. Once the underlying problem is repaired, it may be necessary to force replication to resynch the data at all the controllers.
The Active Directory Sites and Services Console allows you to select the controller to replicate from and the target controller to replicate to; repadmin/replicate behaves similarly. Note this will only correct replication at the target; it will not propagate further to additional controllers to correct a broader failure. You can use repadmin /syncall to synchronize a controller with all its partners; be cautious using this command due to its impact on the network.
Keeping Active Directory running properly is essential for both security and user experience reasons. Get help supporting your Microsoft infrastructure from Prescient Solutions. Certified Microsoft Experts quickly diagnose and resolve problems, along with providing ongoing support. Contact us to learn how Prescient Solutions Microsoft services keep critical IT infrastructure performing at top levels of efficiency and reliability.