Domain Controller - Replication - DFS Functionality Status

Monitoring Benefit

The DFS Functionality test verifies that a configured source domain controller can replicate to the domain controller where the ENow agent is installed, and measures how long that replication takes. If replication fails, or the measured replication time exceeds the Error threshold configured for the test, a critical alert is raised.

This is a directional test. It answers one specific question: is content written on the source DC actually arriving here, and arriving quickly enough? Replication that has silently stopped is one of the more damaging Active Directory faults, because nothing breaks immediately — Group Policy simply becomes stale, SYSVOL contents diverge between sites, and the two domain controllers drift apart until something depends on a change that never arrived.

What the results page shows

The page is titled <server> AD DFS Functionality Status and has three columns:

One row is shown per configured source. Hovering a red status displays the measurement that caused it, in the form:

<source DC>: AD DFS Functionality Replication Time (<measured hours>) exceeded the Error threshold of (<configured hours>).

The measured value is elapsed time, so a replication partnership that has stopped entirely does not report an error — it reports a very large number of hours. A figure in the hundreds or thousands means replication has not completed for that long, not that it is merely slow.

This test is the counterpart to DFS Namespace Status. The two are easily confused and check entirely different things: DFS Functionality measures replication timing between two named servers, while DFS Namespace reads the event log. A healthy namespace result does not mean replication is working, and vice versa.

How do we verify the results?

The ENow agent collects the result periodically. This information is not written to a log, but the most recent result is cached on the monitored server:

\Program Files (x86)\ENow\Mailscape Agent\Cache\

The cache folder holds one *AgentMessage.xml file per agent area. Open the file whose name matches the agent reporting this test to see the values behind the dashboard.

To reproduce the measurement by hand, run a propagation test from the source DC and then report on it from the monitored DC. Substitute your own server names and replication group:

dfsrdiag PropagationTest /TestFile:enowcheck /RgName:"Domain System Volume" /SendingMember:DC01

dfsrdiag PropagationReport /TestFile:enowcheck /RgName:"Domain System Volume" /ReceivingMember:DC02 /ReportFilePath:C:\Temp\propagation.xml

The report shows whether the test file arrived and how long it took — the same two facts the monitoring is evaluating. To see whether there is a queue of unreplicated content between the pair:

Get-DfsrBacklog -SourceComputerName DC01 -DestinationComputerName DC02 -GroupName "Domain System Volume" -FolderName "SYSVOL Share"

Get-DfsrState -ComputerName DC02

A persistent, non-draining backlog and a rising measured replication time are the same fault seen from two directions.

Common warning or error results, and potential solutions