Skip to main content
Databases & applications

DFSR Backlog

Monitor the DFS Replication backlog between two members of a replicated folder on Windows Server.

The DFSR Backlog check monitors DFS Replication on Windows Server. It determines how many files a sending member has not yet transferred to a receiving member for a replicated folder, and reports when this backlog reaches the thresholds. A growing backlog indicates stalled or too slow replication. The check runs on Windows only (x64 and Arm64) through the octoja agent on the device you assign it to, queries the backlog there with dfsrdiag backlog, and runs every 15 minutes by default.

Prerequisites

  • The dfsrdiag command-line tool must be available on the device. It is part of DFS Replication and the DFS Management Tools of Windows Server.
  • Enter the names of the replication group, replicated folder and members as they appear in DFS Management.
  • dfsrdiag backlog always queries both members over WMI, including the partner. The octoja agent runs the check as SYSTEM, so it signs in to the partner with the device's computer account. By default, the DFSR WMI namespace only admits local administrators: the computer account must be a member of the Administrators group on the partner, otherwise access is denied.

Caution: On domain controllers, the local Administrators group is the domain group. This gives the computer account rights equivalent to a domain admin. If you do not want to grant that, monitor DFS Replication problems locally with the Event Log check instead. It shows logged problems, but not the number of pending files.

Steps

  1. Open the Windows device and, on its Checks tab, click Add Check → DFSR Backlog. Alternatively, assign the check with a config package: Assign checks with config packages.
  2. Enter Replication group, Replicated folder, Sending member and Receiving member.
  3. Adjust the thresholds if needed and save the check.

Settings

FieldTypeRequiredDefaultDescription
Replication groupStringYes—DFSR replication group name
Replicated folderStringYes—DFSR replicated folder name
Sending memberStringYes—Source server whose outbound backlog to measure
Receiving memberStringYes—Destination server receiving the replication
Warning backlog filesIntegerNo100Number of pending files at which a warning is raised
Error backlog filesIntegerNo500Number of pending files at which a critical result is raised

The check measures one direction. To monitor both directions between two servers, create two checks with the members swapped.

  • The Error threshold corresponds to the Critical level. octoja rates check results as OK, Warning or Critical.
  • If Warning backlog files is higher than Error backlog files, the check silently swaps the two values before evaluating.
  • Leading and trailing spaces in the four name fields are ignored.

Results

SituationResult
Backlog below Warning backlog filesOK
Backlog reaches or exceeds Warning backlog files / Error backlog filesWarning / Critical
One of the four required fields is emptyWarning, Replication group/folder and members are required
dfsrdiag is not available on the deviceWarning, DFSR tools (dfsrdiag) not available
dfsrdiag reports an error or returns no readable backlog count, for example because of wrong names or an unreachable memberWarning, dfsrdiag failed: followed by the output of dfsrdiag

For a valid result, the check details show the number of backlog files.

Was this helpful?

Still need help? Ask the team