shadragonMemberJun 06, 2018 at 1:14 pm #167579
We have a 2016 file server with 2.6 million files that replicates via DFS to an off-site location.
For three months we’ve had no issues. Over the last week, I come in first thing and run a DFS diagnostic report to find 700K – 2.4 million files backlogged from the primary server to the backup server. This amount decreases throughout the day and I’ve seen it go a couple of days without any backlog, but I can’t see what changes the state on so many files that it queues them again for DFS. Obviously, this slows down the DFS for all other files that change day to day and it takes many extra hours to replicate.
The topology reports are good. All servers can connect with the others. No other shares with DFS seem to be affected. The largest 35 files equal 95GB and my staging files are set to 130GB.
I have no events in the DFS logs save the odd 5014 error: “The DFS Replication service is stopping communication with partner (Primary server) for replication group (Namespace path) due to an error. The service will retry the connection periodically.”
This is followed, literally a second later, with: “The DFS Replication service successfully established an inbound connection with partner (Primary Server) for replication group (Namespace path).”
So there is no extended down time. Even if there was, it does not account for 2+ million files suddenly needing to be replicated again. There’s no bulk AV scans, no Windows search and we turned off Netwrix logging thinking that was doing something strange. The backlogs still appear.
Any ideas? Thanks.
jessewalt375ParticipantSep 08, 2020 at 8:39 am #652386
The Get-DfrsBacklog cmdlet retrieves pending updates between two computers that participate in Distributed File System (DFS) Replication.
Updates can be new, modified, or deleted files and folders. The maximum number of files that this command displays is 100. The Verbose parameter displays a count of all backlogged updates. Any files or folders listed in the DFS Replication backlog have not yet replicated from the source computer to the destination computer. This is not necessarily an indication of problems. A backlog indicates latency, and a backlog may be expected in your environment, depending on configuration, rate of change, network, and other factors.
Hope this information is helpful.
- This reply was modified 4 months, 2 weeks ago by jessewalt375.
You must be logged in to reply to this topic.