locked
SOLVED - DFS backlog is stuck one direction RRS feed

  • Question

  • I am running Windows Server 2008 R2 - 2 servers running DFS-R to synchronize a directory between the two of them. A couple days ago I realized that replication was only happening in one direction - and that the direction which was not working had a backlog consisting of various folders that did not exist on either server, as well as one entry that was completely blank.

    In DFS Management, I deleted the Replicated Folder for this replication group, and recreated it. I let the servers do their thing, synchronizing all of the files again, and now today, I am back in the same boat, kind of.

    The backlog in one direction is clear - all synched up.

    The backlog in the other direction, however, is stuck again. 

    There are 5 entries - 3 files, and two completely blank entries. Any additions/changes/deletions are just adding entries to the backlog.

    Deleting/renaming the files that are in the backlog do not clear that entry.

    At this point, I'm not sure what else to try. Anyone heard of this?


    • Edited by Erpara Monday, November 19, 2012 6:26 PM
    Friday, November 9, 2012 11:38 PM

Answers

All replies

  • Bump?
    Tuesday, November 13, 2012 8:58 PM
  • Hello,

    From server A please make sure you can ping server B using both the internal FQDN as well as the NetBIOS names of the servers and vice-versa.

    Also, make sure that RPC is running on both servers and that the DFS-R service is also running.

    Any errors from the event log would be helpful.

    If you have information being replicated that was not on either servers replicated folders, you may have a database journal wrap.

    You may have too much information replicating too slowly, not enough staging capacity or insufficient journal size. Services getting stopped can lead to this problem, broken WAN links, off-line servers, etc.

    If you wait until you are in wrap condition, it's too late. You have to start over or do an authorative restore. I would try again and monitor the services and replication every few hours to make sure everything is working as it should.


    Miguel Fra | Falcon IT Services, Miami, FL
    www.falconitservices.com | www.falconits.com | Blog





    • Edited by Miguel Fra Tuesday, November 13, 2012 9:35 PM
    Tuesday, November 13, 2012 9:19 PM
  • Thanks!

    I can ping via FQDN and NetBIOS from both servers.

    RPC is running, and the DFS-R is running on both servers.

    There are no relevant error messages in the event logs.

    This share is roughly 200GB in size - I wouldn't call it a large directory really. The files that are sitting in the backlog are much smaller than the staging area size.

    There is no backlog in one direction (primary > replica), but the server that nobody is connected to now (replica > primary) has 9 entries in the backlog (3 of them are blank lines). 

    I'm seeing if there's some firewall issue - but if that doesn't pan out, then I'm still searching...

    Tuesday, November 13, 2012 9:57 PM
  • Hi,

    Can you create another set of linked folders for testing and see if replication works between them?


    Miguel Fra | Falcon IT Services, Miami, FL
    www.falconitservices.com | www.falconits.com | Blog

    Tuesday, November 13, 2012 11:26 PM
  • Hmm nope - same exact problem...

    I'll reply back with the results once I've verified the firewall rules.

    Thanks for the response!

    Tuesday, November 13, 2012 11:43 PM
  • Make sure Ports 135, 5722 are open.

    Miguel Fra | Falcon IT Services, Miami, FL
    www.falconitservices.com | www.falconits.com | Blog


    Wednesday, November 14, 2012 1:27 AM
  • FYI - once all traffic was allowed between the two servers, the backlog immediately cleared. So it was definitely a problem with which ports were open.

    Thanks!

    Monday, November 19, 2012 6:25 PM