SBS 2003 in Journal Wrap Error State from Previous Migration

Discussion in 'Windows Small Business Server' started by JNelsonCPath, May 5, 2010.

  1. JNelsonCPath

    JNelsonCPath Guest

    I apologize ahead of time if this question was already discussed, I realize
    it is an older topic, but I couldn't find my particular "offshoot" of the
    issue so decided to post a question.

    Here is the history of the problem. I have an SBS 2003 Standard domain.
    Two years ago, I moved SBS 2003 to more powerful hardware by way of
    Microsoft's migration procedure. All went well except for the following:
    When I removed the old SBS from the network, I began getting errors in my
    file replication log on the new server complaining that it had a journal wrap
    error (event ID 13568) and could not read from the NTFS USN journal on the
    other server. Well the error is correct because the old server was removed
    and no longer exists. I made an assumption that Microsoft's recommended
    method for migrating SBS 2003 to new hardware would deal with any replication
    issues when the domain data was transferred, but it did not and the new
    server has been looking for the old one ever since.

    For the past two years, the SBS 2003 server has been running tip top and has
    experienced few issues. We now plan to migrate to SBS 2008 to take advantage
    of the newer technologies, and this journal wrap error has become important
    again. I have researched the web for days looking for a solution to my
    particular situation and cannot find anything that makes sense for this
    particular problem. All of the solutions, ideas, posts, KB articles, etc...
    that I have been looking at are solutions and ideas for restoring replication
    between two or more partners. While these solutions and ideas are talking
    about the same error message that I am seeing on my server, I have to assume
    that they don't apply to me becasue there is no location on my network to
    replicate with.
    How can I resolve the NTFRS errors I am seeing with no replication partner
    available? It is important that I resolve this before we migrate. All
    solutions I have seen say that the server will be removed from the
    replication set and then re-added to it which is impossible due to the old
    server being gone.

    Thank you in advance!
     
    JNelsonCPath, May 5, 2010
    #1
    1. Advertisements

  2. See the error message in the event logs?
    Read it.
    See what it says to do? See how it says to add an reg key?
    Do it.
    It's that easy.
     
    Susan Bradley, May 5, 2010
    #2
    1. Advertisements

  3. JNelsonCPath

    JNelsonCPath Guest

    Thank you for your response. I have read that blog several times already and
    from what I understand, I cannot use the burflags option because I do not
    have a replication partner available. This isnt as easy as a standard
    replication breaking between two servers; but rather it was a migration that
    appears incomplete.
    One piece of advice I was given that sounds promising is to restore the
    System State data from a backup, however, I need more detail as I think that
    restoring the System State data would just restore the same problem.

     
    JNelsonCPath, May 5, 2010
    #3
  4. JNelsonCPath

    JNelsonCPath Guest

    Thanks, please forgive my ignorance, but I am very nervous to do this as it
    doesn't appear to fit my situation. Do you have any more detail you can
    offer me about why this will do the trick in my situation?? The "solution"
    in the event log removes the server from the replication set and then re-adds
    it. Is this OK because I only have a single DC?
    I am also reading way too many posts online about how this procedure harms
    the SYSVOL folder and removes data from it. If I can get a clear explanation
    of how to perform this procedure making sure it wont harm my SYSVOL as well
    as stop the replication set from trying over nad over to reach my old server,
    I will be happy.
     
    JNelsonCPath, May 5, 2010
    #4
  5. Copy the event log error and paste it here.

    I'll the explain what it's saying and what it's going to do.

    This is one of those times you have to stop reading all of those posts.
    We're not them. We don't have a second DC. We're SBS and it picks it
    up from the copy of the sysvol.

    You perform the procedure by sticking the reg key in exactly as it says
    and you let AD fix itself up.
     
    Susan Bradley, May 5, 2010
    #5
  6. JNelsonCPath

    JNelsonCPath Guest

    JNelsonCPath, May 6, 2010
    #6
  7. Yup that's me. If the one burflags value doesn't work you do the other one.
     
    Susan Bradley, May 6, 2010
    #7
  8. JNelsonCPath

    JNelsonCPath Guest

    So are you now telling me to skip the event ID procedure and do the burflags
    procedure??
     
    JNelsonCPath, May 6, 2010
    #8
  9. JNelsonCPath

    Duncan McC Guest

    I would recommend you attempt to fix the journal wrap error on the
    server by setting the Burflags registry entry to D4, and restart NTFRS
    service (and not set the "Enable Journal Wrap Automatic Restore"
    registry parameter").

    Backup your server first.

    Do the procedure. The check the SYSVOL share and Event Logs.
     
    Duncan McC, May 6, 2010
    #9
  10. They should be one and the same.

    What does the event log error say. It gives you the exact steps of what
    to do in the event log error.
     
    Susan Bradley, May 7, 2010
    #10
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.