Thanks for the comment. I would agree with what you say that deleting
the files shouldn't fix it but I'm pleased to report that it did! I
started just deleting a couple of files; a similar number (but not
exactly the same number) went from the staging folder and some time
later they went from the list of violations in the ultrasound database.
I got braver and deleted all the files from the hub and (after about 24
hours) all of the sharing violations had gone as had the files in the
staging area on the server which was showing sharing violations.

I've now copied back the problem files and they've replicated fine so
everyone's happy.

I don't understand why this happens - I would have thought that tasks
were queued as you say - but presumably NTFRS is intelligent enough to
process deletes ahead of writes???

I'd also like to know what I can do if this happens again but I'd guess
I'll just wait and hope it doesn't :-)

Steve 

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] 
Sent: 19 July 2004 22:25
To: [EMAIL PROTECTED]
Subject: RE: [ActiveDir] Troubleshooting NTFRS replication

Keep thinking there was a post SP4 hotfix that addresses the sharing
violation problems.  Preliminary search hasn't turned up anything
though.  I'll keep looking... we get that on one or two DCs here and
there.

Even if you delete the files on the hub server, if those files are
registering sharing violations, it's still going to be the same thing,
even for deletions.  NTFRS works sequentially based on the changes in
the journal...

-m

-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Steve Rochford
Sent: Saturday, July 17, 2004 6:42 AM
To: [EMAIL PROTECTED]
Subject: [ActiveDir] Troubleshooting NTFRS replication

I've got a DFS replica set which is basically working but Ultrasound
tells me that I have 757 files in sharing violation for more than a day
(there are over 400,000 files in total). Looking in the frs-staging
folder for the relevant server there are 699 files which are time
stamped at the times shown by Ultrasound and the replica folder on that
server does have at least some of the files listed in Ultrasound
missing.

I've looked at KB 822300 but it doesn't really help. I can't use
"install override" because this server is running Windows 2000.

I've used both process explorer and handle from sysinternals and, as far
as I can tell, none of the files are being held open on either the
source or the destination server (and I'm assuming I only really need to
look at the destination - the files are in the frs-staging folder there)

I've tried stopping and restarting ntfrs; I've even restarted the server
in case something was somehow holding the files open but was hidden.

What can I try? One thing which occurs to me is to delete the files
which are failing to replicate from the "hub" server (the replica set is
configured as hub and spoke and this server is one of the spokes) and
see if that clears the problem. I could then just restore the missing
files.

Steve
List info   : http://www.activedir.org/mail_list.htm
List FAQ    : http://www.activedir.org/list_faq.htm
List archive:
http://www.mail-archive.com/activedir%40mail.activedir.org/

List info   : http://www.activedir.org/mail_list.htm
List FAQ    : http://www.activedir.org/list_faq.htm
List archive:
http://www.mail-archive.com/activedir%40mail.activedir.org/
List info   : http://www.activedir.org/mail_list.htm
List FAQ    : http://www.activedir.org/list_faq.htm
List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/

Reply via email to