Hi Rich,
 
Ooops!  I totally missed your reply to my question.
 
Yes, our system is configured using multi master replication.
 
Umm... the scenario you described could happen in our system.  We are currently conducting some isolated network testing.
 
What we do is to put couple nodes as an island and have them operate separately.  Each island consists of a LDAP.
 
When all the islands are connected together again, we are hoping that the LDAP will automatically resync and I guess that's when the problem occur.
 
I will look into that.  Thanks for your insight.
 
- David

On Tue, Sep 28, 2010 at 4:33 PM, Rich Megginson <rmeggins@redhat.com> wrote:
Chun Tat David Chu wrote:
> Hi all,
>
> I am hitting some problem with my replicated directory server.  I saw
> the following error messages from the errors log file.
>
> [17/Sep/2010:09:47:51 -0400] NSMMReplicationPlugin - conn=1 op=4
> csn=4c9363a9000000020000: Can't resurrect tombstone
> ou=test,dc=example,dc=com to glue reason 'deletedEntryHasChildren',
> error=68
>
> Can anyone tell me what could cause this problem and how to prevent it
> from happening again?
Looks like you deleted an entry on one master, but at the same time, on
another master, you added child entries to the same entry.  I'm not sure
how this happened - was one master down for some period of time, or
disconnected for some period of time, and managed separately during that
time?

You'll have to delete all of the child entries of
ou=test,dc=example,dc=com, then you'll have to delete
ou=test,dc=example,dc=com
>
> Thanks!
>
> David
>
> ------------------------------------------------------------------------
>
> --
> 389 users mailing list
> 389-users@lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/389-users

--
389 users mailing list
389-users@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/389-users