On 02/09/2011 06:39 AM, remy d1 wrote:
Hi Rich,
I reinstalled all my server from scratch and reimported all my data
(with cert files).
If I try to synchronize my data, I can import users from AD to 389-DS
but I can't do the opposite. My 389 server replica is always in status
"in progress" with "replica acquired successfully : incremental update
started", but it can't finish the synchronization job.
Sometimes you have to tell winsync to do a full resync a few times
before it finally works.
I could also continue to launch request to my AD server from my 389-DS
server (ldapsearch...). I successfully add a user to my AD with Apache
Directory Studio (installed on my 389-DS server) with the AD
synchronizing account. So, it's not an access problem.
Moreover I added a schema on my 389-DS for my directory that is not
present on my AD. Do you think I have to add this schema to AD or is
the synchronization done only on AD required attributes ?
No. The schema that winsync uses is hard coded in 389 - you cannot
extend it or change it - it should work with AD, no changes to AD should
be required.
Or,
Is it a cert file problem on my AD ?
or ...?
Any idea would be appreciated
Regards-
2011/1/25 Rich Megginson <rmegg...@redhat.com
<mailto:rmegg...@redhat.com>>
On 01/25/2011 01:29 AM, remy d1 wrote:
Hi Rich,
I tried to raise the log level, but when I did it, I was not able
to stop/restart my dirsrv service.
What log level did you use? What error messages did you see when
you attempted to stop/restart the service? Anything in the errors
log?
To stop it, I must kill the process and remove the pid file. Then
I could start it.
In my error logs, there is a lot of informations :
[root@KingKong ~]# tail /var/log/dirsrv/slapd-KingKong/errors
[24/Jan/2011:16:18:30 +0100] NSMMReplicationPlugin - changelog
program - cl5GetOperationCount: could not get DB object for replica
[24/Jan/2011:16:18:30 +0100] NSMMReplicationPlugin - changelog
program - _cl5GetDBFile: no DB object found for database
/var/lib/dirsrv/slapd-KingKong/changelogdb/1d934402-27b111e0-b651ef2e-02b602d3_4d0b28870000ffff0000.db4
[24/Jan/2011:16:18:30 +0100] NSMMReplicationPlugin - changelog
program - cl5GetOperationCount: could not get DB object for replica
[24/Jan/2011:16:18:40 +0100] NSMMReplicationPlugin - changelog
program - _cl5GetDBFile: no DB object found for database
/var/lib/dirsrv/slapd-KingKong/changelogdb/1d934402-27b111e0-b651ef2e-02b602d3_4d0b28870000ffff0000.db4
[24/Jan/2011:16:18:40 +0100] NSMMReplicationPlugin - changelog
program - cl5GetOperationCount: could not get DB object for replica
[24/Jan/2011:16:18:41 +0100] NSMMReplicationPlugin - changelog
program - _cl5GetDBFile: no DB object found for database
/var/lib/dirsrv/slapd-KingKong/changelogdb/1d934402-27b111e0-b651ef2e-02b602d3_4d0b28870000ffff0000.db4
[24/Jan/2011:16:18:41 +0100] NSMMReplicationPlugin - changelog
program - cl5GetOperationCount: could not get DB object for replica
[24/Jan/2011:16:18:42 +0100] NSMMReplicationPlugin - changelog
program - _cl5GetDBFile: no DB object found for database
/var/lib/dirsrv/slapd-KingKong/changelogdb/1d934402-27b111e0-b651ef2e-02b602d3_4d0b28870000ffff0000.db4
[24/Jan/2011:16:18:42 +0100] NSMMReplicationPlugin - changelog
program - cl5GetOperationCount: could not get DB object for replica
[24/Jan/2011:16:24:18 +0100] NSMMReplicationPlugin - changelog
program - cl5ExportLDIF: failed to locate changelog file for
replica at (dc=mydomain,dc=com)
This problem is very similar to this post :
http://www.redhat.com/archives/fedora-directory-commits/2009-March/msg00005.html
Although I have the last version of 389-DS.
Are you sure this is the correct post you wanted to refer to?
Because this is a patch commit for a fix when moving the changelog
directory - did you move the changelog directory? Because you did
not mention it in your earlier post.
I think I have also some troubleshooting with my hostname because
bind is not configured. However, I have choosen to put it my
/etc/hosts file
[root@KingKong ~]# nl /etc/host.conf
1 multi on
2 order hosts,bind
hostname command reply the full "fqdn" if I choose the option
--all-fqdn, contrary to the option "--fqdn". The reply is just my
hostname without the domain. By the way, if I say
#hostname KingKong.mydomain.com <http://KingKong.mydomain.com>
Eveything is now good for my hostname but I can not launch my
389-console. I think the adress to connect is not ok... I do not
know if this problem is linked to the previous problems...
So, I do #hostname KingKong
Then, I launch the console again. Now, if I try to initiate a
full synchronization, I can see (and I am still stuck on it) the
window "please wait while data is being synchronized...", but
nothing else... Data are not synchronized and I do not see
anything in my Windows event viewer while replica agreement seems
to be ok and PassSync service is installed...
It is very difficult to change your hostname after you have
configured the admin server and console. I suggest starting over
from scratch, and first make sure your hostname is correct.
I also suggest using
http://docs.redhat.com/docs/en-US/Red_Hat_Directory_Server/8.2/html-single/Administration_Guide/index.html#Windows_Sync
to configure Windows Sync.
Thanks for help,
-Regards
2011/1/21 Rich Megginson <rmegg...@redhat.com
<mailto:rmegg...@redhat.com>>
Date:
Fri, 21 Jan 2011 10:25:56 +0100
To:
"General discussion list for the 389 Directory server
project." <389-us...@lists.fedoraproject.org>
<mailto:389-us...@lists.fedoraproject.org>
Hi Rich,
Thanks for this usefull link.
I have successfully initiate replica between Windows AD and
my server 389-DS. Ldapsearch is working. But even if
everything seems to be ok, the update does not work and I do
not see any error in the log files... So, my AD server stay
empty, the accounts are not migrate...
Here you have my access log file which is more verbose...
(mydomain.com <http://mydomain.com> for the example) :
<snip>
Obviously I am connecting to the server 389-DS itself
whereas it can resolve the DNS name of my Windows server...
There is no error in the AD event viewer while I could see
errors on it when it was misconfigured (like DirSync
errors)... So, basically, the Windows server is contacted to
my DS-Server over 2 different networks.
Do you think I have to open the ports described in my message ?
-Regards.
I don't know. There is no winsync information in the access
log. Note that the access log records client accesses to the
directory server, and in winsync, the directory server itself
acts as a client to AD, so winsync will log nothing in the
access log. The errors log could be helpful, and especially
using the replication log level (which is also used for
winsync logging). The Windows Event Viewer is useless for
winsync issues.
--
389 users mailing list
389-us...@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/389-users
