Re: [389-users] Strange replication error

[Dumbo Q]

I think I must have goofed something up with the replica ID.  I thought I recalled it be
set to 65535, but then changing it to 1.  However looking at the idm console it
doesn't appear to have an option to change it. My best assumption is that I botched it
up, but it didn't break until i restarted it.

I deleted both replication agreements, and recreated them (reinitializing one of the
servers).  It seems ok now.  It survived a reboot.


________________________________
From: Rich Megginson <rmeggins(a)redhat.com&gt;
To: General discussion list for the 389 Directory server project.
<fedora-directory-users(a)redhat.com&gt;
Sent: Monday, July 13, 2009 11:10:24 AM
Subject: Re: [389-users] Strange replication error

Dumbo Q wrote:
>  I have 2 servers setup for MMR. It seemed to be working fine
> (although I've only had it running for a few hours).  Today I installed a new SSL
> certificate on both servers.  They both came back up fine, and SSL is working perfectly.
>  
>  However I noticed that replication has stopped.  Here is the error message that I am
> getting.  I've looked around and can't find any information about it.  I imagine I
> could probably reinitialize, but I would really like to know what went wrong.
>  
>  
>  
>  [10/Jul/2009:11:22:13 -0400] - CentOS-Directory/8.1.0 B2009.134.1334 starting up
>  [10/Jul/2009:11:22:13 -0400] - I'm resizing my cache now...cache was 20000000 and is
> now 8000000
>  [10/Jul/2009:11:22:13 -0400] - skipping cos definition
> cn=nsAccountInactivation_cos,dc=mydomain,dc=com--no templates found
>  [10/Jul/2009:11:22:13 -0400] - _csngen_parse_state: replica id mismatch; current id - 1,
> replica id in the state - 65535 
The only thing that changed is the SSL certificate? 
Because this error appears to be caused by some really weird configuration problem, as if
you replaced a working replication configuration with something else.
>  [10/Jul/2009:11:22:13 -0400] NSMMReplicationPlugin -
> _replica_init_from_config: failed to create csn generator for replica
> (cn=replica,cn=\22dc=mydomain, dc=com\22,cn=mapping tree,cn=config) 
Same as agove.
>  [10/Jul/2009:11:22:13 -0400] NSMMReplicationPlugin - Unable to
> configure replica dc=mydomain, dc=com: failed to create csn generator for replica
> (cn=replica,cn=\22dc=mydomain, dc=com\22,cn=mapping tree,cn=config) 
Same as above.
>  [10/Jul/2009:11:22:13 -0400] - skipping cos definition
> cn=nsAccountInactivation_cos,dc=mydomain,dc=com--no templates found
>  [10/Jul/2009:11:22:13 -0400] - slapd started.  Listening on All Interfaces port 389 for
> LDAP requests
>  [10/Jul/2009:11:22:13 -0400] - Listening on All Interfaces port 636 for LDAPS requests
>  [10/Jul/2009:12:08:52 -0400] NSMMReplicationPlugin - conn=18 op=3
> replica="unknown": Unable to acquire replica: error: no such replica 
I
have no idea what happened - I've never seen this before.  I'm not really sure
what you can do except to just start over.  You'll have to shutdown the servers,
remove all cn=replica entries and their children from cn=config (by editing dse.ldif with
a text editor - be sure to make a backup first), then start up the servers, then configure
replication from scratch.
>  
>  
>  ------------------------------------------------------------------------
>  
>  --
>  389 users mailing list
>  389-users(a)redhat.com
>  https://www.redhat.com/mailman/listinfo/fedora-directory-users
>