[389-users] Replication

[crashingdaily]

On Jan 11, 2011, at 3:50 PM, Rich Megginson wrote:

> On 01/11/2011 01:40 PM, harry.devine at faa.gov wrote:
>>
>> We just ran the mmr.pl script from server 1, and the data got  
>> replicated across to server 2.  If we create a new user on server  
>> 2, it does not get replicated to server 1.  It appears to be a 1  
>> way replication.
> mmr.pl has bit rotted - no one maintains it any more - you're on  
> your own if you decide to use it - patches/maintainer welcome


I've successfully used mmr.pl (last time with 389 1.2.6.1). I would  
try restarting your two 389 instances, and if that doesn't magically  
fix it, use the mmr.pl script to --remove the existing replication  
agreement and then add it back while keeping an eye on the 389 logs  
for errors. I've had rare cases where I've had to run mmr.pl twice to  
get it to take.

The --with-ssl option was broken in the last version I used (but  
that's clearly not the cause of Harry's issue). For working patched  
version, should you want SSL replication, see
http://crashingdaily.wordpress.com/2010/11/13/fixing-the-mmr-pl-script-for-ssl/



>>
>> Any thoughts?
>> Harry
>>
>> Harry Devine
>> Common ARTS Software Development
>> AJT-144
>> (609)485-4218
>> Harry.Devine at faa.gov
>>
>> -----Rich Megginson <rmeggins at redhat.com> wrote: -----
>>
>> To: "General discussion list for the 389 Directory server project."  
>> <389-users at lists.fedoraproject.org>
>> From: Rich Megginson <rmeggins at redhat.com>
>> Date: 01/11/2011 03:09PM
>> cc: Harry Devine/ACT/FAA at FAA, Ted Rush/ACT/FAA at FAA
>> Subject: Re: [389-users] Replication
>>
>> On 01/11/2011 12:53 PM, harry.devine at faa.gov wrote:
>>>
>>> OK, so we are testing 389-ds in our environment (version  
>>> 1.2.7.5).  We have 1 machine set up with user accounts, and we  
>>> just set another one up.  The new one doesn't have any accounts or  
>>> anything on it yet.  We would like to get replication going  
>>> between the 2 DS's, but can't find any current documention on how  
>>> to do it.  When we set up the 2nd machine, we set it up as if it  
>>> were its own DS (i.e. not tied into the existing one).  Not sure  
>>> it that is relevant, but I thought I'd mention it for full  
>>> disclosure.
>> Should not be a problem.
>>>
>>> We'd like to have this be a 2-way multi-master replication, so if  
>>> our main DS server ever went down, the 2nd one could pick up and  
>>> service login requests.  Is there any current documention that  
>>> explains this anywhere?
>>
>> http://docs.redhat.com/docs/en-US/Red_Hat_Directory_Server/8.2/html-single/Administration_Guide/index.html#Managing_Replication
>>> Is multi-master the correct method to use for this type of setup?
>> Yes.
>>>
>>> Thanks,
>>> Harry
>>>
>>> Harry Devine
>>> Common ARTS Software Development
>>> AJT-144
>>> (609)485-4218
>>> Harry.Devine at faa.gov
>>>
>>> --
>>> 389 users mailing list
>>> 389-users at lists.fedoraproject.org
>>> https://admin.fedoraproject.org/mailman/listinfo/389-users
>>
>>
>
> --
> 389 users mailing list
> 389-users at lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/389-users