Fedora directory users - Feb 2009 - Replication: How long does it take until replicated servers are in sync after one of them came up....

Hi, 

Consider the following scenario:
You have a replicated environment, let''s say a MMR setup. Now, one of
the masters (M2) is down for a period of time and updates were performed
against the other master (M1). How long does it take until both boxes
are in sync, after M2 is back online. I assume that M1 tries to update
M2 throughout the downtime in an interval setting. If this is right,
what is this interval and secondly, can this be configured.

Thanks,
-Reinhard

Reinhard Nappert wrote:
>
> Hi,
>
> Consider the following scenario:
> You have a replicated environment, let''s say a MMR setup. Now, one
of
> the masters (M2) is down for a period of time and updates were 
> performed against the other master (M1). How long does it take until 
> both boxes are in sync, after M2 is back online. I assume that M1 
> tries to update M2 throughout the downtime in an interval setting. If 
> this is right, what is this interval and secondly, can this be configured.
>
The interval uses a backoff strategy - it first tries after 1 second, 
then after 2 seconds, then after 4 seconds, etc. doubling every time 
until it hits a maximum of every 5 minutes.  This is not configurable, 
but there is a way to send updates now which will break it out of the 
wait/backoff.
>
> Thanks,
> -Reinhard
>
> ------------------------------------------------------------------------
>
> --
> Fedora-directory-users mailing list
> Fedora-directory-users@redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-directory-users
>

Yes,

I know that you can set nsds5ReplicaUpdateSchedule to "0000-2359
0123456", which forces the update, but this would mean that I know that
the other box was down.

Is there are different way to do it?

Thanks,
-Reinhard



-----Original Message-----
From: fedora-directory-users-bounces@redhat.com
[mailto:fedora-directory-users-bounces@redhat.com] On Behalf Of Rich
Megginson
Sent: Monday, February 09, 2009 3:52 PM
To: General discussion list for the Fedora Directory server project.
Subject: Re: [Fedora-directory-users] Replication: How long does it take
untilreplicated servers are in sync after one of them came up....

Reinhard Nappert wrote:
>
> Hi,
>
> Consider the following scenario:
> You have a replicated environment, let''s say a MMR setup. Now, one
of
> the masters (M2) is down for a period of time and updates were 
> performed against the other master (M1). How long does it take until 
> both boxes are in sync, after M2 is back online. I assume that M1 
> tries to update M2 throughout the downtime in an interval setting. If 
> this is right, what is this interval and secondly, can this be
configured.
>
The interval uses a backoff strategy - it first tries after 1 second,
then after 2 seconds, then after 4 seconds, etc. doubling every time
until it hits a maximum of every 5 minutes.  This is not configurable,
but there is a way to send updates now which will break it out of the
wait/backoff.
>
> Thanks,
> -Reinhard
>
> ----------------------------------------------------------------------
> --
>
> --
> Fedora-directory-users mailing list
> Fedora-directory-users@redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-directory-users
>

Reinhard Nappert wrote:
> Yes,
>
> I know that you can set nsds5ReplicaUpdateSchedule to "0000-2359
> 0123456", which forces the update, but this would mean that I know
that
> the other box was down.
>
> Is there are different way to do it?
>   
No.  You would have to have some external agent notice the machine was 
up and tell replication to begin immediate updates.
> Thanks,
> -Reinhard
>
>
>
> -----Original Message-----
> From: fedora-directory-users-bounces@redhat.com
> [mailto:fedora-directory-users-bounces@redhat.com] On Behalf Of Rich
> Megginson
> Sent: Monday, February 09, 2009 3:52 PM
> To: General discussion list for the Fedora Directory server project.
> Subject: Re: [Fedora-directory-users] Replication: How long does it take
> untilreplicated servers are in sync after one of them came up....
>
> Reinhard Nappert wrote:
>   
>> Hi,
>>
>> Consider the following scenario:
>> You have a replicated environment, let''s say a MMR setup. Now,
one of
>> the masters (M2) is down for a period of time and updates were 
>> performed against the other master (M1). How long does it take until 
>> both boxes are in sync, after M2 is back online. I assume that M1 
>> tries to update M2 throughout the downtime in an interval setting. If 
>> this is right, what is this interval and secondly, can this be
>>     
> configured.
>   
> The interval uses a backoff strategy - it first tries after 1 second,
> then after 2 seconds, then after 4 seconds, etc. doubling every time
> until it hits a maximum of every 5 minutes.  This is not configurable,
> but there is a way to send updates now which will break it out of the
> wait/backoff.
>   
>> Thanks,
>> -Reinhard
>>
>> ----------------------------------------------------------------------
>> --
>>
>> --
>> Fedora-directory-users mailing list
>> Fedora-directory-users@redhat.com
>> https://www.redhat.com/mailman/listinfo/fedora-directory-users
>>   
>>     
>
>
> --
> Fedora-directory-users mailing list
> Fedora-directory-users@redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-directory-users
>