You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@directory.apache.org by Emmanuel Lécharny <el...@gmail.com> on 2012/11/15 01:11:21 UTC
Replication, consummers and producers question
Hi,
as I'm currently looking at the code, I'm wondering if we should store
the consummers configuration into the producer. It seems like every
server that has consumers is setting up a persistent search for each
consumer, evn if the consumer are not yet connected.
Wouldn't be better to just create the persistentSearch when a consumer
actually conntects to the master server, and to avoid storing such
information into the server ? (Just wondering)
--
Regards,
Cordialement,
Emmanuel Lécharny
www.iktek.com
Re: Replication, consummers and producers question
Posted by Kiran Ayyagari <ka...@apache.org>.
this is done to handle the events that occurred while the consumer was offline
(either in refreshOnly or refreshNPersist mode). There is another way to handle
without maintaining this history but we choose to maintain history.
On Thu, Nov 15, 2012 at 5:41 AM, Emmanuel Lécharny <el...@gmail.com> wrote:
> Hi,
>
> as I'm currently looking at the code, I'm wondering if we should store
> the consummers configuration into the producer. It seems like every
> server that has consumers is setting up a persistent search for each
> consumer, evn if the consumer are not yet connected.
>
> Wouldn't be better to just create the persistentSearch when a consumer
> actually conntects to the master server, and to avoid storing such
> information into the server ? (Just wondering)
>
> --
> Regards,
> Cordialement,
> Emmanuel Lécharny
> www.iktek.com
>
--
Kiran Ayyagari
http://keydap.com
Re: Replication, consummers and producers question
Posted by Howard Chu <hy...@symas.com>.
Emmanuel Lécharny wrote:
> Hi,
>
> as I'm currently looking at the code, I'm wondering if we should store
> the consummers configuration into the producer. It seems like every
> server that has consumers is setting up a persistent search for each
> consumer, evn if the consumer are not yet connected.
>
> Wouldn't be better to just create the persistentSearch when a consumer
> actually conntects to the master server, and to avoid storing such
> information into the server ? (Just wondering)
>
IMO: definitely. One of the reasons syncrepl is designed as it is is because
you don't need to change the provider's config when you add or remove
consumers. (Of course, this was a greater concern back before we had dynamic
config for OpenLDAP.) Only the actual consumer should need to know about its
replication configuration.
--
-- Howard Chu
CTO, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc/
Chief Architect, OpenLDAP http://www.openldap.org/project/