[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: provider failure behaviour

--On Wednesday, May 05, 2004 5:10 PM +1000 Dave Harrison <d.harrison@isu.usyd.edu.au> wrote:

Hi all,

I'm new to both OpenLDAP and the list so please excuse me if this
question has already been answered.

Having consulted google extensively, and the documentation for OpenLDAP
I can't find documented anywhere, the behaviour of slapd or syncrepl in
the event that the master/provider server were to die.

Do you mean slurpd and syncrepl?

My questions in this scenario are :
    - What is the documented behaviour of the consumer/client ?

If slurpd stops on the master, the client is not affected, but won't receive updates.
If you are using syncrepl, the client will periodically try to contact the master, read slapd.conf on how to set the intervals for retries.

- Does the consumer/client remain query-able ?


    - Will the client periodically attempt a reconnect until it finds
      the master/provider again ?

Yes, if you are using syncRepl.


Quanah Gibson-Mount
Principal Software Developer
ITSS/TSS/Computing Systems
ITSS/TSS/Infrastructure Operations
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html