On 11/05/2015 20:48, Quanah Gibson-Mount wrote:
See ITS#8100: http://www.openldap.org/its/index.cgi/?findid=8100
You need to do at least one modification on the other MMR server.
Doesn't make any difference unfortunately. The contextCSN's duly get updated but CPU use remains flat-out at both ends.
Did you restart both servers after the update?
I tried again this morning with a fresh set of servers, and modifying a record sorted them out without a restart being required.
This is good enough for me - should it not work out, given the size of our database then initialising all peers from the same LDIF will do too.
Thanks for your help, this has been driving me nuts over the last week or so.