Hi, [resubmitted post as my original appears to have vanished] Thanks to those who replied to my double colon issue. I have a second issue. I have set-up a provider/consumer pair with syncrepl. Here's the config in my consumer: syncrepl rid=1 provider=ldap://<ip>:389 type=refreshAndPersist searchbase="" filter="(objectclass=*)" scope=sub attrs="" schemachecking=off updatedn="cn=root" bindmethod=simple binddn="cn=root" credentials=foobar retry=5,+ This works perfectly when provider and consumer are both up and have been started off with the same data from slapadd, consumer started with CSN. If I pull the consumer down, however, make a change on the provider, then restart the consumer (without the CSN) it does not pick up the change *until* i make another change to that node. Unfortunately the application wouldn't know to do this, so cannot be expected to 'touch' nodes to force them to replicate in the case of system failure. Is there a confidence inspiring command or log I can use to ensure data integrity? Apologies if this is obvious. I'm afraid I have found the docs which talk about LDAP sync Replication quite hard to follow. I don't intend this to be a flame on the authors, but there is scope here for improvement, specifically in the area titled 14.3.1 where it talks about the various syncProvider and syncConsumer entries and the options that retrieve or generate them and reasons for them: http://www.openldap.org/doc/admin22/syncrepl.html Thanks again, -- Rob Fielding rob@dsvr.net www.dsvr.co.uk Development Designer Servers Business Serve Plc
Attachment:
signature.asc
Description: This is a digitally signed message part