Problem with SLURPD and multiple databases to single slave (ITS#646)

[jeff.ogden@telecom.co.nz]

Full_Name: Jeff Ogden
Version: 1.2.8
OS: Linux
URL: ftp://ftp.openldap.org/incoming/
Submission from: (NULL) (146.171.254.57)


I have a single OpenLDAP server with two separate LDBM databases on it.  One of
the LDBMs is the root and the other is an OU off that root ( historical reasons
).  This all works perfectly.

I want to replicate the updates to both databases to replicas on a single backup
server.  ( The slave server also has matching separate LDBM databases defined on
it which also both work OK.  )

I have duplicate replica statements in each LDBM defintion. (I am using the same
updatedn ( the rootpw ) for each of the replicas.)  I am using a separate slurpd
log file for each replica.

The problem is I cant get both replicas to update at the same time. Only the
first replica updates, the 2nd does nothing and gives no error messages.  But if
I comment out the slapd.conf statements for the first replica then the second
starts updating the slave.

When I start SLURPD in debug mode it verifies it has read the config succesfuly
and that it is running two replication streams.  The only anomaly I can see is
that the identifier it gives to each stream is the same ( i.e IPaddress:0 ).  (
I have identified the slave using its IP address).  I would have expected the
first stream to be IPaddress:0 and the second IPaddress:1.

Regards

Jeff Ogden