I'm migrating our ldap service from Sun One to OpenLDAP. In the
process I have stumbled upon this issue a couple times.
1. Initialize Master
2. Connect syncrepl slaves
3. Wait a week
4. Replication stops working with `null_callback : error code 0x12` in
slave logs and lots of ugly output in the master logs
5. Reinitialize master, GOTO 2
We started with slapd 2.4.7-6ubuntu3, then upgraded to
2.4.9-0ubuntu0.8.04.2 to see if it would help the issue, but it has
returned. I spoke with Howard at LISA'08, and he mentioned this may be
resolved in a newer release. Has anyone else seen an issue like this,
and is it worth my time to backport a newer version to Ubuntu Hardy
(our server platform of choice)?
Master Log: http://pastebin.ubuntu.com/79362/
Slave Log: http://pastebin.ubuntu.com/79348/
Ubuntu Bug Report:
[config bits attached there]
I'll try to ask again. :)
We are want use slapd-meta for aggregate several databases to one DIT.
We are suppose, users will read and write "o=vega" (virtual) suffix.
Members of cn=sysadmins should have write access to all db objects.
Also, we would like to use ACL's per-databases, not global.
Currently, write access to ou=devel doesn't work and we can't find error
in our acls.
Could somebody help us? We can provide any extended information.
suffixmassage "ou=devel,ou=sites,o=vega" "ou=devel"
suffixmassage "o=vega" "o=vega-main"
access to dn.sub="ou=devel"
by * read
We are using FreeBSD 7.0 amd64 and openldap-2.4.11