El sáb, 25-04-2009 a las 22:38 +0200, John S. Skogtvedt escribió: > Holger Levsen skrev: > > Is there anon svn checkout available? If not, could you make it available, > > please? Alternativly and not really a solution IMO, would be to give me > > normal svn access... I'd like to upload lwat to our lenny-test so we can use > > your changes.. > > > > This should work: > $ svn co http://lwat.org/svn/trunk lwat > > > >> I've attached a diff with proposed changes to to dhcp.ldif in > >> debian-edu-config/ldap-bootstrap, which is intended to make debian-edu > >> work better with the lwat DHCP changes. > > > > has this been included in out schema now? > > > > From a quick look at ldap-bootstrap/dhcp.ldif, the answer seems to be no. I've been testing lwat 0.18~beta-2 and almost everything seems to go perfectly in the servers we have in production since last december. The 'almost" is due to a strange behaviour when adding machines to the dns domain. The structure we have is the same you can see at http://wiki.debian.org/DebianEdu/Lwat?action=AttachFile&do=get&target=dns-ldap.png In this case the domain is 'santaeulalia'. You can see that the hosts are added under the domain name branch. But with lwat, it adds them under the ou=hosts branch. This is the diference: cn=test-pro,ou=Hosts,dc=instituto,dc=extremadura,dc=es cn=test-pro,dc=domain,ou=Hosts,dc=instituto,dc=extremadura,dc=es For the reverse addresses it handles it perfectly, adding them under the dc=in.addr,dc=... I don't understand why the machines are added under ou=hosts. In terms of clarity when you have many machines, I think this is not the right behaviour. I'd like to know if this is a bug or a design decision. In the later case I'd like to know the advantages of using that structure instead of the proposed at http://wiki.debian.org/DebianEdu/Lwat . Regards. José L.
Attachment:
signature.asc
Description: Esta parte del mensaje =?ISO-8859-1?Q?est=E1?= firmada digitalmente