Re: v2.2.24 structural object class modification not allowed

[Curt Blank <curt@uwm.edu>]

Ya know, that is a d*mn good question, I really had to stop and think 
about that. Best answer I can come up with is stay with what is current 
and supported and going to the Berkeley backend does allow us to run 
cached and get better performance. So I guess the bottom line to upgrade 
was better performance.

Back to the topic at hand, I'm really trying to understand this. I 
thought I asked a while ago if theses three new internal attributes, one 
being structualObjectclass, meant anything to me and I thought I was 
told no their internal. But obviously now this one means something to me 
if an application cannot add a user because the value of 
structualObjectclass needs to be changed for that to happen and that is 
not allowed.

This is what I currently have.

top->person->organizationalPerson->inetOrgPerson->uwmPerson->ctCalUser

objectclass ( 2.5.6.6 NAME 'person'
       SUP top STRUCTURAL
objectclass ( 2.5.6.7 NAME 'organizationalPerson'
       SUP person STRUCTURAL
objectclass ( 2.16.840.1.113730.3.2.2 NAME 'inetOrgPerson'
       SUP organizationalPerson
objectclass ( 1.3.6.1.4.1.4170.1.1.25 NAME 'uwmPerson'
       SUP inetOrgPerson STRUCTURAL
objectclass ( 1.3.6.1.4.1.2672.3.3 NAME 'ctCalUser'
       SUP uwmPerson STRUCTURAL

When a new user gets added to LDAP it appears structualObjectclass get 
sets to uwmPerson, then when the Calendar application tries to give that 
person a Calendar account slapd wants to change structualObjectclass to 
ctCalUser and can't so it sends a failure status to the Calendar server 
and the Calendar server fails to add the account.

Now, if structualObjectclass get sets to some value at the time a dn is 
added and this restriction exists how can schema ever be extended?

Bottom line, how do I fix this? And allow for future schema additions?



Pierangelo Masarati wrote:

> > Yeah well that's all fine and dandy but 2.0.27 didn't have this problem
> > and now because slapd(8) enforces that this problem exists.
> >    
>
> Then why don't you keep using 2.0.27?
>
> p.
>
>