Package 2.1.13 -> Re: Segmentation fault on 2.1.13

[Jehan PROCACCIA <Jehan.Procaccia@int-evry.fr>]

Great ! it finally work :-)

I've just compiled RPMS under RH 8.0 , I put them + source RPM in :
http://www.int-evry.fr/mci/user/procacci/SRPMS/ -> openldap-2.1.13-3
Thanks to Frank Swasey an openldap developpers it now includes the patch 
about schema checking problem with attributes in rdn not present in 
entry which caused a seg fault message !

%changelog

* Thu Feb 28 2003 Jehan Procaccia <jehan.procaccia@int-evry.fr> 2.1.13-3
- from Francis Swasey <Frank.Swasey@uvm.edu> 2.1.12-uvm2.8.0:
- sleepycat bdb 4.1.25 plus locking patch
- add patch for schema-check
- add patch for slapd lock-release in group acl processing
- add patch for slapcat a subtree
- add make test to openldap build

Now insteed of seg fault it says:

<= str2entry(cn=brou,ou=People,dc=enstb,dc=fr) -> 0x823aba0
...
slapadd: dn="cn=brou,ou=People,dc=enstb,dc=fr" (line=19): (16) value of 
naming attribute 'cn' is not present in entry

ldif beeing:

dn: cn=brou,ou=People,dc=enstb,dc=fr
cn: Jean Brou
sn: Brou
objectClass: top
objectClass: person
objectClass: inetOrgPerson
mail: Jean.Brou@enstb.fr
givenName: Jean

now if I put in the ldif:
sn=brou,ou=People,dc=enstb,dc=fr

<= index_entry_add( 3, "sn=brou,ou=People,dc=enstb,dc=fr" ) success
:-)

However the error message could be even more explicit :
"(16) value of naming attribute 'cn' is not present in entry" + add " or 
does not have the same value in the entry than in the dn ..." which was 
my case !.

By the way, why this new contraint ? in respect to LDAP RFCs ?

Thanks.




Frank Swasey wrote:
> Today at 6:30pm, Pierangelo Masarati wrote:
>
>
> > > Thanks!
> > >
> > > I'm rebuilding now with that patch applied to schema_check.c.
> > >
> > > I am curious about what part of the schema I am violating that was legal
> > > (or possibly just not checked) in 2.1.12.  But, time will tell -- slow
> > > machine takes a while to build the RPM :-)
> >
> > OpenLDAP never checked that the attributeTypes and values used
> > in the RDN were actually present in the entry.  This was added
> > in 2.1.13 (maybe it was not adequately advertised ...).
> >
> > For details, you may follow the thread originated by ITS#2243.
>
>
> Well, my ldif file isn't supposed to have any entries that violate 
> that.... but looking at that one entry.... Drat!  more crud to go fix... 
> :-)
>
> Thanks,
>   F



--
Jehan Procaccia 			| Ingenieur Systemes & Reseaux
Institut National des Telecommunications| Tel : +33 (0) 160764436
MCI, Moyens Communs Informatiques  | Mail: Jehan.Procaccia@int-evry.fr
9 rue Charles Fourier 91011 Evry France | Fax : +33 (0) 160764321