LDB: schema syntax comparison

Stefan (metze) Metzmacher metze at samba.org
Mon Sep 9 17:13:28 CEST 2013

Am 08.09.2013 17:24, schrieb Simo:
> On Sat, 2013-09-07 at 13:36 +0200, Matthias Dieter Wallnöfer wrote:
>> Hi Simo,
>> I simply tried to revert this specific patch and then the following 
>> exception is raised during provisioning:
>>> ERROR(ldb): uncaught exception - attribute 'dn': value #0 on 
>>> '@ATTRIBUTES' provided more than once
>> And I think this is index-entry related?
> Not necessarily, in any case it is a bug that needs to be addressed
> properly.

The problem is triggered by ldb_setup_wellknown_attributes() setting up
{ "dn", LDB_SYNTAX_DN },

I'll propose a fix in the next days...


> Simo.
>> Matthias
>> Simo schrieb:
>>> On Thu, 2013-09-05 at 16:03 +0200, Matthias Dieter Wallnöfer wrote:
>>>> Hi Simo & others,
>>>> I do not remember the exact problem, but I think it would be very
>>>> tedious to enforce 100%-correct schema validation for index entries as well.
>>>> And then: does comparison strictly require preliminar validation? My
>>>> opinion in this case is no, since if something is comparable let us just
>>>> do the comparison. But okay, this is a little philosophical.
>>> If index DNs are validated and canonicalized before being written, then
>>> I guess it really doesn't make much sense re-validating on read given
>>> the performance penalties. Do we have the guarantee they are properly
>>> vetted before being written ?
>>> I ask because the comment in the bug seem to implay we have 'bad'
>>> provision creating some bad index, and that is certainly not acceptable
>>> IMO.
>>> Simo.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 261 bytes
Desc: OpenPGP digital signature
URL: <http://lists.samba.org/pipermail/samba-technical/attachments/20130909/ad3eb5ee/attachment.pgp>

More information about the samba-technical mailing list