Re: LDAP work at IETF...

On 28/01/2015 14:14, Simo wrote:
> On Tue, 2015-01-27 at 20:50 +0000, Mark R Bannister wrote:
>> On 27/01/2015 15:47, Michael Ströder wrote:
>>> Mark R Bannister wrote:
>>>> The following internet drafts are intended to replace RFC2307 and RFC2307bis.
>>>> [..]
>>>> draft-bannister-dbis-mapping
>>>> draft-bannister-dbis-netgroup
>>>> draft-bannister-dbis-passwd
>>>> draft-bannister-dbis-hosts
>>>> draft-bannister-dbis-devices
>>>> draft-bannister-dbis-automounter
>>>> draft-bannister-dbis-custom
>>>> [..]
>>>> I would prefer to see RFC2307bis removed from the list to be considered above,
>>>> and replaced by the DBIS drafts.
>>> This probably won't happen because there are so many RFC2307(bis) deployments
>>> out there which cannot easily be migrated.
>> DBIS supports maps defined in legacy RFC2307 or RFC2307bis format, as
>> well as using the new format.  Migration is designed to be extremely
>> easy.  In fact, it took me next to no time at all to start using
>> directory entries defined in RFC2307 format from a DBIS installation, it
>> really is just a matter of defining a few configuration maps that do the
>> class and attribute remappings, and they are mostly standard.  See
>> https://sourceforge.net/p/dbis/wiki/ConfigurationMaps-RFC2307/
>>
>>> Also I have my own schema built on top of RFC2307bis (not published yet) which
>>> differs much from DBIS for very good reasons. I aim to present it at LDAPcon
>>> 2015 in Edinburgh (in case the program comittee agrees). A preview version of
>>> that is already in production for almost a year in a large data center.
>>>
>>> So my suggestion would be to brush up RFC2307bis as much as possible (taking
>>> care of Kurt's IANA comments) so that you can base DBIS on that, as I will
>>> base my system on top of that.
>>>
>>> How does that sound to you?
>>>
>>> Ciao, Michael.
>>>
>> DBIS is not a superset nor a child schema that inherits from RFC2307bis,
>> nor is it built on top of RFC2307bis - it is a complete replacement and
>> was designed as such.  RFC2307 and RFC2307bis can be completely retired
>> and DBIS put in their place and there would be no omissions.  It's an
>> evolution that began life by taking RFC2307bis and splitting it into
>> logic units (separate internet drafts) that make for a less unwieldy
>> document and make it easier for others to extend.  Then I standardised
>> the client-side configuration, fixed some of the NIS incompatibilities
>> that had been inherited from RFC2307, added a few much needed enterprise
>> features, and that's it: a modern, modular, extensible, enterprise-ready
>> RFC2307bis under another name ("2307bis" wasn't a very friendly name,
>> after all, and I wasn't going to call my improvements "2307ter", but
>> they could have been called that, because realistically that's what they
>> are).
>>
>>> So my suggestion would be to brush up RFC2307bis as much as possible (taking
>>> care of Kurt's IANA comments) so that you can base DBIS on that, as I will
>>> base my system on top of that.
>>>
>>> How does that sound to you?
>>>
>>> Ciao, Michael.
>>>
>> I don't see a reason to try to improve RFC2307bis, I think that would be
>> a backwards step.  The improvements needed to RFC2307bis have already
>> been made, I've done the work myself and named it "DBIS" because I
>> couldn't wait for everyone else to catch-up, and I needed it last year
>> :-)  I think if people started looking deeper into the DBIS internet
>> drafts, making suggestions for improvement etc. we'd get to a much
>> better end point than if you have people pore over RFC2307bis again.
> It hurts me to curb enthusiasm, but I think your drafts are not a step
> forward, at most a step sideways, and ignore what's out there right now.

Given that they were written to fix specific deficiencies in RFC2307bis 
that were causing
pain in a number of very large enterprises I have worked for, I don't 
see how they
could be considered a step sideways.

When you say my drafts ignore what's out there right now, what are they 
ignoring?  I
don't quite understand that point.  I'm very open to make changes that 
would make
the drafts more useful in other organisations too, but I cannot make 
improvements
when faced with generic statements that have no detail behind them.

> The software out there and now expect either RFC2307 or at most
> RFC2307bis (and then really they just test against Active Directory's
> schema and things happen to mostly work). Any other schema would just
> fail to be useful for a generic LDAP server.
>
> Simo.
>
To what software do you refer specifically?  NSS is a big subscriber, 
and I've addressed
that one.  Then there is the automounter, and sudo, and really only a 
handful of
others all of which can be fixed to use a more modern schema.

I'm not going to spend 18 months developing DBIS to be turned away 
because software
doesn't support it.  Of course software doesn't support it - yet - it's new!

Best regards,
Mark.

_______________________________________________
Ldapext mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ldapext