[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
Re: (c.harding 30612) RE: (c.harding 30479) RE: (c.harding 30393) Re: Tw bobs worth on TOP, LDAP standards process and subschemasub entry attribute usage in rootDSE
Dear Alan,
I agree that the more things you can test, the better. The trouble is
that it is impossible to test everything, and the more thorough the
tests are, the more expensive they are to produce. That is why there
must be a commitment to conform even to those parts of the standards
that are not tested!
>Chris - thanks for that. Its just that testing, which I am always for,
>sometimes misses the real mark because it is only based on protocols and
>PICS etc.
>
>eg. Our brilliant distributed directory service ( :-) ) can, in each DSA
>support (theoretically) attribute values of 16k * 128mb, 16k attribute
>values per attribute, 16k attributes per entry and 4.25 billion entries
>per DSA and NNNN DSAs can interconnect.
> And we have seen some "LDAP servers" explode when adding entries with
>attribute values of over 10 mb.
>
>So a real operational test would be read a massive entry from one server
>and write it to another via LDAP.
>
>Other tests like unplugging the power during updates also shows up a few
>issues.
>
>IMHO Operational compatability tests should be included in the test/
>conformance suite as it is this area that determines a product or just a
>flimsy-theoretical protocol only implementation.
>
>regards alan
>
>----------
>From: Chris Harding
>To: Alan Lloyd
>Cc: ''ietf-ldapext@netscape.com ' '; 'directory@opengroup.org '; 'Mark
>Wahl '; 'd.w.chadwick@iti.salford.ac.uk '
>Sent: 7/15/99 4:44:58 AM
>Subject: Re: (c.harding 30479) RE: (c.harding 30393) Re: Tw bobs worth
>on TOP, LDAP standards process and subschemasub entry attribute usage in
>rootDSE
>
>Dear Alan,
>
>The strength, as well as the technical coverage, of our LDAP
>certification scheme is something that is still to be discussed.
>So I can only talk about how previous schemes have worked.
>
>The real basis on which previous Open Group conformance schemes have
>relied is not the test suite but the vendor's commitment to conform
>to the standards. This is actually much stronger. No test suite
>in practice ever covers more than a fraction of the possible cases,
>while the commitment does cover everything. The test suite is
>extremely valuable in showing that the vendor conforms as regards the
>cases that it does cover, but it is important that the vendor conforms
>as regards the other cases too.
>
>The real issue is what standards a branded LDAP client or server
>can inter-operate using. What is covered by the test suite is
>secondary (but still important).
>
>>I hope the conformance testing goes a bit beyond that of the LDAP
>>client-server actions of log on a read via LDAP a few common
>attributes.
>>This issue of corrupt TOP definitions will affect the creation of
>>entries, what one gets back if one reads all attributes, the definition
>>of access controls, the configuration approaches of directory
>management
>>clients, the definition of filters and entry information selection for
>>searches, the relationships of attributes within and between entries
>and
>>of course the LDAP replicate everything to everywhere requirements
>>(schema mapping)
>>
>>What will be interesting is when one LDAP server with a corrupted TOP
>>refers to another sever with a different corruption of TOP and what the
>>entry selection will make of it...
>>
>>There is nothing like an inconsistent user interface and inconsistent
>>operations management interface caused by baseline corrupted
>definitions
>>:-(
>>
>>regards alan
>>
>>
>>
>>----------
>>From: Chris Harding
>>To: Mark Wahl; d.w.chadwick@iti.salford.ac.uk
>>Cc: 'ietf-ldapext@netscape.com '; directory@opengroup.org
>>Sent: 7/12/99 6:14:32 PM
>>Subject: Re: (c.harding 30393) Re: Tw bobs worth on TOP, LDAP standards
>>process and subschemasub entry attribute usage in rootDSE
>>
>>Hi -
>>
>>The Open Group will be working out a conformance program for LDAP.
>>
>>We currently operate several conformance programs under the name of the
>>Open Brand. This enables a buyer to require conformance to a profile of
>>standards. A supplier can not claim conformance unless he has passed a
>>set of tests AND has made a commitement to maintain his products in
>>conformance with the standards. There is a legal remedy (under
>trademark
>>law) against a vendor that makes a false claim.
>>
>>The Open Brand does not replace the contract between buyer and vendor
>>but it makes it much easier to draw up a contract and make it stick.
>>It has been used in procurements totalling nearly $25bn in value,
>>mostly for UNIX systems.
>>
>>The LDAP conformance program will be different from the UNIX one in
>>many respects. The essential requirement is for interoperability
>>rather than for conformance to an API. We will be working out the
>>details of this over the next few months, starting at the Open Group
>>meeting next week in Montreal.
>>
>>
>>At 05:20 PM 11/07/99 -0500, Mark Wahl wrote:
>>>
>>>My message tried to state that there was nothing the _IETF_ did to
>>>enforce vendor behavior.
>>>
>>>In the IETF there is no contractual arrangement between an implementor
>>>and a working group. However, in large deployments there may be a
>>>contractual arrangement between an implementor and a deployer of
>>>that technology. This contractual arrangement may reference Internet
>>>standards-track RFCs as criteria for a correct implementation, and the
>>>deployer may choose to use conformance or interoperability test
>results
>>>as the basis for evaluating this criteria. There are standard
>business
>>>practices for an implementor and deployer to work out what to do if
>>this
>>>criteria is not met.
>>>
>>>Mark Wahl, Directory Product Architect
>>>Innosoft International, Inc.
>>>
>>>
>>>
>>
>>Regards,
>>
>>Chris
>>+++++
>>
>>-----------------------------------------------------------------------
>-
>>-
>> Chris Harding
>> T H E Development Manager
>> O P E N Apex Plaza, Forbury Road, Reading RG1 1AX, UK
>>G R O U P Mailto:c.harding@opengroup.org Ph: +44 118 950 8311 x2262
>> WWW: http://www.opengroup.org Fx: +44 118 950 0110
>>
>>OSF/1, Motif, UNIX and the "X" device are registered trademarks in
>>the US and other countries, and IT DialTone and The Open Group are
>>trademarks of The Open Group.
>>-----------------------------------------------------------------------
>-
>>-
>>
>>
>>
>
>Regards,
>
>Chris
>+++++
>
>------------------------------------------------------------------------
>-
> Chris Harding
> T H E Development Manager
> O P E N Apex Plaza, Forbury Road, Reading RG1 1AX, UK
>G R O U P Mailto:c.harding@opengroup.org Ph: +44 118 950 8311 x2262
> WWW: http://www.opengroup.org Fx: +44 118 950 0110
>
>OSF/1, Motif, UNIX and the "X" device are registered trademarks in
>the US and other countries, and IT DialTone and The Open Group are
>trademarks of The Open Group.
>------------------------------------------------------------------------
>-
>
>
>
Regards,
Chris
+++++
-------------------------------------------------------------------------
Chris Harding
T H E Development Manager
O P E N Apex Plaza, Forbury Road, Reading RG1 1AX, UK
G R O U P Mailto:c.harding@opengroup.org Ph: +44 118 950 8311 x2262
WWW: http://www.opengroup.org Fx: +44 118 950 0110
OSF/1, Motif, UNIX and the "X" device are registered trademarks in
the US and other countries, and IT DialTone and The Open Group are
trademarks of The Open Group.
-------------------------------------------------------------------------