[OTDev] A&A

Tobias Girschick tobias.girschick at in.tum.de
Mon Oct 25 11:12:37 CEST 2010


Hi,

I copied below what I found in an old OT-dev mailinlist thread. It's
part of a discussion of Luchesar and Andreas Maunz.

>         
>         > and a short comment on the algorithm example
>         >
>         > curl -X GET -d 'username=usr' -d 'password=pwd'
>         
>  
>         > http://localhost:8080/OpenTox-dev/algorithm/test -iv
>         >
>         > should use the token from the opensso authentication
>         >
>         > curl -X GET -d 'token_id=token'
>         > http://localhost:8080/OpenTox-dev/algorithm/test -iv
>         >
>         > you authenticate once and then use the token afterwards.
>         Each
>         > webservices proofs authorization with the token.
>         
>         
>         The problem is, that in this case, the user would have to
>         authenticate
>         on his own before using a 'restricted' web service. Or is it
>         in this
>         case a 'security breach', as the service get's the uid and
>         pwd? It would
>         be convenient to just use my pwd and my username if I want to
>         issue just
>         one service call...
>         Has this been discussed and I am not finding and/or
>         remembering the
>         results or is this still an open issue? 
> 
> It has been discussed number of times,  look for discussions of
> service being able to authenticate on behalf of users .

-- Quoted:

> One final thing that seems to be not yet decided: so far we were
> discussing whether the services should always expect user credentials
> or they would allow totally unauthenticated requests as well. How
> about the user credentials themselves: as you remember, we can decide
> between:
> z) accepting only user/password -- then the service does the
authentication.
> x) accepting only tokens -- then the user or requesting service does
it.
> y) accepting both -- or rather, "any", as "both", literally, wouldn't
> make sense.
> Actually, it's a choice between (x) and (y), because each service (or
> most of them) must accept tokens, anyway -- because of the requests
> from other services -- and this means (z) doesn't suit us at all.

Since user/password are confidential information, we should think about 
passing only tokens whenever possible. In that sense, x) would be best.
Also, a workflow would have no chance of getting started, if the client 
software can not obtain a token (in case SSO service is down). I
believe 
this is a benefit.

-------

I haven't found any decision statements ...


> 
> I am not sure if there was a final decision , but would be good to
> have support of both options.

I will add both now on the web, but we should maybe discuss and decide
in next weeks SWDT.


best regards,
Tobias

> 
> Best regards,
> Nina
> 
> 
>  
> 
>         
>         
>         best regards,
>         Tobias 
>         
>         
>         >
>         > best regards,
>         >
>         > Micha
>         >
>         >
>         > Tobias Girschick schrieb:
>         > > Hi,
>         > >
>         > > ok. In that case, I propose to use the same scheme as in
>         the A&A
>         > > Deliverable Example session for communicating with the
>         openSSO service
>         > > to prevent confusion (also the parameter names can be
>         reused). There
>         > > form parameters are used:
>         > > curl -i -d 'username=<uid>' -d 'password=<sec>'
>         > > http://opensso.in-silico.....
>         > >
>         > > For an algorithm that would mean:
>         > > curl -X GET -d 'username=usr' -d 'password=pwd'
>         > > http://localhost:8080/OpenTox-dev/algorithm/test -iv
>         > >
>         > > The question is where to add it in the wiki (API)? Only in
>         the A&A
>         > > section or in every component?
>         > >
>         > > Any opinions on that? NTUA, how are you handling that?
>         ALU? All?
>         > > best regards,
>         > > Tobias
>         > >
>         > > On Thu, 2010-10-21 at 16:16 +0300, Nina Jeliazkova wrote:
>         > >
>         > >> Hi Tobias,
>         > >>
>         > >>
>         > >> On 21 October 2010 16:07, Tobias Girschick
>         > >> <tobias.girschick at in.tum.de> wrote:
>         > >>         Hi Nina, Andreas, All,
>         > >>
>         > >>         while adding the A&A functionality to the TUM
>         webservices I
>         > >>         stumbled
>         > >>         upon one question: is there a uniform way how the
>         user
>         > >>         credentials
>         > >>         (username, password) are passed to the services
>         (not the A&A
>         > >>         services
>         > >>         but an algorithm for example)?
>         > >>
>         > >>         Meaning do I issue:
>         > >>         curl -X GET
>         > >>
>         http://localhost:8080/OpenTox-dev/tox/test?username=usr&password=pwd -iv
>         > >>         or
>         > >>         curl -X GET -d 'username=usr' -d 'password=pwd'
>         > >>         http://localhost:8080/OpenTox-dev/tox/test -iv
>         > >>
>         > >>         I didn't find anything in the A&A Deliverable or
>         the API...
>         > >>
>         > >> Your observation is absolutely correct, AFAIK this is not
>         (yet!)
>         > >> defined in the API .
>         > >>
>         > >>
>         > >>         I think I have seen the first version in the
>         AMBIT code (might
>         > >>         be
>         > >>         deprecated version)
>         > >>
>         > >>         Object username =
>         > >>
>         request.getResourceRef().getQueryAsForm().getFirstValue(username_tag);
>         > >>
>         > >>
>         > >> It's not deprecated, but was done quickly to test the the
>         AA and is
>         > >> used only for a test resource.
>         > >>
>         > >> I think it would be best if you just pick up parameter
>         names which you
>         > >> prefer (hopefully the same as ALU and NTUA are using) and
>         document it
>         > >> in the wiki.  I don't have preferences and it will be
>         easy to change
>         > >> ambit code to reflect your choice.
>         > >>
>         > >> Best regards,
>         > >> Nina
>         > >>
>         > >>
>         > >>         best regards,
>         > >>         Tobias
>         > >>         --
>         > >>         Dipl.-Bioinf. Tobias Girschick
>         > >>
>         > >>         Technische Universität München
>         > >>         Institut für Informatik
>         > >>         Lehrstuhl I12 - Bioinformatik
>         > >>         Bolzmannstr. 3
>         > >>         85748 Garching b. München, Germany
>         > >>
>         > >>         Room: MI 01.09.042
>         > >>         Phone: +49 (89) 289-18002
>         > >>         Email: tobias.girschick at in.tum.de
>         > >>         Web: http://wwwkramer.in.tum.de/girschick
>         > >>
>         > >>
>         > >>
>         > >
>         > >
>         > _______________________________________________
>         > Development mailing list
>         > Development at opentox.org
>         > http://www.opentox.org/mailman/listinfo/development
>         
>         
>         -- 
>         
>         Dipl.-Bioinf. Tobias Girschick
>         
>         Technische Universität München
>         Institut für Informatik
>         Lehrstuhl I12 - Bioinformatik
>         Bolzmannstr. 3
>         85748 Garching b. München, Germany
>         
>         Room: MI 01.09.042
>         Phone: +49 (89) 289-18002
>         Email: tobias.girschick at in.tum.de
>         Web: http://wwwkramer.in.tum.de/girschick
>         
>         _______________________________________________
>         Development mailing list
>         Development at opentox.org
>         http://www.opentox.org/mailman/listinfo/development
>         
> 

-- 
Dipl.-Bioinf. Tobias Girschick

Technische Universität München
Institut für Informatik
Lehrstuhl I12 - Bioinformatik
Bolzmannstr. 3
85748 Garching b. München, Germany

Room: MI 01.09.042
Phone: +49 (89) 289-18002
Email: tobias.girschick at in.tum.de
Web: http://wwwkramer.in.tum.de/girschick




More information about the Development mailing list