[Planetlab-devel] geniwrapper integration
rmoran at CS.Princeton.EDU
rmoran at CS.Princeton.EDU
Tue May 20 15:55:46 EDT 2008
Tony and I have been going over the code, but it is not all that clear
to us either. I'm hoping Soner can interject at some point and give us
more of a developers idea of what is going on. His code is not similar
to our API or NM code at all and uses a Java GUI that has a lot of code
within it as well.
We are continuing to dig through it and determine how to integrate it
best, but it is not the patch job we had assumed it would be. It is
going to be non trivial to translate his code to work on our system, as
well as fit it into our command line setup and GUI.
The CRED_FILE is indeed the credential file we all know and love that
will be returned by a GetCredential call, but the ACC_FILE I am not
sure of. It may be part of the overall credential, but we will need to
dig deeper to unravel this further.
Lets keep this discussion open here so we can help each other.
Quoting Scott Baker <smbaker at gmail.com>:
> As a first step in the NM/AC wrapper, I'm working on getting Soner's
> geniwrapper working so that I have a reference implementation of the
> security design. Tony and Reid, I know that you're also working on
> geniwrapper integration (I'm assuming into the PLC_API trunk?). Most
> of the issues I've run into have been related to integration (getting
> the right pieces of geniwrapper connected to the right pieces of
> PLC_API for example), so I'd like to coordinate our efforts if we can.
>
> Also, the API document is handy, but do we have a discussion anywhere
> about ACC_FILE and CRED_FILE? I'm not sure which objects in the
> overall geni design these items map into. My assumption is that the
> ACC_FILE is a list of certs for a chain of authorities, and a
> CRED_FILE is akin to a SliceCredential, but I'm not sure if I'm
> correct.
>
> Scott
>
> _______________________________________________
> Devel mailing list
> Devel at lists.planet-lab.org
> https://lists.planet-lab.org/mailman/listinfo/devel
>
More information about the Devel
mailing list