Sundar Yamunachari wrote:
> Hi,
> 
>    The functional specifications for AI transport (Web Server redesign) 
> is available at 
> http://wikis.sun.com/display/OSOLInstall/Web+Server+Redesign+Functional+Specification.
>  
> Please review the specs and provide your feedback by 06/15/09
> 

Most of this looks pretty good.  Some things to consider:

General: Please retitle this from redesign to version 2, or similar.

1.b: I've never encountered use of the security attributes noted here 
for torrents; is there a reference you can point to on cases where this 
has been done?

2.f.: Could use discussion of goals for multiple servers might interact, 
and how to scale the server infrastructure.  Will a transport be 
expected to work correctly with content caching infrastructures such as 
squid?  What would a high-availability server environment look like? 
Are we relying merely on clustering for HA?

5.c: Are all transports required to provide secure options?

5.f: There seem to be requirements here to enhance the TFTP service. 
Should discuss with networking team.

6.b: This is identical, or nearly so, to 6.a, which isn't surprising 
since they are really just two pieces of the same puzzle.  Recommend 
consolidating and up-leveling: rather than specific reference to 
solaris.zlib or solarismisc.zlib (both are undefined terms in this 
document), I would suggest referring generally to "archives required to 
assemble the complete installation environment" or something of that ilk.

6.d: I'm not sure how well this use case maps to a P2P transport; they 
typically operate on a pull model, yet this appears to be a push model.

6.e: Standard term is derived manifests.

6.f: Is P2P really a use case?  It seems like a solution to (at least 
some of) the other use cases, rather than a use case in its own right.

A use case for archive-based installation seems worth adding.

Finally, I'm wondering how this relates to a "serverless" automated 
installation, which is what the Virtual Machine Constructor seems to be 
requiring.  Is there a sort of "null" transport that would be used 
there?  Just trying to sort out how the architecture is anticipated to 
adapt to that case.

Dave

Reply via email to