<tt><font size=2>> From: Steven Dake <sdake@redhat.com></font></tt>
<br><tt><font size=2>...<br>
> The crux of the problem is how do you obtain critical mass for <br>
> custom one-off solutions? Lets assume two possible solutions
to <br>
> this problem that these vendors could take. If there are more,
<br>
> please feel free to explain them:<br>
> <br>
> 1) Implement a ReST server which the vendor's image talks to ReST
<br>
> server to obtain bootstrapping information<br>
> 2) SSH into the machine from an external configuration server process<br>
</font></tt>
<br><tt><font size=2>It looks to me like the OneConvergence guys had in
mind</font></tt>
<br>
<br><tt><font size=2>(3) Client is authorized by presenting a username/password
pair, as mentioned in the Puppet case (</font></tt><a href="https://puppetlabs.com/blog/managing-f5-big-ip-network-devices-with-puppet"><tt><font size=2>https://puppetlabs.com/blog/managing-f5-big-ip-network-devices-with-puppet</font></tt></a><tt><font size=2>)
cited elsewhere in this thread; server authentication/authorization is
not a concern; conversation confidentiality either is not a concern or
is handled by the client/server protocol without additional configuration;
client is code running in the heat engine</font></tt>