On Thu, 2005-10-20 at 11:22 +0200, Thomas Anders wrote:
> FWIW, the original *intention* of --disable-agent wrt. snmptrapd
> seems to have been:
>
> [x] build snmptrapd without agent
> But I don't think it ever worked.
> So should we try to complete what has been started back then?
Well, first let's be clear about what exactly *are* the
dependencies that snmptrapd has on the agent code.
AIUI, there are probably three main elements:
a) running as an AgentX subagent
b) access control processing
c) invoking external commands ('run_shell_command')
a) feels fairly straightforward - if the agent
isn't being built, it's probably safest to
disable the snmptrapd AgentX support as well.
I'm not sure whether b) is currently a dependency
or not. But it presumably will become so, when
Wes' latest proposals come on board.
c) is fairly fundamental. It's used to implement
the 'traphandle' directive, which feels a fairly
core part of snmptrapd behaviour!
I suspect that the cleanest way to handle c) might
be to compile 'agent/mibgroup/utilities/execute'
directly into the snmptrapd binary (rather than
accessing it via the agent library).
Thoughts?
Dave
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
