>>>>> On Mon, 26 Jul 2004 23:02:28 +0200, Thomas Anders <[EMAIL PROTECTED]> said:

Thomas> I'm (re-)designing an application that processes SNMPv3
Thomas> INFORMS from tens of thousands of devices, each of them
Thomas> associated with a unique usmUser. The credentials get
Thomas> negotiated (by other means) just before sending the INFORM.

Ohh...  so many ways to implement this... sigh.

The simplest might be actually appending the new user info to a new
file, and registering a new signal and then sending that signal and
having snmptrapd call a read_config routine to read *just that file*
(IE, much less than a full -HUP signal).

Thomas> A snmpUsmMib implementation (like in snmpd) looks overkill for
Thomas> snmptrapd. A designated admin interface or different backends
Thomas> don't yet exist. Shall I (really?) use SNMPv3 INFORMs
Thomas> themselves to do the job?

Well, actually, you're doing user management right?  And SNMP provides
a way to do that right?  You seem to be trying to avoid something that
already is known to work and the code is already written for.  It'd be
trivial to make snmptrapd start the usm code and act as a full agent
(on yet another port) so you could do user management through it.

-- 
Wes Hardaker
Sparta


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
_______________________________________________
Net-snmp-coders mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders

Reply via email to