On 25/08/06, Arnaud BODENAN <[EMAIL PROTECTED]> wrote:
> When I start the master agent, 'ls -ltr master' gives:
> srwxrwxr-x   1 arnboden rms            0 Aug 25 11:09 ../master
>
> When I start the subagent, the same command gives the same result:
> srwxrwxr-x   1 arnboden rms            0 Aug 25 11:09 ../master

Yes.  That's as expected.
The master agent creates the socket (and listens on it, and deletes it
when it's finished).
The subagent simply connects to it.

The master agent *controls* the socket - the subagent *uses* it.


> But, I'm a bit confused with this parameter (agentxsocket). When must I use:
> - a named socket?
> - an address?

There's no "must" - you can use either, depending on your particular
requirements.
OK - that's not quite true.   If your subagents are running on a
different system than the master agent, then you need to use a TCP
socket (i.e. an address).  If everything's running on the same box,
then you can use either.


> For security reasons, is it preferable to use named socket or address?

A named socket.
That allows you to use filesystem permissions to restrict access to it.
A network socket is open to anybody - one of the biggest flaws in
AgentX security.

Dave

-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Net-snmp-users mailing list
[email protected]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users

Reply via email to