[Click] SNMPTrapSource

[Eddie Kohler]

Well, that is what IP routing and ARPQuerier are for!

If you know the trap dest is elsewhere, I would do something like

trapSource -> SetIPAddress(ADDR_OF_GATEWAY) -> ARPQuerier(WHATEVER) -> ToDevice;

If you don't know whether trap dest is here or elsewhere, something like

trapSource -> StaticIPLookup(LOCALNET 0, 0/0 GATEWAY 0) -> ARPQuerier(WHATEVER) 
-> ToDevice;

(where LOCALNET is something like 131.24.0.0/16, and GATEWAY is an IP addr like 
131.24.0.1)

If you're working at user level, a bit more may be required.

Eddie


Paine, Thomas Asa wrote:
> Eddie,
>         Thank you very much.  I had the trap part working, I guess I was curious what people's configurations might look like after the trap source.  For my prototype, I just ran the UDP packets, from the SNMPTrapSource, through an EtherEncap to at least get a sensible frame on the wire... (where eth5 below is really the management interface for the box, for SSH etc, as well)
> 
> <snip>
> ee::EtherEncap(0x0800, 1:1:1:1:1:1, ff:ff:ff:ff:ff:ff);
> trapSource -> ee -> trapQueue::Queue -> trapIf::ToDevice(eth5);
> </snip>
> 
>         However, the managment DST could easilly be off the local network, and I probably should not be broadcasting the traps even if it  wasn't.  :)  Just curious what people were doing to address trap destinations off the interface's subnet, getting the interface's hardware address, a gateway's MAC address, etc, at runtime.  I do see your usage of "Script", which may be the solution I need.  Or, using proc to modify the EtherEncap after it's up, I don't know.
> 
>         Feedback/discussion is more than welcome.
> 
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 
>    Thomas Paine {paineta at uwec.edu)}
>    University of Wisconsin - Eau Claire
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> 
> -----Original Message-----
> From: Eddie Kohler [mailto:kohler at cs.ucla.edu]
> Sent: Wednesday, October 03, 2007 2:00 PM
> To: Paine, Thomas Asa
> Cc: 'click at amsterdam.lcs.mit.edu'
> Subject: Re: [Click] SNMPTrapSource
> 
> Thomas,
> 
> I've just checked in a sample script, uncommented unfortunately, into the snmp
> packages directory, in conf/test-trap.click.  This script requires the very
> latest click as well (for the "goto begin CONDITION" construct).
> 
> http://www.read.cs.ucla.edu/gitweb?p=click-packages;a=blob;f=snmp/conf/test-trap.click;hb=HEAD
> 
> Eddie
> 
> 
> Paine, Thomas Asa wrote:
>>         Does anyone have a working example configuration, which uses SNMPTrapSource, they would be willing to share?
>>
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>    Thomas Paine {paineta at uwec.edu)}
>>    University of Wisconsin - Eau Claire
>> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>
>>
>>
>> _______________________________________________
>> click mailing list
>> click at amsterdam.lcs.mit.edu
>> https://amsterdam.lcs.mit.edu/mailman/listinfo/click