Hi Joe

On Sep 22, 2012, at 7:35 PM, Joe Patterson <[email protected]> wrote:

> When you say "custom", how custom is it?  Is it a matter of taking the
> existing examples and making small tweaks, or a major coding project?

The first one.

> 
> As an alternative, if you're using DNA to access packet data, are
> those packets rendered unavailable through other non-DNA means?  So if
> you have one application using DNA with symmetric RSS, could you have
> another application that's just using standard pf_ring and seeing the
> same packets (albeit not as efficiently)?

Yes, if you are willing to sacrifice performance it is possible to use a 
"userspace ring" as "reflector device" (we introduced the 
pfring_set_reflector_device() call for this). 
This is a dirty solution with respect to the libzero-based equivalent one and 
also requires a small patch to the daq module.

Regards
Alfredo

> 
> Thanks,
> 
> -Joe
> 
> On Sat, Sep 22, 2012 at 1:13 PM, Alfredo Cardigliano
> <[email protected]> wrote:
>> Bob
>> please see inline
>> 
>> On Sep 21, 2012, at 8:43 PM, Bob Bomar <[email protected]> wrote:
>> 
>> I’m doing some testing currently with PF_RING + DNA.  And I had 2 questions
>> that I wanted to ask:
>> 
>> 1)      If we’re running the driver with 8 RSS queues, is it splitting the
>> traffic across those 8 queues, or only the ones that we’re listening to?
>> I.e, if we’ve got 8 queues and we’ve got snort running on 2 of them, are we
>> ignoring 75% of the traffic?
>> 
>> 
>> Yes, as RSS distributes packets to queues according to the hash,  you are
>> ignoring 75% of the traffic.
>> 
>> 2)      Can we, and if so how, get more than one application receiving the
>> same traffic?  So if we have snort listening on dna0@[0-7], can we also fire
>> up nprobe for dna0@[0-7] and have it see the same data?
>> 
>> 
>> This is available with standard pf_ring or with libzero for DNA (not with
>> standard DNA). In the second case you need a custom "pfdnacluster_master"
>> application for distributing/cloning traffic in zero-copy to nProbe and
>> snort, and a custom DAQ-DNA compatible with libzero.
>> 
>> Best Regards
>> Alfredo
>> 
>> 
>> 
>> --
>> Bob Bomar
>> Secure Information Services
>> terremark worldwide
>> Confidentiality Notice: This e-mail message, including any attachments, is
>> for the sole use of the intended recipient(s) and may contain confidential
>> and privileged information. Any unauthorized review, use, disclosure or
>> distribution is prohibited. If you are not the intended recipient and
>> received this in error, please contact the sender by reply e-mail and you
>> are hereby notified that the copying, use or distribution of any information
>> or materials transmitted in or with this message is strictly prohibited.
>> 
>> 
>> _______________________________________________
>> Ntop-misc mailing list
>> [email protected]
>> http://listgateway.unipi.it/mailman/listinfo/ntop-misc
>> 
>> 
>> 
>> _______________________________________________
>> Ntop-misc mailing list
>> [email protected]
>> http://listgateway.unipi.it/mailman/listinfo/ntop-misc
>> 
> _______________________________________________
> Ntop-misc mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop-misc

_______________________________________________
Ntop-misc mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop-misc

Reply via email to