Hi Andy, Cat6500 is a distributed environment. Not only PFC (or any other one part for this matter) is responsible for capabilities.
Alex. On Jan 22, 2013 2:27 PM, "Andy Ellsworth" <[email protected]> wrote: > If the PFC doesn't support it, it's done in software (or not at all). This > is Cat6500 fundamentals. > > Q.E.D. > > > On Tue, Jan 22, 2013 at 5:32 AM, Alex K. <[email protected]> wrote: > > > Yes I know. I'm looking for a best match. > > > > I've already sent an email to my local SE. > > > > The point is that I need something official that will state 'yes, it's > done > > by software, cpu impact is expected'. > > > > Best Regards, > > Alex. > > On Jan 22, 2013 11:58 AM, "Oliver Boehmer (oboehmer)" < > [email protected]> > > wrote: > > > > > Alex, > > > > > > not sure what you're looking for. "Not supported" means you're on your > > > own, use it at your own risk and expect things can go wrong. It could > be > > > switched in software in one release (which might be fine and serve your > > > purpose as long as the traffic stays below given threshold or it > doesn't > > > affect other features you are using), or hell could freeze over in > other > > > releases, we don't test this. So I guess you could call your setup > > > "mis-configured". > > > > > > you will not find a document stating "NBAR implementation is software > > > based on the PFC/7600". > > > > > > oli > > > > > > > > > > > > On 22/01/2013 10:47, "Alex K." <[email protected]> wrote: > > > > > > >Hi Oliver, > > > >Exactly - not supported. It implies that *if it works (not on > SIP-200), > > > >it > > > >must be software'. > > > >I came across this document before I sent the question. As it seems, > > that > > > >what I'll use. > > > >I'm looking for a document that say explicitly 'NBAR implementation is > > > >software based' to be sure we didn't run into some sort of > > > >bug/mis-configuration. > > > >Thank you. > > > > > > > >Best Regards, > > > >Alex. > > > >On Jan 22, 2013 8:04 AM, "Oliver Boehmer (oboehmer)" < > > [email protected]> > > > >wrote: > > > > > > > >Alex, > > > > > > > > > > > >On 22/01/2013 01:19, "Alex K." <[email protected]> wrote: > > > > > > > >>Hi Pete, > > > >> > > > >>We're running 12.2(33)SRA6. > > > >> > > > >> > > > >>On SIP-200 it's running fine (as expected). Configuring > > > >>NBAR-using-policy-map on an *SVI*, causes high CPU Interrupts. > > > >> > > > >> I do believe it's being punted to a CPU. > > > >> > > > >>But this time I need a document that clearly states that i.e. on > > > >>SIP-200 > > > >>by hardware, on SVI by software and this is not a bug/some other > > > >>malfunctioning. > > > >> > > > >>I'm asking for a document from which we can understand that, yes, > using > > > >>NBAR on an SVI will make those packets punted. Technically I agree > with > > > >>you > > > >>completely, most likely that¹s what happening. > > > > > > > > > > > > > > > > > > http://www.cisco.com/en/US/docs/routers/7600/ios/15S/configuration/guide/q > > > >o > > > >s.html says "The PFC does not support Network-Based Application > > > >Recognition (NBAR).", this is valid for earlier SW releases as well. > So > > > >your config on the SVI is not supported. > > > > > > > >SIP200 Datasheets clearly state NBAR support. > > > > > > > > oli > > > > > > > > > > > > _______________________________________________ > > cisco-nsp mailing list [email protected] > > https://puck.nether.net/mailman/listinfo/cisco-nsp > > archive at http://puck.nether.net/pipermail/cisco-nsp/ > > > _______________________________________________ > cisco-nsp mailing list [email protected] > https://puck.nether.net/mailman/listinfo/cisco-nsp > archive at http://puck.nether.net/pipermail/cisco-nsp/ > _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
