Applying an ACL to block the next packet may be faster than shutting the
interface. I've never tested the reaction speed of EEM in anything less
than a few seconds. Maybe a policer plus the EEM would be enough time to
get the policy through?
-nick
On Fri, Nov 5, 2010 at 2:09 PM, Rich Collins <nilsi2002_at_gmail.com> wrote:
> Hi,
>
> Yes it is an isolated environment and I will not affect other testing.
>
> I tried your suggestion with an ACL with the log statement. The EEM
> looks for the pattern in the syslog and then shuts down the interface.
> The problem is that this is not fast enough - about 3-4 seconds
> elapses before the interface is shut down "enable, config term,
> interface, shut". I'm not sure if some other EEM action would be much
> faster.
>
> This also opens up an old question about logging on ACL's. Only the
> first packet is logged and not the following ones if they have the
> same characteristics - at least for the next five minutes. How can
> one disable this default or how can one reinitialize this buffer?
> Clear access-list counter does not do the job.
>
> Thnks
> Rich
>
>
>
> On Thu, Nov 4, 2010 at 8:44 PM, Nick Matthews <matthn_at_gmail.com> wrote:
> > If it was an isolated environment, and maybe even not then, you could do
> > something like this:
> >
> > Create a trigger:
> > Turn on 'debug ip packet detail'
> > or
> > Use an access list with a 'log' statement
> >
> > Write an EEM script to trigger when something in the log matches either
> the
> > packet details or the log statement
> > Have the EEM script write an ACL to block the rest of the packets
> >
> > At that point I would probably manually disable to ACL to re-test. You
> > could get fancy and write a watchdog EEM to do this as well.
> >
> > -nick
> >
> > On Thu, Nov 4, 2010 at 3:27 PM, Jay McMickle <jay.mcmickle_at_yahoo.com>
> wrote:
> >>
> >> What about VACL's or MACL's? You could block this at the layer 2 frame.
> >>
> >>
> >> Regards,
> >> Jay McMickle- CCNP, CCSP, CCDP, MCSE
> >> http://mycciepursuit.wordpress.com/
> >>
> >>
> >>
> >>
> >>
> >> ________________________________
> >> From: Rich Collins <nilsi2002_at_gmail.com>
> >> To: Cisco certification
> >> <ccielab_at_groupstudy.com>
> >> Sent: Thu, November 4, 2010 9:48:37 AM
> >> Subject:
> >> Router trick - how to allow only one single packet
> >>
> >> Hi all,
> >>
> >> I am trying to
> >> test a client application in the lab and need a method
> >> to block subsequent
> >> requests to a server. The retries (UDP packets
> >> with same length, port number)
> >> etc. from this client should not reach
> >> the server. The retries occur less
> >> than a second later and continue.
> >>
> >> Limiting by CAR would still pass some of
> >> the requests a few seconds
> >> later. I can't record and spoof this first packet
> >> because of the
> >> encoding in the packet.
> >>
> >> I was also thinking of load balancing
> >> by packet and creating numerous
> >> sinkholes at dummy destinations.
> >>
> >> Any ideas or
> >> EEM?
> >>
> >> Thanks
> >> Rich
Blogs and organic groups at http://www.ccie.net
Received on Sat Nov 06 2010 - 12:04:33 ART
This archive was generated by hypermail 2.2.0 : Sun Dec 05 2010 - 22:14:55 ART