Re: cam timer tuning

Step (4) does not happen! Switches don't send ARP requests. Hosts do!

--
Marko Milivojevic - CCIE #18427 (SP R&S)
Senior CCIE Instructor - IPexpert
On Fri, Dec 28, 2012 at 12:48 PM, Joe Sanchez <marco207p_at_gmail.com> wrote:
> I did have this issue several years back with the 6500's but not a 4948, I
> couldn't dig up any bug's or old documentation on it though.
>
> Naufal,
>
> please let us know if this works for you.
>
> JS
>
> On Fri, Dec 28, 2012 at 12:39 PM, Marko Milivojevic <markom_at_ipexpert.com>
> wrote:
>>
>> Quite right. I don't think this is the result of STP. I believe this
>> is the result of "silent" hosts, or hosts that disappear from the
>> network, while someone is still sending them traffic (and knows about
>> their existence).
>>
>> --
>> Marko Milivojevic - CCIE #18427 (SP R&S)
>> Senior CCIE Instructor - IPexpert
>>
>> On Fri, Dec 28, 2012 at 10:24 AM, Joseph L. Brunner
>> <joe_at_affirmedsystems.com> wrote:
>> > No b  stp is a totally different feature. The root bridge in each vlan
>> > keeps
>> > the network in a converged state b  and that includes bpdub s that keep
>>
>> > flowing ever 2 seconds to keep blocking links blocking, etc.
>> >
>> >
>> >
>> > The goal of unifying the timers is make sure no unknown unicasts flood
>> > your
>> > network when the router has superior information it saved with its
>> > longer
>> > default timers than the cam table.
>> >
>> >
>> >
>> > Is this even an issue?
>> >
>> >
>> >
>> > This is usually 1% or less of unknown unicast activity.
>> >
>> >
>> >
>> > If you have lots of unknown unicasts from stale arp timers b  I would
>> > suspect
>> > you have other issues b  like a port scan or virus looking for machines
>> > that
>> > are not up friend.
>> >
>> >
>> >
>> > From: Naufal Jamal [mailto:naufalccie_at_yahoo.in]
>> > Sent: Friday, December 28, 2012 1:13 PM
>> > To: Joseph L. Brunner; Marko Milivojevic
>> >
>> >
>> > Cc: ccielab_at_groupstudy.com
>> > Subject: Re: cam timer tuning
>> >
>> >
>> >
>> > If we change the arp/cam timers to 600 secs across the entire L2 domain
>> > ie.
>> > including csw's and asw's is there any STP reconvergence that could
>> > happen?
>> > would we see any STP outage? I dont think there should be.
>> >
>> >
>> >
>> > ________________________________
>> >
>> > From: Joseph L. Brunner <joe_at_affirmedsystems.com>
>> > To: Naufal Jamal <naufalccie_at_yahoo.in>; Marko Milivojevic
>> > <markom_at_ipexpert.com>
>> > Cc: "ccielab_at_groupstudy.com" <ccielab_at_groupstudy.com>
>> > Sent: Friday, 28 December 2012 11:35 PM
>> > Subject: RE: cam timer tuning
>> >
>> >
>> > You should unify all timers at 600 seconds.
>> >
>> > That number is usually good to me.
>> >
>> > thanks
>> >
>> > -----Original Message-----
>> > From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of
>> > Naufal Jamal
>> > Sent: Friday, December 28, 2012 12:43 PM
>> > To: Marko Milivojevic
>> > Cc: ccielab_at_groupstudy.com
>> > Subject: Re: cam timer tuning
>> >
>> > Hi Marko,
>> >
>> > Nexus  ----------------trunk----------4948
>> > Arp timer:1500 secs
>> >                 Arp timer: 4 hours
>> > Cam timer:1800 secs                    Cam
>> > timer: 300 secs
>> >
>> > I know cam timer should be greater than arp timer to avoid flood. Should
>> > we
>> > try to make the cam timer in 4948 more than 1500 secs or make the arp
>> > timer
>> > in nexus less than 300 (aging arp in 300 secs ?? ) Any inputs?
>> > ________________________________
>> > From: Marko Milivojevic
>> > <markom_at_ipexpert.com>
>> > To: Naufal Jamal <naufalccie_at_yahoo.in>
>> > Cc:
>> > "ccielab_at_groupstudy.com" <ccielab_at_groupstudy.com>
>> > Sent: Friday, 28 December
>> > 2012 10:29 PM
>> > Subject: Re: cam timer tuning
>> >
>> > When I've seen this in the past,
>> > it was usually discrepancy between
>> > host ARP and switch MAC aging timers. Try lowering timers on your
>> > switches
>> > to be under/same as the ARP timers on hosts and see if there are any
>> > differences.
>> >
>> > --
>> > Marko Milivojevic - CCIE #18427 (SP
>> > R&S)
>> > Senior CCIE Instructor - IPexpert
>> >
>> > On Fri, Dec 28, 2012 at 1:40 AM,
>> > Naufal Jamal <naufalccie_at_yahoo.in> wrote:
>> >> Hello,
>> >>
>> >> We have a pair of nexus
>> > 7K's (without vpc) running HSRP and MSTP.
>> >> They are connected to access layer
>> > switches 4948's. whenever we bring up a new
>> >> trunk link between Nexus 7K and
>> > 4948 we see unicast flooding caused due to
>> >> TCN's. I am wondering if it has
>> > something to do with the difference in cam
>> >> aging timer in both the
>> > platforms. Nexus (1800 secs) and 4948 (300 secs). can
>> >> anyone put some light
>> > here please? I am thinking that changing the aging timer
>> >> on nexus to 300
>> > secs should help.. any thoughts?
>> >>
>> >> Naufal
>> >>
>> >>
>> >> Blogs and organic groups at
>> > http://www.ccie.net
>> >>
>> >>
>> > _______________________________________________________________________
>> >>
>> > Subscription information may be found at:
>> >>
>> > http://www.groupstudy.com/list/CCIELab.html
>> >
>> >
>> > Blogs and organic groups at http://www.ccie.net
>> >
>> > _______________________________________________________________________
>> > Subscription information may be found at:
>> > http://www.groupstudy.com/list/CCIELab.html
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net