[c-nsp] Dot1q Interface detect switchport failure

Rodney Dunn rodunn at cisco.com
Wed Sep 1 21:37:57 EDT 2004


Never stops amazing me the different deployment scenarios
that are out there.


On Wed, Sep 01, 2004 at 08:09:28PM -0400, Matthew Crocker wrote:
> Rodney,
> 
>   Thanks for passing on the info.   We use a Seranoa IPeX WANPort to 
> terminate our DS-1 circuits from customers.  The Seranoa maps the HDLC 
> frames to dot1q frames and sends them out a gigE port.  We use a 3550 
> with a bunch of interface vlans to terminate the IP packets from the 
> customer.  This setup works great with the only major problem being the 
> 3550 can't see the DS-1 and therefore doesn't know when the DS-1 is 
> down.   We also provide DSL backup to our DS-1 customers.  We use a 
> Cisco 1721 to  terminate the DS-1 & DSL (PPPoE) sessions at the 
> customer end.  Customers that have DS-1 & DSL are mapped to a 7513 so 
> we can use rtr/SSA to manage their routes.  It works like a charm.  rtr 
> saves us from running OSPF with all of our customers.
> 
> Again,  thanks for the info,  eventually I'll have enough customers to 
> melt the 3550 and will upgrade to a 7609 or 12k
> 
> -Matt
> 
> 
> On Sep 1, 2004, at 5:57 PM, Rodney Dunn wrote:
> 
> > Matt,
> >
> > Sorry but I don't know the answer to that.
> > I'm guessing it will be when the full rtr/SSA
> > support is in a 12.2S based branch of code that
> > the 3550 ships from.
> >
> > I'll pass the feedback along and let them know
> > someone asked about the support though.
> >
> > This is an interesting feature that was just targeted
> > towards dial backup and it's turned out to be used
> > in a lot more scenarios than I imagined.
> >
> > Rodney
> >
> > On Wed, Sep 01, 2004 at 11:14:23AM -0400, Matthew Crocker wrote:
> >>
> >> Rodney,
> >>
> >>   We do the Object tracking stuff with our 7513,  works great.  We 
> >> setup
> >> an RTR to ping the customer IP and a task tied to the rtr to drop the
> >> static route.  Our customers have DSL as backup which is announced via
> >> OSPF.     Do you know when the full rtr command set will be supported
> >> on the 3550 series?   Right now I can configure the rtr but I can't
> >> configure the task.   We have the EMI feature set (I think).
> >>
> >> Thanks
> >>
> >> -Matt
> >>
> >> On Sep 1, 2004, at 9:24 AM, Rodney Dunn wrote:
> >>
> >>> The problem he is trying to solve is to know when the
> >>> next hop is actually down.
> >>>
> >>> We developed something called object tracking and
> >>> started hooking other protocols in to it.
> >>> ie: PBR, static routes, etc.
> >>> One of the main goals was to solve the DSL dial backup
> >>> problem.  Since then it's been used in a lot of corner
> >>> case scenarios.
> >>>
> >>> Go to CCO and search on "object tracking static route".
> >>>
> >>> http://www.cisco.com/en/US/partner/products/sw/iosswrel/ps5413/
> >>> products_feature_guide09186a00801d862d.html
> >>>
> >>> It hasn't been ported to a code train that runs on the
> >>> 76xx.  The feature was developed originally for lowend
> >>> systems.
> >>>
> >>> There is no other way than a routing protocol to accomplish
> >>> what you need.
> >>>
> >>> You would write an offline script that does it for you.
> >>> Every 30 seconds telnet in and ping the next hop.
> >>> If the next hop doesn't respond add the backup route.
> >>> Repeat it and when the next hop comes back add the
> >>> route back.
> >>>
> >>> Rodney
> >>>
> >>>
> >>>
> >>>
> >>>
> >>> On Tue, Aug 31, 2004 at 09:56:40PM -0400, James Hampton wrote:
> >>>> Try using a floating static route, basically you have two static
> >>>> routes, one of them gets a slightly higher administrative distance
> >>>> than the other. When the first route is not reachable and dropped 
> >>>> from
> >>>> the table, the second one, with the lower A.D. takes over.
> >>>>
> >>>> James
> >>>>
> >>>> On Wed, 1 Sep 2004 01:54:47 +0800, Tay Chee Yong 
> >>>> <tcy at pacific.net.sg>
> >>>> wrote:
> >>>>> Hi all,
> >>>>>
> >>>>> I have a 7600 connecting to a Catalyst 3550, via a GE Trunk.
> >>>>>
> >>>>> As i know that the sub-interface on any router will always be up,
> >>>>> irregardless of the port status on the switch. I would like to find
> >>>>> out if the dot1q interface on a 7600 is capable of detecting an
> >>>>> outage on the switchport of the catalyst 3550, as i need fullfill
> >>>>> some kind of backup routing using static routing, so that when the
> >>>>> FE port on the switch is down, the router is capable of re-routing
> >>>>> traffic to a secondary link. BGP is not an option here. My scenario
> >>>>> doesn't allow me to connect the FE directly to the router also, due
> >>>>> to the network setup.
> >>>>>
> >>>>> Any advise is greatly appreciated.
> >>>>>
> >>>>> Regards,
> >>>>> Cheeyong
> >>>>> _______________________________________________
> >>>>> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> >>>>> https://puck.nether.net/mailman/listinfo/cisco-nsp
> >>>>> archive at http://puck.nether.net/pipermail/cisco-nsp/
> >>>>>
> >>>> _______________________________________________
> >>>> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> >>>> https://puck.nether.net/mailman/listinfo/cisco-nsp
> >>>> archive at http://puck.nether.net/pipermail/cisco-nsp/
> >>> _______________________________________________
> >>> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> >>> https://puck.nether.net/mailman/listinfo/cisco-nsp
> >>> archive at http://puck.nether.net/pipermail/cisco-nsp/
> >>>
> >


More information about the cisco-nsp mailing list