Hi all,
My this is a good going debate. Firstly, I don’t have the actual lab set up
at the mo, so doing this from memory, from the PDF diagrams and from scanning
the notepad initial config files.
I’m afraid I disagree.
This is EXACTLY what the network backdoor command in BGP is
for!! It is used when you have a route
learned via IGP and also via EBGP, and
you want to use the IGP route in preference to the EBGP route. That is why it
artificially bumps up the AD
of the BGP route to 200. So what if both
are using the same physical link in this case (as opposed to a separate link as
diagrammed in Doyle)!!
Think…
Before you start the lab, someone has configured BGP to
advertise 1.1.1.1 from R1. This
advertisement will be seen to originate from AS100 with an origin code of IGP.
If you remove the advertisement from R1, even if it is
redistributed from somewhere else, the network is not being originated from the
correct ASN as far as BGP is concerned.
To my thinking, this is “removing an existing feature” and
will get a big fat zero points in the real lab.
There is no instability if the OSPF route is lost. For this to happen, you
lose R1 or you lose
the link between R1 and R2. Depending on
the failure mode, OSPF will eventually reach dead time and remove the route to
1.1.1.1.
The BGP route to 1.1.1.1 will be asserted, but only until
the BGP session times out and then it too will be lost. This is not
instability. This is not flapping behaviour, merely BGP
failing differently to OSPF.
If OSPF comes back before BGP session times out, then great,
the OSPF route is returned to the routing table. If the OSPF route comes back
after BGP times
out, the BGP session will re-establish.
This is NOT instability, inherent within the network, but catering for
failure scenarios you are not asked to troubleshoot.
To my mind, network backdoor is a perfectly valid solution
and definitely better than removal of the network command on R1.
There are many other solution. Off the top of my head:
Change ospf distance to 19 for route 1.1.1.1.
Put distribute list under BGP to stop BGP route to 1.1.1.1
in routing table, but keeping it in BGP table.
Change BGP distance on EBGP routes to 111.
> Date: Fri, 14 Sep 2012 17:54:51 -0700
> From: [email protected]
> To: [email protected]
> CC: [email protected]
> Subject: Re: [OSL | CCIE_RS] Lab 3-1 Troubleshooting Ticket 5
>
> Thank you to all for the support and responses. I took a different approach
> and asked myself "is this the best use of bgp backdoor?" and the answer is no.
>
> BGP backdoor is used when you want to increase the AD of a BGP learned route
> in favor of an OSPF learned route, for example. It is not good to use in
> this case as it would once again cause instability when the ospf learned
> route is lost, as Joe mentioned.
>
> I believe it would be best to remove the network statement in BGP in this
> particular case, as it is still learned through ospf and redistributed back
> into BGP at another point in the network.
>
> Thanks again for the feedback.
>
> Nick
>
>
> ________________________________
> From: Joe Sanchez <[email protected]>
> To: Nick Bonifacio <[email protected]>
> Cc: Marko Milivojevic <[email protected]>; "[email protected]"
> <[email protected]>
> Sent: Friday, September 14, 2012 7:56 PM
> Subject: Re: [OSL | CCIE_RS] Lab 3-1 Troubleshooting Ticket 5
>
> Nick, believe me... I understand, I have the same Best Friend and getting
> through the lab exam in time is very tough for me because I psychoanalys
> everything.
>
> Regards,
> Joe Sanchez
>
> ( please excuse the brevity of this email as it was sent via a mobile device.
> Please excuse misspelled words or sentence structure.)
>
> On Sep 14, 2012, at 12:48 PM, Nick Bonifacio <[email protected]> wrote:
>
> > But best practice is out the window in the lab, right?
> >
> > Are we both correct and satisfy the task?
> >
> > Sorry, but I have this friend and his initials are O.C.D.
> >
> >
> > Sent from my iPhone
> >
> > On Sep 14, 2012, at 1:40 PM, Marko Milivojevic <[email protected]> wrote:
> >
> >> I like what I read. I like it very much :-)
> >>
> >> --
> >> Marko Milivojevic - CCIE #18427 (SP R&S)
> >> Senior CCIE Instructor - IPexpert
> >>
> >> On Fri, Sep 14, 2012 at 8:29 AM, Joe Sanchez <[email protected]> wrote:
> >>> Nick, when I did this lab I thought the same thing, but after looking at
> >>> the idea of using some other method of learning the lo0 via BGP and an
> >>> IGP I thought about how would the backdoor command come into play; which
> >>> it would never help because if you loose the IGP route and the 1.1.1.1
> >>> was populated into the BGP table, we would drop the connection anyways,
> >>> therefore removing the network statement from R1 was the only logical
> >>> answer I could come up with. In this particular case there is a single
> >>> connection with two peers from R1s perspective.
> >>>
> >>> Regards,
> >>> Joe Sanchez
> >>>
> >>> ( please excuse the brevity of this email as it was sent via a mobile
> >>> device. Please excuse misspelled words or sentence structure.)
> >>>
> >>> On Sep 14, 2012, at 8:03 AM, Nick Bonifacio <[email protected]> wrote:
> >>>
> >>>> Spoiler alert: Do not read ahead if you have not done this lab/question
> >>>> yet!
> >>>>
> >>>>
> >>>> Ticket 5 is as follows:
> >>>> "The BGP peering between R1 and R2/R4 is flapping and needs to be
> >>>> stabilized. Please correct the problem."
> >>>>
> >>>>
> >>>> From my first look at R2 and R4, it appears that they are learning about
> >>>> 1.1.1.1 (update source) through BGP which would of course cause
> >>>> problems. The solution is to remove network 1.1.1.1 mask
> >>>> 255.255.255.255 from router 1, but to me this seems like it would go
> >>>> against "keep the spirit of the config".
> >>>>
> >>>> What I did on R2 and R4:
> >>>>
> >>>>
> >>>> router bgp 2456
> >>>> network 1.1.1.1 mask 255.255.255.255 backdoor
> >>>>
> >>>> This makes the preferred route to 1.1.1.1 over OSPF instead of BGP by
> >>>> making the distance of the 1.1.1.1 route 200 and keeps the config intact.
> >>>>
> >>>> Is this a better/worse solution? Do I not understand the function of
> >>>> the BGP backdoor?
> >>>>
> >>>> thanks,
> >>>> Nick
> >>>> _______________________________________________
> >>>> For more information regarding industry leading CCIE Lab training,
> >>>> please visit http://www.ipexpert.com/
> >>>>
> >>>> Are you a CCNP or CCIE and looking for a job? Check out
> >>>> http://www.platinumplacement.com/
> >>>>
> >>>> http://onlinestudylist.com/mailman/listinfo/ccie_rs
> >>> _______________________________________________
> >>> For more information regarding industry leading CCIE Lab training, please
> >>> visit www.ipexpert.com
> >>>
> >>> Are you a CCNP or CCIE and looking for a job? Check out
> >>> www.PlatinumPlacement.com
> >>>
> >>> http://onlinestudylist.com/mailman/listinfo/ccie_rs
> _______________________________________________
> For more information regarding industry leading CCIE Lab training, please
> visit www.ipexpert.com
>
> Are you a CCNP or CCIE and looking for a job? Check out
> www.PlatinumPlacement.com
>
> http://onlinestudylist.com/mailman/listinfo/ccie_rs
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit
www.ipexpert.com
Are you a CCNP or CCIE and looking for a job? Check out
www.PlatinumPlacement.com
http://onlinestudylist.com/mailman/listinfo/ccie_rs
