[AusNOG] VPLS OSPF question

Skeeve Stevens skeeve+ausnog at eintellegonetworks.com
Wed Apr 17 21:37:52 EST 2013


I never recommend EIGRP on the basis I recommend everyone to always avoid
any vendor specific protocols if it can helped.

When there are protocols in abundance, why do it to yourself?




...Skeeve

*Skeeve Stevens - *eintellego Networks Pty Ltd
skeeve at eintellegonetworks.com ; www.eintellegonetworks.com

Phone: 1300 239 038; Cell +61 (0)414 753 383 ; skype://skeeve

facebook.com/eintellegonetworks ;  <http://twitter.com/networkceoau>
linkedin.com/in/skeeve

twitter.com/networkceoau ; blog: www.network-ceo.net


The Experts Who The Experts Call
Juniper - Cisco - Cloud


On Wed, Apr 17, 2013 at 9:32 PM, Tom Berryman <tom at connectivityit.com.au>wrote:

>  Any reason EIGRP doesn’t get a look in here? (I will presume naively
> that it’s a full Cisco operation- or some brave other vendor implements it).
> ****
>
> I’d suggest it has an advantage in the simplicity and support stakes.****
>
> Tom.****
>
> ** **
>
> *From:* ausnog-bounces at lists.ausnog.net [mailto:
> ausnog-bounces at lists.ausnog.net] *On Behalf Of *J Williams
> *Sent:* Wednesday, 17 April 2013 5:39 PM
> *To:* Matt Carter
> *Cc:* ausnog at lists.ausnog.net
> *Subject:* Re: [AusNOG] VPLS OSPF question****
>
> ** **
>
> RSVP could create significant administrative overhead. Are you using auto
> tunnel or automesh, or some scripting?****
>
> BBA with PPPoE, nice for hub-and-spoke topology and very ISP-ish. ^_^****
>
> ** **
>
> Discussed this briefly with my colleague, and come up with following:****
>
> ** **
>
> BGP with dual RR. Can use BFD where needed for need fast convergence.****
>
> If this is an all Cisco shop, then DMVPN.****
>
> IS-IS is nice as doesn't rely on IP, but not supported on many low end
> routers.****
>
> OSPF is ok, but consider the CPU requirements, especially if using IPv6 as
> well.****
>
> ** **
>
> ** **
>
> Cheers,****
>
> Jules****
>
> ** **
>
> On Wed, Apr 17, 2013 at 10:45 AM, Matt Carter <mattc at mansol.net.au> wrote:
> ****
>
>  What about MPLS-TE  ?****
>
>
> …Consider BGP has no understanding of interface bandwidth, in reality it
> barely knows anything about where traffic should go and hides non-best-path
> information, passing only along the “best route” where advertising a second
> path forces a withdrawl of the first. This is a inherent problem if you
> want flexibility & scale, using an underlying IGP (ISIS/OSPF) and something
> like RSVP allows a full end to end path to be built with consideration for
> the bandwidth & available paths end to end, every router has a copy of the
> TE traffic engineering database. (no loops possible.) …… BGP MEDs on scale
> don’t really help much for the pain they impose (anyone?) and you tend to
> find a closest-exit regime anyway. ****
>
>  ****
>
> In short, BGPs ability to manipulate traffic in an MPLS environment is
> limited, it may be great for exchanging edge information using
> multiprotocol extensions offered by BGP, but for the core transport you
> really want to leverage off the TE capabilities provided by ISIS or OSPF.
> ****
>
>  ****
>
> At the very least this lets you adjust your network and service offering
> based on things like throughput and delay, bandwidth and paths available in
> the underlying core network, which you simply can’t get by doing a simple
> “virtual-mesh” topology using l2 switching vlans / QinQs etc creating
> shortcuts everywhere to make routers appear to be 1 hop from each other.
> There’s no/little feedback or dynamic capability from the underlying
> network architecture in that model without relying on STP or multiple
> shortcut vlans, messy messy messy.****
>
>  ****
>
> My 2c.****
>
>  ****
>
>  ****
>
> *From:* ausnog-bounces at lists.ausnog.net [mailto:
> ausnog-bounces at lists.ausnog.net] *On Behalf Of *Johann Lo
> *Sent:* Wednesday, 17 April 2013 10:09 AM
> *To:* Tom Storey****
>
>
> *Cc:* ausnog at lists.ausnog.net
> *Subject:* Re: [AusNOG] VPLS OSPF question****
>
>  ****
>
> True. I guess I misunderstood your question.****
>
>  ****
>
> But I’d just K.I.S.S. and split the VPLS into VLANs (one for each state
> etc.) and then run each as an OSPF area. I’d assume with a VPLS cloud that
> size you’d have Q-in-Q. ****
>
>  ****
>
> The real impetus with deploying BGP would be if you want to run MPLS and
> VRFs internally but even then you could get away with VLANs and VRF-lite
> and leak routes via real devices e.g. inter VRF aggregation firewalls.  **
> **
>
>  ****
>
> There’s also operational overhead, good luck getting any decent
> troubleshooting out of the lower levels before it escalates to senior.
> Ditto with implementations. ****
>
>  ****
>
> I’ve had a very bad experience with an extremely complex, BGP as internal
> topology network with 27 VRFs and multiple points of redistribution, any
> routing change turned into a CCIE lab. ****
>
>  ****
>
> *From:* Tom Storey [mailto:tom at snnap.net <tom at snnap.net>]
> *Sent:* Wednesday, 17 April 2013 9:51 AM
> *To:* Johann Lo
> *Cc:* Mitchell Warden; ausnog at lists.ausnog.net
> *Subject:* Re: [AusNOG] VPLS OSPF question****
>
>  ****
>
> Im aware iBGP generally requires an IGP as well, to distribute loopback
> information at a minimum, and that is is generally slower to propagate
> updates than the likes of an IGP - thats all 101 stuff.****
>
>  ****
>
> What I dont get is why this would make BGP any less useful internally. I
> mean, if I had to peer 200 sites across a VPLS I'd probably do it with some
> form of BGP rather than an IGP.****
>
>  ****
>
> On 17 April 2013 00:05, Johann Lo <Johann.Lo at aptel.com.au> wrote:****
>
> No, not really…. Why do you think most people run their iBGP on top of
> something else?****
>
>  ****
>
> Also BGP convergence time is inferior to OSPF/EIGRP/ISIS****
>
>  ****
>
>  ****
>
> *From:* ausnog-bounces at lists.ausnog.net [mailto:
> ausnog-bounces at lists.ausnog.net] *On Behalf Of *Tom Storey
> *Sent:* Wednesday, 17 April 2013 4:40 AM
> *To:* Mitchell Warden****
>
>
> *Cc:* ausnog at lists.ausnog.net
> *Subject:* Re: [AusNOG] VPLS OSPF question****
>
>  ****
>
> Whats wrong with using BGP internally. Isnt that what iBGP is for?****
>
>  ****
>
> On 16 April 2013 10:34, Mitchell Warden <wardenm at wardenm.net> wrote:****
>
> Hi Brad,
>
> Some ideas below. There are a lot of considerations...
>
> OSPF
> - Will run fine on a VPLS service. I've done this with up to 30 or so
> sites in the past and it works well.
> - 200 is a lot of sites - I would try to break it down to multiple smaller
> domains.
> - 50 routers in an area isn't a big deal. It will depend on the CPU and
> the number of updates, but even 200 is unlikely to be a problem.
> - 200 neighbour adjacencies however might be a big deal (they're all on
> the same broadcast domain). I think 200 is too many.
>
> BGP
> - All of the routers could be in the same subnet so they would be able to
> reach each other directly without an IGP, if you build neighbors with
> interface addresses instead of loopbacks.
> - You would need to use route reflectors to avoid having to mesh all 200
> routers.
> - I can't think of any reason BGP would reduce available bandwidth.
> - BGP isn't really designed to be used internally, and I would try to
> avoid using it that way.
> - BGP is probably better than OSPF if you can't reduce the size of the
> domain. It will be more scalable and probably more reliable.
>
> Cheers.
> Mitchell****
>
>  ****
>
>  ****
>
> *Johann Lo*****
>
> *Senior IP Network Engineer*****
>
>  ****
>
> <http://www.aptel.com.au/>****
>
> *    Asian Pacific Telecommunications*****
>
> ****
>
>    Level 14, 1 Queens Road, Melbourne, Victoria, 3004****
>
>    *p:* 03 9863 9863 *f:* 03 9863 7701 ****
>
>    *e:* Johann.Lo at aptel.com.au  *w:* www.aptel.com.au****
>
>  ****
>
>    <http://www.aptel.com.au>****
>     ------------------------------
>
>  ****
>
> *Notices* - (1)  If it appears that this email has been sent to you in
> error, please delete it (including any attachments) immediately and let the
> sender know by reply email.  This email may contain confidential
> information and may be privileged.  You may be acting unlawfully if you
> use, disclose, keep or rely upon that information.  (2)  This email and any
> attachment may not be free of viruses or defects.  The sender is not liable
> for anything whatsoever including damage, loss and liability that you
> experience because you have received this email and notes that you should
> ensure that your IT system is properly safeguarded.  (3)  If this email is
> not sent in direct connection with the company’s business, the company does
> not endorse the content.****
>
>  ****
>
>  ****
>
> *Johann Lo*****
>
> *Senior IP Network Engineer*****
>
>  ****
>
> <http://www.aptel.com.au/>****
>
> *    Asian Pacific Telecommunications*****
>
> ****
>
>    Level 14, 1 Queens Road, Melbourne, Victoria, 3004****
>
>    *p:* 03 9863 9863 *f:* 03 9863 7701 ****
>
>    *e:* Johann.Lo at aptel.com.au  *w:* www.aptel.com.au****
>
>
>
> ----- Original Message -----
> From: Brad McGinn
> [mailto:the_xorach at yahoo.com]
> To: ausnog at lists.ausnog.net
> [mailto:ausnog at lists.ausnog.net]
> Sent: Tue, 16 Apr 2013 17:14:02
> +1000
> Subject: [AusNOG] VPLS OSPF question
>
>
> > Hi AusNog list,
>
> Long time listener, first or second time caller.
>
> I
> > know this list is pretty specific to Service Providers so I'm hoping any
> of
> > you who not only know carrier networks, but also have an insight into
> > enterprise networks maybe able to help me to get a view (or even help
> > understanding) of the pros and cons of running OSPF or BGP across a VPLS
> > network.
>
> I respectfully ask your advice.
>
> I am an enterprise
> > network engineer, not a service provider however I hope you don't hold
> that
> > against me.  We run OSPF in our Data Centre and BGP into a MPLS network
> > that all of our sites connect into.
>
> My fairly basic understanding of VPLS
> > is kind of like EoMPLS or even one big broadcast domain.  I assume any
> > IGP could potentially work across it but some factors must be taken into
> > consideration:  eg flapping sites, latency, reference bandwidth, DR/BDR
> > placement, multicast transmission and so on.
>
> So, with that in mind, I'm
> > wondering the following:
> -    would it be wise to run an IGP across a
> > VPLS backbone with over 200 sites? or would BGP be better? or even
> something
> > else?
> -    if an IGP is the go, would one use OSPF?
> -    if OSPF, do
> > you think it would be wiser to run a separate OSPF process for the VPLS
> > connected sites and a separate OSPF process for the DC?  and then
> > redistribute or just summarise right there? (so as to protect the DC from
> > OSPF recalculations when sites go up and down)
> -    if BGP would be the
> > go I'm wondering how one might go about it..  I know that all iBGP
> > neighbours must have a route to the peering IP of all other iBGP
> routers so
> > I would assume an IGP must be run anyway???
> -    cisco say that
> > anything more than 50 routers on an area is a bad idea, so if I have over
> > 200 sites potentially on the VPLS, will OSPF cut it?
>
> I guess i'm just
> > trying to get my head around the different technology.  I'd love to keep
> > the stability that BGP brings, but also would like to be able to make use
> > of the bandwidth that VPLS gives.
>
> Any hints or tips will be gratefully
> > received and thank you for any help.  If you would like to keep from
> > cluttering up subscriber's inboxes, please reply offlist.
>
> Again, thanks
> > for any help.
>
> Regards,
>
> Brad David****
>
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog****
>
>  ****
>
> ***************************************
>
> This email has been scanned by Asian Pacific Telecommunications Hosted Security.****
>
> Powered by TrendMicro.****
>
> For more information please visit www.aptel.com.au****
>
> ***************************************
>
>    ****
>
>  ****
>
> ***************************************
>
> This email has been scanned by Asian Pacific Telecommunications Hosted Security.****
>
> Powered by TrendMicro.****
>
> For more information please visit www.aptel.com.au****
>
> ***************************************
>
>    ****
>
>
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog****
>
>  ** **
>
> _______________________________________________
> AusNOG mailing list
> AusNOG at lists.ausnog.net
> http://lists.ausnog.net/mailman/listinfo/ausnog
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20130417/744f5487/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 11835 bytes
Desc: not available
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20130417/744f5487/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.gif
Type: image/gif
Size: 2969 bytes
Desc: not available
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20130417/744f5487/attachment.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image003.jpg
Type: image/jpeg
Size: 81748 bytes
Desc: not available
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20130417/744f5487/attachment-0001.jpg>


More information about the AusNOG mailing list