I’m sure that just saying what everyone else is here but here’s my 2¢...
I’d use iBGP to have full routing tables on the more upline routers (not necessarily only the peering routers) so that the best upload and download path can be calculated from the routing table. Then advertise the default route from all of those iBGP routers into the OSPF routers. Then you can advertise all your IP’s with all of your uplines and use prepends on some advertisements to create shorter paths to load balance to the needed degree. Alternatively you can have multiple routing tables one for IP’s you send to one BGP peer and one for the other… this gets complicated when wanting to fail over cleanly and it doesn’t take advantage of shorter paths for content on one carrier over the other. Also, you can most likely route out one carrier and back on the other as long as your not dropping invalid sessions at your edge routers because the session is valid on other routers. Sincerely, Joshaven Potter Google Hangouts: j...@g2wireless.co Cell & SMS: 1-517-607-9370 supp...@joshaven.com > On May 3, 2016, at 3:41 PM, That One Guy /sarcasm <thatoneguyst...@gmail.com> > wrote: > > We plan on not having to ever do it. but we stacked our priority customers on > one /24 so if there were service issues on one of the upstreams or if we > unexpectedly saturated one we could force them to use just the one > > On Tue, May 3, 2016 at 2:09 PM, John Babineaux <john.babine...@reach4com.com > <mailto:john.babine...@reach4com.com>> wrote: > I would advertise all networks on each connection. > > > > You can Prepend to an extent to help prefer traffic coming back on a certain > connection. This cannot be prepended to high or some strip it off. It broke > the internet once by someone putting a really large one….. > > > > If one connection is really bad and the other is really good (to many hops or > very few) you will really only use one most of the time. BGP will send the > traffic out of the connection that is closer. If one of the connections goes > out it will stop advertising on that link to the world. The working > connection will be the only one advertising. > > > > If you are expecting problems I normally filter all traffic to keep things > from flapping on that connection until the work is done. > > ____________________________________ > > John Babineaux > > System Administrator > > REACH4 Communications | Website: www.REACH4Com.com <http://www.reach4com.com/> > Phone: 337-783-3436 x105 <tel:337-783-3436%20x105> | Email: > john.babine...@reach4com.com > > 927 N Parkerson Ave, Crowley, LA 70526 > > > > > > > > From: Af [mailto:af-boun...@afmug.com <mailto:af-boun...@afmug.com>] On > Behalf Of That One Guy /sarcasm > Sent: Tuesday, May 03, 2016 1:20 PM > To: af@afmug.com <mailto:af@afmug.com> > Subject: Re: [AFMUG] BGP and OSPF > > > > thank you guys > > > > Now another question, one of our providers is solid, the other..well. What > kind of issues can come up with a basic BGP implementaion (we are taking the > full tables) that will hurt us. Like is there some way that even if we stop > announcing one of the /24 on their circuit theyll aggregate it on their own > into the /22 of the ASN? > > > > You have to remember this is the upstream that moved our bandwidth from an > ethernet port to an SPF one morning without mentioning it to us and without > verifying we had a module, they also send a shitface drunk tech, and for > kicks one day after a failed routing migration, they went ahead and > implemented the changes anyway in the middle of the week, just because they > could. > > > > So any upstream BGP shenanigans I fully expect to see > > > > > > > > On Tue, May 3, 2016 at 1:09 PM, John Babineaux <john.babine...@reach4com.com > <mailto:john.babine...@reach4com.com>> wrote: > > One more thing BGP will pass a Default route to OSPF that will propagate it > so that’s how it will know. > > > > From: Af [mailto:af-boun...@afmug.com <mailto:af-boun...@afmug.com>] On > Behalf Of John Babineaux > Sent: Tuesday, May 03, 2016 1:08 PM > To: af@afmug.com <mailto:af@afmug.com> > Subject: Re: [AFMUG] BGP and OSPF > > > > It’s pretty simple. > > > > You create a connection with your up streams using your ASN and their ASN. > You need ip connectivity to the other router (prob your gateway but could be > another router). And a password if required or preferred. Next you setup > filters to only allow what networks you want to pass upstream and what you > want to accept. Then you add what networks you what to share to the world > statically or to pass them from OSPF. > > > > Keep in mind they will create filters to block anything that you didn’t tell > them that you will pass. If you say x.x.x.x/22 they will only allow that > exact thing. You should only pass nothing lower than a /24 as most will > block it. It’s your choice for full routing tables or just /8 or /16 etc. > You can also get a default route if you don’t get full tables. > > > > Most of the other things I read was for getting things to work when you have > to connect to the other BGP router that’s not directly connected or your > gateway. > > > > ____________________________________ > > John Babineaux > > System Administrator > > REACH4 Communications | Website: www.REACH4Com.com <http://www.reach4com.com/> > Phone: 337-783-3436 x105 <tel:337-783-3436%20x105> | Email: > john.babine...@reach4com.com <mailto:john.babine...@reach4com.com> > 927 N Parkerson Ave, Crowley, LA 70526 > > > > > > > > From: Af [mailto:af-boun...@afmug.com <mailto:af-boun...@afmug.com>] On > Behalf Of That One Guy /sarcasm > Sent: Tuesday, May 03, 2016 12:36 PM > To: af@afmug.com <mailto:af@afmug.com> > Subject: Re: [AFMUG] BGP and OSPF > > > > I understand this, but its still a second new component to something I > already am not competent in > > > > On Tue, May 3, 2016 at 12:25 PM, can...@believewireless.net > <mailto:can...@believewireless.net> <p...@believewireless.net > <mailto:p...@believewireless.net>> wrote: > > iBGP and eBGP are the same thing as far as configuration goes. (i) means you > are using BGP internal to your network > > and (e) means you are using BGP to your providers. How you set them up is > exactly the same. > > > > The only different with multi-hop is that you have to set the maximum number > of hops away the BGP peer can be. > > > > On Tue, May 3, 2016 at 1:23 PM, That One Guy /sarcasm > <thatoneguyst...@gmail.com <mailto:thatoneguyst...@gmail.com>> wrote: > > No, I cant add any topology changes or try to bring in unfamiliar things like > ibgp or mpls/vpls while bringing in something unfamiliar like BGP unless I > run out of interim options > > > > On Tue, May 3, 2016 at 12:13 PM, can...@believewireless.net > <mailto:can...@believewireless.net> <p...@believewireless.net > <mailto:p...@believewireless.net>> wrote: > > With Mikrotik, depending on your typology, running iBGP over layer 3 is > probably better than EoIP. In the > > past, I wasn't happy with the way things worked over EoIP. It's now over a > direct, layer 2 connection and > > works 100%. You can run setup multi-hop BGP routing to do it over layer 3 > (single setting in Winbox) but > > it's not as clean. > > > > Could you have the path between the two routers be layer 2 via VPLS or just > using switches? > > > > On Tue, May 3, 2016 at 12:27 PM, That One Guy /sarcasm > <thatoneguyst...@gmail.com <mailto:thatoneguyst...@gmail.com>> wrote: > > VPLS is not currently present, thats on my "to learn" list > > > > On Tue, May 3, 2016 at 11:26 AM, Jesse DuPont <jesse.dup...@celeritycorp.net > <mailto:jesse.dup...@celeritycorp.net>> wrote: > > The two BGP routers do not need to be on the same L2 network for the iBGP > connection. > > > > Jesse DuPont > > Network Architect > email: jesse.dup...@celeritycorp.net <mailto:jesse.dup...@celeritycorp.net> > Celerity Networks LLC > > Celerity Broadband LLC > Like us! facebook.com <http://facebook.com/>/celeritynetworksllc > > Like us! facebook.com <http://facebook.com/>/celeritybroadband > <image001.png> > > On 5/3/16 10:25 AM, Mike Hammett wrote: > > A BGP speaker would be a router speaking BGP. In this case, most likely your > routers at the edge of your network that connect to your providers. > > Are the routers that are between your two BGP routers capable of running BGP, > resource wise? > > Can you do a VPLS tunnel between your two BGP routers? If not, what about a > VLAN? > > > > ----- > Mike Hammett > Intelligent Computing Solutions <http://www.ics-il.com/> > <https://www.facebook.com/ICSIL> > <https://plus.google.com/+IntelligentComputingSolutionsDeKalb> > <https://www.linkedin.com/company/intelligent-computing-solutions> > <https://twitter.com/ICSIL> > Midwest Internet Exchange <http://www.midwest-ix.com/> > <https://www.facebook.com/mdwestix> > <https://www.linkedin.com/company/midwest-internet-exchange> > <https://twitter.com/mdwestix> > The Brothers WISP <http://www.thebrotherswisp.com/> > <https://www.facebook.com/thebrotherswisp> > > > <https://www.youtube.com/channel/UCXSdfxQv7SpoRQYNyLwntZg> > From: "That One Guy /sarcasm" <thatoneguyst...@gmail.com> > <mailto:thatoneguyst...@gmail.com> > To: af@afmug.com <mailto:af@afmug.com> > Sent: Tuesday, May 3, 2016 11:13:36 AM > Subject: Re: [AFMUG] BGP and OSPF > > Mike, i said helmet, explain it to me like you would a 10 year old, then dumb > it down to my level from there. > > > > I dont know what a bgp speaker is > > > > On Tue, May 3, 2016 at 11:10 AM, Mike Hammett <af...@ics-il.net > <mailto:af...@ics-il.net>> wrote: > > Your OSPF network will just use default routes to get to your BGP speakers. > > Your BGP speakers with full routes will choose the best path. Your BGP > speakers should be connected together, via direct connection, layer 2 tunnel > (VPLS) or via intermediary iBGP speakers. Those iBGP speakers in the middle > of your network will route the correct way, based on BGP. > > > > ----- > Mike Hammett > Intelligent Computing Solutions <http://www.ics-il.com/> > <https://www.facebook.com/ICSIL> > <https://plus.google.com/+IntelligentComputingSolutionsDeKalb> > <https://www.linkedin.com/company/intelligent-computing-solutions> > <https://twitter.com/ICSIL> > Midwest Internet Exchange <http://www.midwest-ix.com/> > <https://www.facebook.com/mdwestix> > <https://www.linkedin.com/company/midwest-internet-exchange> > <https://twitter.com/mdwestix> > The Brothers WISP <http://www.thebrotherswisp.com/> > <https://www.facebook.com/thebrotherswisp> > > > <https://www.youtube.com/channel/UCXSdfxQv7SpoRQYNyLwntZg> > From: "That One Guy /sarcasm" <thatoneguyst...@gmail.com > <mailto:thatoneguyst...@gmail.com>> > To: af@afmug.com <mailto:af@afmug.com> > Sent: Tuesday, May 3, 2016 10:41:52 AM > Subject: [AFMUG] BGP and OSPF > > We currently have a /22 with 2 /24 statically routed in each of our > providers. We are moving to BGP. > > > > What Im still unclear on is how my OSPF network is going to decide on the > best path for data to flow externally > > > > can somebody give me the helmet version of how this is accomplished > > > > -- > > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team. > > > > > > > > > -- > > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team. > > > > > > > > > > > -- > > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team. > > > > > > > > > -- > > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team. > > > > > > > > > -- > > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team. > > > > > > > -- > > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team. > > > > > -- > If you only see yourself as part of the team but you don't see your team as > part of yourself you have already failed as part of the team.
