Date: Thu, 19 Mar 2009 20:08:09 -0400 From: Steve Bertrand <steve@ibctech.ca> To: Christopher Arnold <chris@arnold.se> Cc: isp@freebsd.org Subject: Re: ISPs? Message-ID: <49C2DE69.2070409@ibctech.ca> In-Reply-To: <alpine.BSF.2.00.0903200000180.5570@localhost> References: <153046.19925.qm@web63901.mail.re1.yahoo.com> <907077794.20090317173752@homelink.ru> <49C05E35.8070609@ibctech.ca> <001501c9a795$07058de0$1510a9a0$@com> <49C1C3D0.5060304@neely.cx> <CC6BF6C0-D134-4DE6-9D47-17E01AA71BBB@ekalb.net> <5F9EF08A583352985E262800@tok> <49C24561.5090301@spekreijse.net> <49C253FE.3010408@ibctech.ca> <49C2583D.30502@spekreijse.net> <alpine.BSF.2.00.0903191554160.5077@localhost> <49C2CDAC.60500@ibctech.ca> <alpine.BSF.2.00.0903200000180.5570@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
Christopher Arnold wrote: > > On Thu, 19 Mar 2009, Steve Bertrand wrote: >> Is anyone in a position to possibly eBGP multi-hop a full table to a >> route server I have internally here? Of course I would completely >> null-route the learnt routes, obey any no-export communities, and forbid >> my internal route server from distributing the routes into the network. >> > Shure no problem, give me a host and you will receive a table from London. > AS39779 ip 87.117.214.90 Chris, given that you provided your info publicly, I figured you wouldn't mind me providing my config to the list. Before I do, I'd like to share a thought I had, to see if there is any interest. I have two dual-CPU, 4GB servers sitting in my lab idle. These units can be expanded to 32GB of RAM. Both of the boxes are identical Rackable Systems units. If there is enough interest, perhaps we could configure these units up, one with FreeBSD/Quagga, and the other with Open*, and feed these boxes with as many full tables as we collectively can. I could then set up lab networks in behind each one, and test things such as convergence time and the like. We could also use this as a test to measure how particular OS 'tweaks' affect routers with a large scale of peers and tables. If others are interested in such a project, let me know. It would be a community effort, so access to the boxes via the management plane would obviously be available. Anyway, here is my quick & dirty peering arrangement for the route server: bgp router-id 208.70.111.101 neighbor rs-feeds peer-group neighbor rs-feeds ebgp-multihop 254 neighbor rs-feeds update-source lo101 neighbor rs-feeds soft-reconfiguration inbound neighbor rs-feeds maximum-prefix 310000 neighbor rs-feeds prefix-list TESTING-DENY-OUT out neighbor rs-feeds route-map ROUTE-SERVER-IN in neighbor 87.117.214.90 remote-as 39779 neighbor 87.117.214.90 description RS Peer chris@arnold.se route-map ROUTE-SERVER-IN permit 10 set community no-export no-advertise additive set ip next-hop 192.0.2.2 ip prefix-list TESTING-DENY-OUT seq 5 deny 0.0.0.0/0 le 32 ...and the platform currently (this is one USB thumb stick image I haven't upgraded yet): # zebra -v zebra version 0.99.11 (running bgpd, ospf, zebra, vtysh) # uname -a FreeBSD rs.eagle.ca 7.1-PRERELEASE FreeBSD 7.1-PRERELEASE #0: Fri Aug 29 12:46:00 EDT 2008 Steve
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49C2DE69.2070409>