IRC Logs

Network: freenode Channel: #cisco

	Enter your search terms Submit search form
	Web www.irclog.org

[00:23]<fzrwr_>actually it should be default
[00:23]<zzwffzdnz>AS1 needs to peer with AS3?
[00:24]<fzrwr_>maybe a filter is stopping it?
[00:24]<msxjv>I take this to mean R1 peers to AS1 and R2 to AS3
[00:24]<msxjv>and they want to share routes
[00:24]<msxjv>internally
[00:24]<msxjv>so R1 and R2 are a part of AS_X
[00:25]<gjsw>yes.
[00:25]<gjsw>R1 and R2 are part of AS2.
[00:25]<msxjv>should be a simple IBGP config
[00:25]<gndyvx>yup
[00:25]<msxjv>be sure not to transit
[00:25]<msxjv>but you probably already know that =P
[00:25]<gjsw>actually
[00:25]<gjsw>the goal is to get the middle routers to transit
[00:25]<msxjv>that's fine too
[00:26]<gjsw>They are already neighbors, what else is there to do?
[00:26]<msxjv>if that's what you intend, just means you aren't filtering
[00:26]<gjsw>There are no filters at this time. very simple config router bgp 2, neighbor IP remote-as #, network 2.0.0.0 mask 255.0.0.0
[00:27]<gjsw>I dont see why they arent sharing externally learned routes
[00:27]<fmnwvnz>nosx: are your network statements correct?
[00:27]<gjsw>yes, they are. =P
[00:27]<gjsw>If R1 learns networks advertised by AS1, shouldnt it tell R2 as they are IBGP peers?
[00:27]<fmnwvnz>do you have a 2.0.0.0 subnet assigned to an interface?
[00:27]<fmnwvnz>or is there a 2.0.0.0 route?
[00:27]<gjsw>lo0
[00:28]<fmnwvnz>BGP will not announce a network unless an entry is in the routing table
[00:28]<gndyvx>or you redistribute
[00:28]<gjsw>i assigned each asn a /16 loopback. it shows up as directly connected in the routing table
[00:28]<gndyvx>but.. ibgp peers will share information about external routes to each other
[00:29]<gndyvx>make sure to turn off sycronization
[00:29]<gjsw>sync OFF? ill do no sync right now 1 sec.
[00:29]<fzrwr_>redist connected
[00:29]<fmnwvnz>sh ip bgp/sh ip bgp neigh -> pastebin.org
[00:29]<gndyvx>i don't recommend redistribt connected
[00:30]<gndyvx>expecially in the ibgp enviroment
[00:30]<[mzvzyw]>hey, what letter needs to be in the image name for VRF Lite support?
[00:30]<gndyvx>ibgp peers won't advertise routers that they originated to other ibgp peers
[00:30]<gndyvx>[Matrix]: the one up from ip base
[00:31]<fzrwr>damn wireless dropping out
[00:31]<gndyvx>got your nic back
[00:31]<[mzvzyw]>is that like is-mz?
[00:31]<[mzvzyw]>I'll just check feature navigator and quit being lazy. :-p
[00:32]<gndyvx>good call
[00:32]<gjsw>http://pastebin.com/738059 <~~pasted sh ip bgp / nei as requested.
[00:33]<gndyvx>show ip bgp sum is the good one :)
[00:33]<gndyvx>so back to what you are trying to do
[00:33]<gndyvx>what again are you trying to do?
[00:34]<gjsw>heh.
[00:34]<gndyvx>heh
[00:34]<gjsw>i just want the AS between to act as transit.
[00:34]<gjsw>its not telling the external peers on one side about the others.
[00:34]<gjsw>the external routes dont seem to make it from one ibgp peer to the other.
[00:34]<gjsw>(at least not their routing tables)
[00:35]<2zrgvj-l2>turn off synchronization
[00:35]<gjsw>did that.
[00:35]<gndyvx>so.. [as 1] [ Rt1<--ibgp-->Rt2 ] [ as 3]
[00:35]<2zrgvj-l2>you clear the session?
[00:35]<gndyvx>right?
[00:35]<2zrgvj-l2>soft?
[00:35]<gjsw>correct nemith
[00:35]<gndyvx>yeah
[00:35]<gndyvx>clear ip bgp * soft
[00:35]<gjsw>ok
[00:36]<jygc>I hope it's not a peering router and he's not going to leak his view ;-)
[00:36]<gndyvx>heh
[00:37]<gjsw>what did the clear accomplish?
[00:37]<gndyvx>bgp doesn't do anything untril you restart the session
[00:37]<gndyvx>any (well most) config changes require a clear
[00:38]<gndyvx>soft just says don't kill the tcp connection
[00:39]<gjsw>ok thats good to know. anything else im missing here?
[00:40]<fzrwr>to make a 5300 voip capable do you just need VFC's & supporting IOS?
[00:43]<gjsw>http://pastebin.com/738079 So the remote route is showing up in sh ip bgp, but its not in the routing table, and its not getting sent to external peers so we can act as a transfer AS. How do i tell it to act as a transfer? Redistribute something?
[00:47]<dzzc1lyvn>good guide
[00:47]<dzzc1lyvn>http://www.certmag.com/articles/templates/CM_gen_Article_template.asp?articleid=2018&zoneid=223
[00:49]<deee>can a vlan span two geographically separate switches ?
[00:49]<dzzc1lyvn>yep
[00:49]<dzzc1lyvn>span
[00:49]<dzzc1lyvn>rspan
[00:50]<gndyvx>mzzz: how geographically apart?
[00:50]<deee>50 miles connected by 10mb mpls
[00:51]<gndyvx>can you do layer 2 mpls?
[00:51]<gndyvx>rspan cannot "span" layer 3 links
[00:52]<2zrgvj-l2>espan for that, I believe...
[00:52]<2zrgvj-l2>but who runs code that recent?
[00:52]<dzzc1lyvn>yea
[00:54]<gndyvx>if you have l3 links you could get some high end hardware and do l2tpv3
[00:54]<deee>at the moment we have two sites (10.1.0.0/16 and 10.2.0.0/16) connected via mpls (cisco 2800 at each end) and we'd like to have 10.3.0.0/16 'overlapping' across both sites for load balancing/high availability bits .. is this possible ?
[00:55]<gndyvx>ah.. you don't want span then
[00:55]<sazdn>layer 2 vpn
[00:55]<2zrgvj-l2>or anycast
[00:55]<gndyvx>it is possible, either your mpls provider can do layer 2 vpn for you
[00:55]<gndyvx>or you can get some high end hw and to a puesuedo wire with l2tpv3
[00:56]<2zrgvj-l2>what kind of application are you running?
[00:56]<gndyvx>2800 will not do l2tpv3
[00:57]<gndyvx>nor will it do layer 2 vpn's over mpls
[00:57]<deee>we're in the process of looking at our (current non-existent) business continuity plan .. basically looking at the best way of load balancing/failing over an ISA server and IIS servers
[00:57]<sazdn>a 2800 should beable to do the 28800pps neccesary for wirespeed 10mbit
[00:57]<gndyvx>sure
[00:57]<gndyvx>bandwidth isn't a problem

Back to #cisco
Or go to some related logs:
#debian
#gentoo
#macosx
#gentoo
#gentoo