http://permalink.gmane.org/gmane.network.quagga.user hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.network.quagga.user/12986 <pre>Did you try the following: router bgp xxx redistribute kernel +1 on route tagging, it would be very useful. Regards, Barry On Thu, May 16, 2013 at 3:14 PM, David H &lt;ispcolohost-Re5JQEeQqe8AvxtiuMwx3w&lt; at &gt;public.gmane.org&gt; wrote: _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q&lt; at &gt;public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users </pre> Barry Friedman 2013-05-16T23:26:55 http://permalink.gmane.org/gmane.network.quagga.user/12985 <pre>I'm in the process of setting up Quagga for use as an RTBH router and am trying to make it as simple as possible for staff to add routes to be injected. At first I was going to do just bgpd but I need to apply different communities under certain conditions, so I added zebra and am doing redistribute static with a route map on the bgpd side. Now I can go into zebra and add static routes and they get populated to bgp and off to the border routers and upstreams just fine, but looking to make that easier. I noticed on linux that "ip route show" lists routes I've added for blocking in zebra, for example: blackhole 1.2.3.4 proto zebra I was curious if there's a linux command I could use to add those routes myself and then zebra would pick them up? That would make it very easy to script an interface for our noc staff to use to quickly blackhole or block someone. Perhaps something like "ip route add blackhole 1.2.3.4/32 proto zebra"? I did that and it ended up in the routing table of the kernel but not in z</pre> David H 2013-05-16T22:14:45 http://permalink.gmane.org/gmane.network.quagga.user/12984 <pre>Howdy, I was looking for an equivalent of `clear ip ospf process` which would restart ospfd. Is there any similar thing in quagga? I see a few `clear bgp ...` thingies. -- Ricky Charlet Software Dev / Routing Dude: Aries team, Roseville CA ricky.charlet-VXdhtT5mjnY&lt; at &gt;public.gmane.org&lt;mailto:ricky.charlet-VXdhtT5mjnY&lt; at &gt;public.gmane.org&gt; USA: 916.785.2090 </pre> Charlet, Ricky 2013-05-16T21:22:04 http://permalink.gmane.org/gmane.network.quagga.user/12983 <pre>Hi Kamil, Thanks your response, my problem is solved. Cristian Em 16-05-2013 06:04, szastan escreveu: </pre> CPN 2013-05-16T13:12:43 http://permalink.gmane.org/gmane.network.quagga.user/12982 <pre>I'm curious too. We've seen this crashes a few times but were not able to pinpoint the cause. It seems like bgpd crashes because of an assert that checks prefix start/end values. Thanks, Sergey On 5/15/13 4:03 PM, Bulend Malik wrote: </pre> Sergey Polzunov 2013-05-16T09:10:25 http://permalink.gmane.org/gmane.network.quagga.user/12981 <pre>Hi Cristian, as Banyan said, this is rather related to your Linux config. It's a blind shot, but check if you have in /etc/sysctl.conf disabled spoof protection: net.ipv4.conf.default.rp_filter=0 net.ipv4.conf.all.rp_filter=0 regards Kamil Wiadomość napisana przez CPN &lt;cpn-UF1JytKP+LQ&lt; at &gt;public.gmane.org&gt; w dniu 15 maj 2013, o godz. 23:58: </pre> szastan 2013-05-16T09:04:05 http://permalink.gmane.org/gmane.network.quagga.user/12980 <pre> Hello,   yesterday around 16:10 CET we experienced kernel crash on a box running quagga and peering to Cogent. According to our findings we suspect some malicious/malformed/unsupported IPv6 updates. Our other upsream peering on exactly the same box did not experience any problems.   We are running quagga 0.99.20 on debian squeeze 3.0.28 x86_64.   Anybody else experienced similar issues with Cogent or other ISP?   Thanks.   Miro _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q&lt; at &gt;public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users </pre> mironto 2013-05-16T08:19:48 http://permalink.gmane.org/gmane.network.quagga.user/12979 <pre>Hi! The crash never happened again and I assume it was not software-related. We did not upgrade to 0.99.22, it still runs 0.99.21. </pre> Kurt Jaeger 2013-05-16T04:44:58 http://permalink.gmane.org/gmane.network.quagga.user/12978 <pre>Hi Cristian, I think you send to the wrong list if the packets being dropped is not quagga related. You need the linux users for the issue. Basically, you need to check why it's dropped, make the dump first. Best regards, ------------ Banyan He Blog: http://www.rootong.com Email: banyan-b+asLkKy1cdBDgjK7y7TUQ&lt; at &gt;public.gmane.org On 5/15/13 2:58 PM, CPN wrote: </pre> Banyan He 2013-05-15T22:21:03 http://permalink.gmane.org/gmane.network.quagga.user/12977 <pre>Hi ALL... I have quagga (version 0.99.15) and i have two backbone. I having problem when the packet in at backbone A and out backbone B, the linux is droping the packet. Someone know what can happening? regards, Cristian </pre> CPN 2013-05-15T21:58:44 http://permalink.gmane.org/gmane.network.quagga.user/12976 <pre> have you solved this issue ? is it enough to upgrade to quagga-0.99.22 ? Thanks </pre> Bulend Malik 2013-05-15T14:03:22 http://permalink.gmane.org/gmane.network.quagga.user/12975 <pre>Hi! [...] It never happened again, so I assume it was no software issue. </pre> Kurt Jaeger 2013-05-15T18:15:37 http://permalink.gmane.org/gmane.network.quagga.user/12974 <pre>Dear all, Does the BGP daemon support more than what the official documentation says it does ? If so, is there any other documentation available that has an updated list of commands supported ? If not then fine. thanks and regards, balaji venkat PS : I looked at www.nongnu.org/Quagga documentation. _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q&lt; at &gt;public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users </pre> Balaji venkat Venkataswami 2013-05-08T19:18:58 http://permalink.gmane.org/gmane.network.quagga.user/12973 <pre> Hello guys Is there result or soluton about that ? </pre> Bulent Malik 2013-05-07T20:40:45 http://permalink.gmane.org/gmane.network.quagga.user/12972 <pre>Hi, For 3 years, I ran quagga on a what was a (effectively) desktop at our HQ. We had 2 full BGP feeds. The desktop config was: 1. Pentium 2.x GHz 2. Quad-port Intel NIC 3. 2GB RAM 4. HW RAID-1 (2x160GB PATA) 5. Debian 4 or so (this was 2006) the machine ran flawlessly, without a hiccup. It also was the DHCP/DNS Server (multiple internal subnets), and squid as a proxy for outgoing traffic. It was routing 2 6M WAN links, and multiple internal networks. Your Dell 1450 should not break a sweat, or even notice the load. The only issue may be that your PCI bus gets saturated, you may want a quad-port NIC on PCI-E. The other minor quibble is hardware failure (which is why I was using a desktop, I had 40 exact same configs being used by staff, so if I had had a power-supply or motherboard failure, I could have swapped with whoever was on leave today). </pre> Sanjeev Gupta 2013-05-04T03:16:27 http://permalink.gmane.org/gmane.network.quagga.user/12971 <pre>Hi, I have to set up 2 quagga that will have to handle 2 BGP (full table) neighbors, an IBGP session between them and some ospf, like this : || Datacenter A || Datacenter B || AS ISP 1 AS ISP 2 || AS ISP 1 AS ISP 2 | | || | | | | || | | | | || | | | | Our AS | | +---+------+---+ || +---+------+---+ | | || | | | Quagga A | || | Quagga B | | |/___________IBGP OSPF____________\| | +------+-------+ || +------+-------+ |</pre> Ace Slash 2013-05-04T00:03:35 http://permalink.gmane.org/gmane.network.quagga.user/12970 <pre>When I run /etc/init.d/quagga restart (or start) I am getting this message "*..* *Loading capbility module if not done already * *Starting quagga daemons (prio:10): zebra/usr/lib/quagga/zebra: symbol lookup error: /usr/lib/quagga/zebra: undefined symbol: setsockopt_ipv6_multicast_hop*s " I am using quagga version 0.99.20.1 The same thing is working over other VMs except this one. All my VM use Ubunutu 12.04. I tried rebooting. reinstalling quagga etc. I also tried various combinations, without zebra enabled in daemons file its works fine, as soon as I enable zebra, it gives this error. It will be great if I can get some help form the community. Also let me know what information I should share to assist debugging this issue. </pre> Arpit Gupta 2013-05-03T20:08:49 http://permalink.gmane.org/gmane.network.quagga.user/12969 <pre>Yes, I'm planing to use quagga + pmacctd with bgp_daemon running. I run it on debian 6.0. Next I'm planing to export the date using pmacctd and nfprobe plugin and show the date on other machine using as-stats. I have eth0 to one peer and eth1 to second peer. Now I'm trying to configure second instance pmacctd to work with second ethernet interfaces and put all together to netflow to other machine to as-stats. I'l let you know how the solution will by stable. </pre> sidlo.marek-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org 2013-05-02T23:00:03 http://permalink.gmane.org/gmane.network.quagga.user/12968 <pre>Are you planning to use bgp_daemon in the same router you use quagga? That was my "fear", freebsd+quagga is unstable enough by themselves, so I didn't wanna complicate it more including bgp_daemon. let us know your progress mate! cya. </pre> André Gustavo N. Lopes 2013-05-02T21:46:52 http://permalink.gmane.org/gmane.network.quagga.user/12967 <pre>Andre, thank you for information! I'll try to connect quagga with pmacctd using bgp_daemon and if it fails or will be unstable then I'll try to do in in your way. Regards </pre> sidlo.marek-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org 2013-05-02T20:49:47 http://permalink.gmane.org/gmane.network.quagga.user/12966 <pre>My edges are running pmacctd like this: debug: false daemonize: true interface: bce0 logfile: /var/log/pmacct.log aggregate: src_host, dst_host, src_port, dst_port, src_as, dst_as, proto, tos plugins: nfprobe, memory nfprobe_receiver: mynetflowserver:9992 nfprobe_version: 5 networks_file: /usr/local/etc/pmacct_probe_nets.conf pmacctd_net: file pmacctd_as: file /usr/local/etc/pmacct_probe_nets.conf looks like this: ASN,prefix/lenght 15169,1.0.0.0/24 56203,1.0.4.0/22 2519,1.0.16.0/23 2519,1.0.18.0/23 ... etc There is a place where this information is updated in a daily basis: http://data.caida.org/datasets/routing/routeviews-prefix2as/2013/05/ Hope it helps </pre> André Gustavo N. Lopes 2013-05-02T18:20:16 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.network.quagga.user