http://blog.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://comments.gmane.org/gmane.network.quagga.user/9835 _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users Kaiyu 2008-11-26T06:28:56 http://comments.gmane.org/gmane.network.quagga.user/9833 Hello! From http://icons.apnic.net you can read this announce. Which version of Quagga is ready for this or a patch is needed? Very thanks Regards --- Sim Sim 2008-11-19T15:11:41 http://comments.gmane.org/gmane.network.quagga.user/9830 _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users Kristiadi Himawan 2008-11-19T06:52:52 http://comments.gmane.org/gmane.network.quagga.user/9829 Hello. Does Quagga/bgpd supports conditional adv. feature from Cisco? I was digging on the website and found nothing. Wonder if someone has worked this out. People are asking for this feature about 7 years. I'm using BGP/static routes, (no OSPF) Thanks in advance. tx0 2008-11-19T00:24:17 http://comments.gmane.org/gmane.network.quagga.user/9828 This is a repeatable crash in 0.99.9. I had in my config the following (this is the session to my other quagga router, iBGP): router bgp 12345 neighbor 10.0.0.8 remote-as 12345 neighbor 10.0.0.8 description iBGP to ash2 over localnet neighbor 10.0.0.8 update-source 10.0.0.5 neighbor 10.0.0.8 next-hop-self Upon the command "no neighbor 10.0.0.8" bgpd crashes with the following in the log: BGP: Received signal 11 at 1227014986 (si_addr 0x0, PC 0x25e858); aborting... Program counter: /usr/lib/quagga/libzebra.so.0(sockunion_cmp+0x18)[0x25e858] Backtrace for 12 stack frames: /usr/lib/quagga/libzebra.so.0(zlog_backtrace_sigsafe+0x31)[0x26eff1] /usr/lib/quagga/libzebra.so.0(zlog_signal+0x25f)[0x26f65f] /usr/lib/quagga/libzebra.so.0[0x27c9ca] /lib/tls/libc.so.6[0x47d910] /usr/sbin/bgpd[0x806d3e5] /usr/sbin/bgpd[0x806eada] /usr/sbin/bgpd[0x806eee2] /usr/lib/quagga/libzebra.so.0(work_queue_run+0xd5)[0x27db75] /usr/lib/quagga/libzebra.so.0(thread_call+0x3f)[0x2613bf] /usr/sbin/bgpd(main+0x2db)[0x805c66b] /lib/ Dennis Gerasimov 2008-11-18T19:52:06 http://comments.gmane.org/gmane.network.quagga.user/9827 I'm in the process of doing an implementation of the new BGP MIB in Quagga. This MIB supports IPv6. My test environment consists of two Ubuntu virtual machines in Parallels. I have v6 reachability between the two VMs. VM #1 is running my modified code, basd on 0.99.11. VM #2 is running the ubuntu distribution of 0.99.9. Neither VM is running the zebra daemon since I don't require FIB modifications. The configuration on VM#2 is as follows: router bgp 7676 bgp router-id 10.0.0.2 neighbor 10.211.55.6 remote-as 7675 neighbor fec0::1 remote-as 7675 neighbor fec0::1 disable-connected-check ! address-family ipv6 network fec0::/48 neighbor fec0::1 activate exit-address-family Note that 10.211.55.6 is attached to the same peer and is used to test IPv4. The IPv6 session is coming up, however the fec0::/48 network is not being advertised. I'd appreciate any suggestions to help me advertise IPv6 reachability. Jeffrey Haas 2008-11-18T16:54:12 http://comments.gmane.org/gmane.network.quagga.user/9823 _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users audrius-uDjapQuiMQBeoWH0uzbU5w< at >public.gmane.org 2008-11-17T06:05:00 http://comments.gmane.org/gmane.network.quagga.user/9817 Hi, I'm using 2.6.24-19 kernel. The following message type is logged after upgrading to Quagga 0.99.11 Nov 12 22:22:20 RS2 bgpd[13434]: Can't set IP_TOS option for fd 10 to 0xc0: Operation not permitted Nov 12 22:22:20 RS2 bgpd[13434]: Can't set IP_TOS option for fd 11 to 0xc0: Operation not permitted Nov 12 22:22:20 RS2 bgpd[13434]: Can't set IP_TOS option for fd 12 to 0xc0: Operation not permitted Nov 12 22:22:20 RS2 bgpd[13434]: Can't set IP_TOS option for fd 13 to 0xc0: Operation not permitted Of corse, it really doesn't set precedence 6 (0xc0) and the messages is filling my syslog log, specially when I have a lot of neighbors configured. I know how to avoid the messages: disable logging of warning messages for daemons in syslog.conf. example: daemon.*;daemon.!=warning -/var/log/daemon.log But how can I avoid this error ??? Marius Postelnicu 2008-11-13T07:26:14 http://comments.gmane.org/gmane.network.quagga.user/9811 _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users Dan Cave 2008-11-12T17:18:10 http://comments.gmane.org/gmane.network.quagga.user/9809 Erwin, Take a look in /usr/include/linux/tcp.h: #define TCP_MD5SIG 14 /* TCP MD5 Signature (RFC2385) */ /* for TCP_MD5SIG socket option */ #define TCP_MD5SIG_MAXKEYLEN 80 struct tcp_md5sig { struct __kernel_sockaddr_storage tcpm_addr; /* address associated */ __u16 __tcpm_pad1; /* zero */ __u16 tcpm_keylen; /* key length */ __u32 __tcpm_pad2; /* zero */ __u8 tcpm_key[TCP_MD5SIG_MAXKEYLEN]; /* key (binary) */ }; Can you find that definition? -- Pedro Em Qua 12 Nov 2008, Erwin Lubbers escreveu: Pedro Torres 2008-11-12T12:25:17 http://comments.gmane.org/gmane.network.quagga.user/9799 Hello fellow Quaggers! I am trying to figure out how to configure iBGP between two quagga routers. The network diagram is very simple: [bandwidth provider 1]--bgp--[rtr1 10.0.0.1]---{internal network}---[rtr2 10.0.0.2]--bgp--[bandwidth provider 2] I would like to have the two routers in an active-active faiolover configuration, for which I will have VRRP running between them. To get there, I need first to be able to use both links, and I have been told that one way to do it is through iBGP. Unfortunately, there is very little info on how to configure it in the quagga documentation. My current understanding is that rtr1 will have the following in its config: router bgp 12345 bgp router-id 10.0.0.1 neighbor 10.0.0.2 remote-as 12345 neighbor 10.0.0.2 description iBGP-PEER1 neighbor 10.0.0.2 route-reflector-client Whereas rtr2 will have the same, but with IP's reversed and a different description. Here are my questions: What distinguishes iBGP from regular BGP configuration? Since both routers Dennis Gerasimov 2008-11-11T15:28:00 http://comments.gmane.org/gmane.network.quagga.user/9798 Hi, What does LS age is equal to MaxAge mean exactly and why would I get it over and over again in a continual stream? Both my directly connected uplink CISCO's neighbors show as full: Neighbor ID Pri State Dead Time Address Interface RXmtL RqstL DBsmL 172.16.10.47 1 Full/DROther 35.462s 10.255.14.29 gre2:10.255.14.30 1 0 0 172.16.10.45 1 Full/DROther 36.319s 10.255.13.29 gre1:10.255.13.30 247 0 0 but I have no ospf routes: K>* 0.0.0.0/0 via 2xx.x.x.1, vr1 O 10.254.150.0/24 [110/10] is directly connected, vr0, 00:00:14 C>* 10.254.150.0/24 is directly connected, vr0 C>* 10.255.3.10/32 is directly connected, vr0 C>* 10.255.4.10/32 is directly connected, vr0 O 10.255.13.28/30 [110/10] is directly connected, gre1, 00:00:16 C>* 10.255.13.28/30 is directly connected, gre1 O 10.255.14.28/30 [110/50] is directly connected, gre2, 00:00:14 C>* 10.255.14.28/30 is directly connected, gre2 C>* 127.0.0.0/8 is directly connected Steve Clark 2008-11-11T13:35:14 http://comments.gmane.org/gmane.network.quagga.user/9782 Hi, I'm trying to compile Quagga 0.99.11 on Linux kernel 2.6.27.5 with RFC2385 compiled into the kernel. But the configure command of Quagga shows me: configure:30317: checking whether TCP_MD5SIG is declared configure:30415: gcc -c -Os -fno-omit-frame-pointer -g -std=gnu99 - Wall -Wsign-compare -Wpointer-arith -Wbad-function-cast -Wwrite-s trings -Wmissing-prototypes -Wmissing-declarations -Wchar-subscripts - Wcast-qual conftest.c >&5 conftest.c: In function `main': conftest.c:194: error: `TCP_MD5SIG' undeclared (first use in this function) conftest.c:194: error: (Each undeclared identifier is reported only once conftest.c:194: error: for each function it appears in.) configure:30421: $? = 1 configure: failed program was: | /* confdefs.h. */ | #define PACKAGE_NAME "Quagga" | #define PACKAGE_TARNAME "quagga" | #define PACKAGE_VERSION "0.99.11" | #define PACKAGE_STRING "Quagga 0.99.11" Could someone give me some directions where to look to solve it? Regards, Erwin Erwin Lubbers 2008-11-10T16:41:08 http://comments.gmane.org/gmane.network.quagga.user/9777 I will try, but the route over fiber doesent even show under show ip route ospf or the kernel routing table, so I'm not sure that's the soltuion? Mind you this only happens once you unplug fiber, then plug it back in. ----- Original Message ----- From: quagga-users-bounces Sent: 11/04/2008 03:08 PM To: Chris Beach <chrisb-Ymy/fnJKlnTQT0dZR+AlfA< at >public.gmane.org> Cc: quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org Subject: [quagga-users 10103] Re: OSPFD routing between networks, and/or "received on link eth0 but no ospf_interface" Dear Chris, set higher interface cost at the links over the wlan. Kind regards, Ingo Flaschberger _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users Chris Beach 2008-11-04T21:43:05 http://comments.gmane.org/gmane.network.quagga.user/9766 [trying again, my mail client is messing with my "From:" headers] On Oct 30, 2008, at 1:12 PM, Peter van den Heuvel wrote: For some reason this is often described confusingly. I'll try to summarize. A ---- B ---- C Three routers A,B,C are all on a broadcast network. A announces things to B. B announces things to C. (Other announcements may happen in the reverse direction, we're not concerned with that.) If A announces route R to B, and B announces R to C, then B will normally make A the next hop in its announcement to C. That's because there's no reason for C to send traffic for R to B when it can send it to A directly. (ICMP redirects are based on similar reasoning.) ...except there actually may be a very good reason for C to pass traffic back via B, rather than directly to A. For example, A may refuse packets coming from C. Or the underlying network might be rather different than a classic broadcast network (ATM, for example), and packets from C might not be able to reach A direct Alexis Rosen 2008-10-31T17:20:37 http://comments.gmane.org/gmane.network.quagga.user/9763 _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users Chris Beach 2008-10-31T14:55:22 http://comments.gmane.org/gmane.network.quagga.user/9760 _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users Kaiyu 2008-10-31T10:21:31 http://comments.gmane.org/gmane.network.quagga.user/9750 Good morning, I have quagga (0.98.6) and ospfd running on a CentOS 5.1 (2.6.18-53.1.4.el5). I can't get the networks attached to the loopback virtual interfaces to advertise, it may be pilot error but it was working and then stopped. The adjacencies form correctly and I can see routes added and removed from the other two Cisco devices attached to area 0 on the Quagga machine but I do not see the Quagga machine's networks on the Cisco devices. When I 1st had this working none of the redistribute statements, abr-type or stub commands were present. These were added in an attempt to remedy the issue. Ascii art and config below, TIA. 4.4.4.0/24 (area 4)Quagga-------------Cisco-12.3-----1.1.1.0/24 (Area 1) 5.5.5.0/24 (area 5) | | Area0 | Cisco-Pix-6.3-----2.2.2.0/24 (area 2) Quagga config ---------------------------- Wade Blackwell 2008-10-29T16:53:47 http://comments.gmane.org/gmane.network.quagga.user/9748 Hi there. I have 2 uplinks on my quagga 0.99.10 on mandrake 2.6.26.3. The first (AS34211) is main, the second (AS8470) is backup. I have another severe problem today. I am hosting a PBX, and a voIP-extension is connected remotely via another ISP. The extension died and was down 20 minutes. I tried to figure out what's going on and got the following info. Pinging the extension failed. Traceroute to it gave dead end (192.168.0.1 that is default route in case both uplinks are down). thunder_bgpd# sh ip bgp 78.107.83.67 BGP routing table entry for 78.106.0.0/15 Paths: (2 available, no best path) Not advertised to any peer 34211 3216 8402, (suppressed due to dampening) 81.9.48.65 (metric 1) from 81.9.48.65 (81.9.48.65) Origin IGP, metric 10, localpref 110, weight 2000, valid, external Dampinfo: penalty 999, flapped 12 times in 00:46:36, reuse in 00:06:12 Last update: Wed Oct 29 12:37:42 2008 8470 3216 8402, (suppressed due to dampening) 87.118.220.62 (metric 1) from 87.118.220 billy-SdyTfEM+II2HXe+LvDLADg< at >public.gmane.org 2008-10-29T10:31:21 http://comments.gmane.org/gmane.network.quagga.user/9741 Hello, Is there some way to determine what is left in the "Link state request list" when quagga hangs in the Loading state? We have a number, greater than 50, remote sites connected back to a two cisco systems on different ds3 from different providers that are running ospf over gre/vpn. We are having trouble with one site where quagga is getting stuck in : Neighbor ID Pri State Dead Time Address Interface RXmtL RqstL DBsmL 172.16.10.45 1 Loading/DROther 30.524s 10.255.13.213 gre1:10.255.13.214 0 2 0 172.16.10.47 1 Loading/DROther 34.442s 10.255.14.213 gre2:10.255.14.214 0 2 0 The cisco systems show the adjacency as full. My question is how to I go about debugging this problem? I understand that the quagga side is missing some part of its database that is suppose to be sent from the cisco, if I do a 'sh ip os database' I see a lot of the database, how do I tell what quagga is waiting for? We were first using 0.99.6 and had th Steve Clark 2008-10-28T14:18:23 http://comments.gmane.org/gmane.network.quagga.user/9739 Has anybody seen this: # show ip route 0.0.0.0 Routing entry for 0.0.0.0/1 Known via "ospf", distance 110, metric 327716, best Last update 03:50:05 ago * 172.30.18.241, via eth0 * 172.30.18.245, via eth1 There is no 0.0.0.0/1 on the adjacent routers. Here's the ospfd.conf: interface eth0 description Router Link #1 ip address 172.30.18.242/30 ip ospf cost 500 ip ospf network point-to-point ip ospf priority 0 ipv6 nd suppress-ra ! interface eth1 description Router Link #2 ip address 172.30.18.246/30 ip ospf cost 500 ip ospf network point-to-point ip ospf priority 0 ipv6 nd suppress-ra ! router ospf ospf router-id 172.30.0.222 passive-interface lo passive-interface lo:0 network 172.30.0.222/32 area 0.0.0.0 network 172.30.18.240/29 area 0.0.0.0 Fortunately, our ospfd instance isn't advertising the 0.0.0.0/1. Anybody have insights on what's causing this? We're running 0.98.6 that comes with CentOS 5. Thanks for any tips. Jiann-Ming Su 2008-10-28T00:57:49 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.network.quagga.user