gmane.network.quagga.user

[quagga-users 9885] Re: Help RIPNG can send but not receive

Larry Marquis — 2008-09-05T17:57:14

Greg, ip6tables was blocking the ripng msg receive. I am using centos5 Linux that is a redhat clone. The ripng msgs sarted flowing when I disabled the ip6tables service with "service ip6tables stop". Then I added this line "-A RH-Firewall-1-INPUT -p udp --dport 521 -j ACCEPT" to /etc/sysconfig/ip6tables then restarted ip6tables "service ip6tables start" Now all is well. Thanks for your help, Larry Greg Troxel public.gmane.org> Sent by: To quagga-users-boun Larry Marquis/Boca ces-UOy77sIEA+f4/q2+jIuWiA< at >public.gmane.org Raton/Contr/IBM< at >IBMUS net cc quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org Subject 09/05/2008 08:55 [quagga-users 9884] Re: Help RIPNG AM can send but not receive Larry Marquis public.gmane.org> writes: [formatting recovered] It sounds like multicast receive is broken, somewhere between quagga, your kernel, and the ethernet interface. If your test program doesn't work, I would suggest getting trying to get support from Centos, and only when that works reexamine quagga. I presume you are doing an IP_ADD_MEMBERSHIP setsockopt, and that you used netstat to look at the list of joined groups ('netstat -a' on BSD, don't know about Linux). On my NetBSD/sparc64 system, I can see ff02::9. You can also try running tcpdump on the same interface, to see if putting it in promiscuous mode works around bugs that may be present in the multicast filter programming code. (See attached file: attmy2md.dat) _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users _______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users

[quagga-users 9884] Re: Help RIPNG can send but not receive

Greg Troxel — 2008-09-05T12:55:23

_______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users

[quagga-users 9883] Is OSPF6D usable ?

Quagga User — 2008-09-04T17:27:11

I have Quagga ospf6d running on a simple ring of routers (area 0) for 3 sites connected with IPv6 tunnels and another 2 routers in one of the a adjacent areas. The ospf6d daemons seem to crash frequently, perhaps, so it appears sometimes, when a link goes down briefly. Does OSPF6d have known problems like this? Is it usable at all? -Thanks

[quagga-users 9882] Help RIPNG can send but not receive

Larry Marquis — 2008-09-04T15:47:28

Help ? I'm trying to use the RIPng component of the Quagga routing GPL software package. My problem is RIPng can send RIPng protocol messages but does not receive them.. I am monitoring the network with Wireshark and can see RIPng messages from other routers, and can see the RIPng messages broadcast by my RIPng node. In an effort to determine if the problem is my Linux configuration or Quagga RIPng I constructed a simple example program (based on the RIPng code) to read messages from the group address "ff02::9" on the RIPng port 521. The simple example program behaves like Quagga RIPng in that it does not receive the group multicast messages. So my assumption is that my Linux configuration is not correct for passing multicast group messages to my application (RIPng.) The sysctl value net.ipv6.conf.all.forwarding is 1. If there something else I need to do to enable receive of group multicast message ? I'm running Centos5 linux. Thanks, Larry

[quagga-users 9881] Is OSPF6D usable at all?

Quagga User — 2008-09-04T06:03:08

[quagga-users 9880] OSPFD gone wild ...

Joe Greco — 2008-09-03T22:24:43

We're another FreeBSD site having trouble with Quagga and FreeBSD 7, but we are also seeing some strangeness with legacy installs that used to work just fine. For example, this is a FreeBSD 6 box running Quagga 0.98.2. We've been seeing a strange failure mode where Quagga will "go insane". For ex: # sho ip osp OSPF Routing Process, Router ID: 206.55.65.149 Supports only single TOS (TOS0) routes This implementation conforms to RFC2328 RFC1583Compatibility flag is enabled OpaqueCapability flag is disabled SPF schedule delay 1 secs, Hold time between two SPFs 1 secs Refresh timer 10 secs Number of external LSA 0. Checksum Sum 0x00000000 Number of opaque AS LSA 0. Checksum Sum 0x00000000 Number of areas attached to this router: 1 Area ID: 0.0.1.0 Shortcutting mode: Default, S-bit consensus: ok Number of interfaces in this area: Total: 9, Active: 9 Number of fully adjacent neighbors in this area: 0 Area has simple password authentication Number of full virtual adjacencies going through this area: 0 SPF algorithm executed 2 times Number of LSA 1 Number of router LSA 1. Checksum Sum 0x0000b85b Number of network LSA 0. Checksum Sum 0x00000000 Number of summary LSA 0. Checksum Sum 0x00000000 Number of ASBR summary LSA 0. Checksum Sum 0x00000000 Number of NSSA LSA 0. Checksum Sum 0x00000000 Number of opaque link LSA 0. Checksum Sum 0x00000000 Number of opaque area LSA 0. Checksum Sum 0x00000000 # sho ip osp nei Neighbor ID Pri State Dead Time Address Interface RXmtL RqstL DBsmL # sho ip osp ro ============ OSPF network routing table ============ [...] N 206.55.68.160/27 [30020] area: 0.0.1.0 directly attached to fxp0 N 206.55.69.160/27 [30010] area: 0.0.1.0 directly attached to fxp1 ============ OSPF router routing table ============= ============ OSPF external routing table =========== # sho ip osp da OSPF Router with ID (206.55.65.149) Router Link States (Area 0.0.1.0) Link ID ADV Router Age Seq# CkSum Link count 206.55.65.149 206.55.65.149 1499 0x80000114 0xd697 9 # sho ip osp int fxp0 is up Internet Address 206.55.68.172/27, Broadcast 206.55.68.191, Area 0.0.1.0 Router ID 206.55.65.149, Network Type BROADCAST, Cost: 30020 Transmit Delay is 1 sec, State DR, Priority 1 Designated Router (ID) 206.55.65.149, Interface Address 206.55.68.172 No backup designated router on this network Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:10 Neighbor Count is 0, Adjacent neighbor count is 0 fxp1 is up Internet Address 206.55.69.172/27, Broadcast 206.55.69.191, Area 0.0.1.0 Router ID 206.55.65.149, Network Type BROADCAST, Cost: 30010 Transmit Delay is 1 sec, State DR, Priority 1 Designated Router (ID) 206.55.65.149, Interface Address 206.55.69.172 No backup designated router on this network Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:10 Neighbor Count is 0, Adjacent neighbor count is 0 Okay, which looks like neither of the attached networks are running OSPF. Yet one of the FreeBSD routers that's responsible for the 206.55.68.160/27 network: # sho ip osp nei Neighbor ID Pri State Dead Time Address Interface RXmtL RqstL DBsmL [...] 206.55.65.173 1 2-Way/DROther 00:00:39 206.55.68.162 em3:206.55.68.161 0 0 0 206.55.73.10 1 Full/Backup 00:00:38 206.55.68.164 em3:206.55.68.161 32 0 0 206.55.73.11 10 Full/DR 00:00:30 206.55.68.165 em3:206.55.68.161 26 0 0 206.55.73.12 1 2-Way/DROther 00:00:32 206.55.68.168 em3:206.55.68.161 0 0 0 206.55.65.148 1 2-Way/DROther 00:00:38 206.55.68.171 em3:206.55.68.161 0 0 0 206.55.65.149 1 Init/DROther 00:00:34 206.55.68.172 em3:206.55.68.161 0 0 0 206.55.64.22 1 2-Way/DROther 00:00:36 206.55.68.189 em3:206.55.68.161 0 0 0 [...] # sho ip osp int em3 em3 is up Internet Address 206.55.68.161/27, Broadcast 206.55.68.191, Area 0.0.1.0 Router ID 206.55.64.21, Network Type BROADCAST, Cost: 1 Transmit Delay is 1 sec, State DROther, Priority 1 Designated Router (ID) 206.55.73.11, Interface Address 206.55.68.165 Backup Designated Router (ID) 206.55.73.10, Interface Address 206.55.68.164 Timer intervals configured, Hello 10, Dead 40, Wait 40, Retransmit 5 Hello due in 00:00:01 Neighbor Count is 7, Adjacent neighbor count is 2 Now, notice the "Init/DROther" ... So, if I turn on packet tracing on the afflicted box, I see that there is no packet reception. 2008/09/03 17:07:17 debugging: OSPF: Hello sent to [224.0.0.5] via [fxp0:206.55.68.172]. 2008/09/03 17:07:17 debugging: OSPF: Hello sent to [224.0.0.5] via [fxp1:206.55.69.172]. 2008/09/03 17:07:27 debugging: OSPF: Hello sent to [224.0.0.5] via [fxp0:206.55.68.172]. 2008/09/03 17:07:27 debugging: OSPF: Hello sent to [224.0.0.5] via [fxp1:206.55.69.172]. But they're there on the wire, visible via tcpdump. 17:21:02.701101 IP 206.55.68.165 > 224.0.0.5: OSPFv2, Hello, length: 72 17:21:02.901490 IP 206.55.68.161 > 224.0.0.6: OSPFv2, LS-Update, length: 1480 17:21:02.901504 IP 206.55.68.161 > 224.0.0.6: ospf 17:21:02.901511 IP 206.55.68.161 > 224.0.0.6: OSPFv2, LS-Update, length: 56 17:21:02.902419 IP 206.55.68.164 > 224.0.0.5: OSPFv2, LS-Update, length: 812 17:21:02.907621 IP 206.55.68.171 > 224.0.0.6: OSPFv2, LS-Update, length: 812 17:21:02.915266 IP 206.55.68.189 > 224.0.0.6: OSPFv2, LS-Update, length: 812 17:21:02.989948 IP 206.55.68.162 > 224.0.0.6: OSPFv2, LS-Ack, length: 584 So what should I be looking at here? We're seeing this on a variety of FreeBSD and Quagga versions, intermittently. ... JG

[quagga-users 9879] Re: question mark "?" inside as_path

Steve Bertrand — 2008-09-03T16:14:57

CTRL-V works for me (forgive the wrap): router.eagle.ca(config)# ip as-path access-list 150 permit ^1234(_.*)?(1111|2222)$ router.eagle.ca# sh ip as-path-access-list AS path access list 150 permit ^1234(_.*)?(1111|2222)$

[quagga-users 9878] question mark "?" inside as_path

Mikhail A. Grishin — 2008-09-03T15:34:28

Good day. I am unable to enter the question mark "?" from bgpd vty (as element of the configuration inside regexp) For example, from startup config, bgpd successfully accepts: ip as-path access-list 159 permit ^1234(_.*)?_(1111|2222)$ but if I enter this from vty, the string looks like: ip as-path access-list 159 permit ^1234(_.*)_(1111|2222)$ For Cisco devices, there is a solution: ======= To type a question mark in a regular expression on the router, first press Ctrl-V (Escape for CLI), and then you can enter ?. ======= Is there any solution for Quagga?

[quagga-users 9877] Route server MD5 performance tests

Valentin Rosic — 2008-09-03T08:56:15

Hello, Did anyone do some performance tests with Quagga as route server with MD5 support? I'm interested in speed of route propagations. regards, Valentin

[quagga-users 9876] Re: bgpd stops responding when zebra isbusy

Oliver Hookins — 2008-09-03T05:40:23

I'd still like any suggestions for how to avoid bgpd locking up when it tells zebra to withdraw and add a large number of routes. As another list member pointed out, the commands "clear ip bgp X soft (in|out)" and "clear ip bgp X (in|out)" are the same command (as I've shown from the code snippets below). Is the behaviour I'm seeing in bgpd a bug or just a known limitation? ... ... ... ...

[quagga-users 9875] Re:Static route metric always '0' in Kernel routetable

Daniel Ng — 2008-09-03T01:26:49

configured or Well, it looks like I was confusing the concepts of 'Administrative Distance' and 'Metric'. The following sets up a static route with an Administrative Distance of 24 in zebra: ip route 1.2.3.4/32 24 So now my question is, why isn't there a way to set the initial Metric of a specific static route? It can be done under Linux (using "route add...."), why not in Quagga?

[quagga-users 9874] Re: RIP routing

nanda — 2008-09-02T13:26:59

_______________________________________________ Quagga-users mailing list Quagga-users-UOy77sIEA+cAd7ICUelF/Q< at >public.gmane.org http://lists.quagga.net/mailman/listinfo/quagga-users

[quagga-users 9873] RIP routing

Monali Bhattacharya — 2008-09-02T11:44:52

hi, i'm not able to do rip using the quagga. when i'm doing show ip rip it is showing the connected ones but not the RIP. if i'm trying to do write file, it is showing " Can't open configuration file /usr/local/etc/zebra.conf.pl678.Pls help. Larsen & Toubro Limited www.larsentoubro.com This Email may contain confidential or privileged information for the intended recipient (s) If you are not the intended recipient, please do not use or disseminate the information, notify the sender and delete it from your system.

[quagga-users 9872] ospf6

Tomasz Grobelny — 2008-09-01T20:41:59

My config on 4 routers looks like this: # cat ospf6d.conf hostname password ! interface eth1 ipv6 ospf6 cost 1 ipv6 ospf6 hello-interval 1 ipv6 ospf6 dead-interval 40 ipv6 ospf6 retransmit-interval 5 ipv6 ospf6 priority 1 ipv6 ospf6 transmit-delay 1 ipv6 ospf6 instance-id 0 ! interface vlan0 ipv6 ospf6 cost 1 ipv6 ospf6 hello-interval 1 ipv6 ospf6 dead-interval 40 ipv6 ospf6 retransmit-interval 5 ipv6 ospf6 priority 1 ipv6 ospf6 transmit-delay 1 ipv6 ospf6 instance-id 0 ! router ospf6 redistribute connected interface vlan0 area 0.0.1.0 interface eth1 area 0.0.1.0 ! line vty ! I would expect that paths are redistributed to all those 4 machines but unfortunately this is not the case. How can I debug the problem? Do I need to add anything to my config? OSPFv2 on the same network works just fine. Thanks in advance,

[quagga-users 9871] Re: Building quagga-0.99.10

2008-09-02T09:06:31

I have installed 'grep' from www.sunfreeware.com. The description from site: # grep-2.5.1a-sol9-sparc-local.gz The GNU version of grep which finds lines that match patterns - installs in /usr/local. Dependencies: libiconv, pcre, and /usr/local/lib/libgcc_s.so.1 needs to exist by installing libgcc-3.4.6 or gcc-3.4.6. SMCgrep just the name of package, it doesn't means source, only wish of packager (Steve Christensen), as I understand. Take package from Solaris 10 is a wrong way, I think. 02.09.08, 12:28, "Nigel Marett" public.gmane.org>: 01.09.08, 17:59, "Nigel Marett" public.gmane.org>:

[quagga-users 9870] Re: Building quagga-0.99.10

Nigel Marett — 2008-09-02T08:28:17

Quagga-0.99.10 configures for me on Solaris, but then I'm using a different grep.. ;) Quagga configuration -------------------- quagga version : 0.99.10 host operationg system : solaris2.10 source code location : . compiler : gcc compiler flags : -Os -fno-omit-frame-pointer -g -std=gnu99 -Wall -Wsign-compare -Wpointer-arith -Wbad-function-cast -Wwrite-strings -Wmissing-prototypes -Wmissing-declarations -Wchar-subscripts -Wcast-qual make : make includes : linker flags : -lresolv -lcrypt -lumem -lnsl -lsocket state file directory : /var/run config file directory : /usr/local/etc example directory : /usr/local/etc user to run as : quagga group to run as : quagga group for vty sockets : config file mask : 0600 log file mask : 0600 Cheers, Nigel. 2008/9/1 public.gmane.org>:

[quagga-users 9869] RIP default-info originate broken?

Daniel Ng — 2008-09-02T08:01:47

Hi, Using 0.99.10, it seems 'default-information originate' in ripd behaves the same way as 'default-information originate always' in ospfd ie.- Regardless of whether a 0.0.0.0/0 static route has been configured in the node, the node will still advertise itself as the Default Gateway. Shouldn't 'default-information orginate' behave the same way in both ospfd and ripd? ie.- If (and only if) a 0.0.0.0/0 static route has been configured in the node, the node will advertise itself as the Default Gateway. If so, can someone suggest a way to fix ripd so that we can achieve this? TIA, Daniel

[quagga-users 9868] Re: hi,

Sargun Dhillon — 2008-09-02T06:16:31

Did you turn of SELinux? "setenforce 0" Did you set the permissions properly? On Mon, Sep 1, 2008 at 10:54 PM, Monali Bhattacharya public.gmane.org> wrote:

[quagga-users 9867] hi,

Monali Bhattacharya — 2008-09-02T05:54:38

hi, i'm not able to run the rip or ospf routing protocol with the quagga. if i'm trying to do write file it is showing the error "Can't open configuration file /usr/local/etc/ripd.conf.PlkF22 . I also want to know how can i know what is the log location. Pls help Larsen & Toubro Limited www.larsentoubro.com This Email may contain confidential or privileged information for the intended recipient (s) If you are not the intended recipient, please do not use or disseminate the information, notify the sender and delete it from your system.

[quagga-users 9866] Re: Static route metric always '0' in Kernelroute table

Daniel Ng — 2008-09-02T04:54:21

configured or I've noticed this bug in 0.99.10 as well. Has anyone got any clues about how to fix this?

[quagga-users 9865] Re: Best hardware for maximum throughput andPPSperformance

Charles Wyble — 2008-09-01T18:43:58

See http://www.liberouter.org/ and http://www.openpattern.org/ for more on open source FPGA routers. Oh yeah. Many orders of magnitude more.