Srinivas K | 18 Dec 13:29 2014

[quagga-users 13859] OSPF floods Type-4 MaxAge LSAs in to stub area

R1 -----area 0-----Quagga------area 1-----R3 (Cisco)

1. Initially area 1 is configured as normal area. Once Adjacencies are formed Quagga generated Type 4 ASBR
summary LSA and flooded into R3
2. After that I have configured area 1  as stub in Cisco and Quagga respectively.
3. Quagga sending MaxAge Type 4 ASBR summry LSA  to Cisco. But cisco rejects to access to displays below
error message.

*Dec 18 17:49:00.463: %OSPF-4-BADLSATYPE: Invalid lsa: Bad LSA type Type 4, Length 28, LSID
from,, FastEthernet0/0 

This issue is not solved even after bringing down neighbour at Quagga. (by shut-downing interface at Cisco)

Quagga vs Cisco differences :

1. Once configured as normal area as stub area Cisco brings down adjacency. But not by Quagga
2. Quagga acknowledge MaxAge Type 4 LSA if Cisco router is replaced by Quagga.


[quagga-users 13857] Re: The route between quagga servers in full mesh is not optimal

В Срд, 17/12/2014 в 14:24 -0500, Steve Clark пишет:
> On 12/17/2014 01:01 PM, Покотиленко Костик wrote:
> > Thanks for reply and for understanding this mess
> > 
> > В Срд, 17/12/2014 в 12:06 +0800, Feng Lu пишет:
> > > Hi,
> > > 
> > > I think the quagga behavior is normal in both of your two cases.
> > > 
> > > The remote interface addresses are usually meaningless and useless
> > > to the calculating router in the OSPF AS.
> > > 
> > > So I don't understand why Loc1 needs to know the path to
> > > and If it needs to reach Loc0, it knows and
> > > If it needs to reach Loc2, it knows and
> > > The addresses on the links between Loc0 & Loc2 are
> > > meaningless to Loc1.
> > Not quite so. I can omit using 10.253. addresses when making
> > connections, but, example:
> > 
> > - during the time links between Loc0 and Loc1 were down
> > - Loc0 (router itself) makes tcp connection to Loc1 using "good" address
> >
> > - Since the only way for Loc0 to reach Loc1 is now through Loc2 the
> > source address for packets would be either or
> > - links between Loc0 and Loc1 now restored
> > - packets in direction Loc0->Loc1 ( -> now use
> > optimal route since there are no problems routing 10.1.
> > - packets in direction Loc1->loc0 ( -> will now go
(Continue reading)


[quagga-users 13853] The route between quagga servers in full mesh is not optimal


We have several physical locations we'd love to interconnect with HA
VPN. I'm now testing things.

The idea is to connect each site with each other by 2 openvpn links, one
incomming (server) and one outgoing (client) and use quagga ospf.
Starting with 3 locations.

For routing to work properly regarding path cost we need to use one
openvpn instance for each link, it should look like "p2p", one client
for each incoming instance.

It comes that there are 2 ways to do this with openvpn:

1. Use TUN device and setup real P2P, topology p2p, 2 addresses /32 for
each link

          |      |
                    +-------< | Loc1 | >-------+
                   /    +---> |      | <---+    \
                  /    /       ------       \    \
                 /    /\    \
                /    /                        \    \ /    / \    \
              ------                            ------
             |      | |      |
(Continue reading)

Daniel Paez Sanchez | 11 Dec 15:55 2014

[quagga-users 13848] Fwd: Problem with Quagga

---------- Forwarded message ----------
From: Daniel Paez Sanchez <dapaez360 <at>>
Date: 2014-12-11 9:54 GMT-05:00
Subject: Fwd: Problem with Quagga
To: security-Xce1itCos/

---------- Forwarded message ----------
From: Daniel Paez Sanchez <dapaez360 <at>>
Date: 2014-12-11 9:53 GMT-05:00
Subject: Problem with Quagga
To: maintainers-Xce1itCos/

I am doing a project where I am using Quagga last version (, and have told me that this version has some fixes included that now I can use IS-IS.

My project is about comparing IS-IS and OSPF for IPv6, in security area, so before this version I could only test OSPFv3

The thing is when I try to install quagga (I follow the steps: configure, make, make install). When I type : CONFIGURE, there is an error:

configure: error: GNU awk is required for lib/memtype.h made by memtypes.awk.
BSD awk complains: awk: gensub doesn't support backreferences (subst "\1")

Anyone knows how can I fix this error and install quagga??

Daniel Páez

Daniel Páez
Tecnología de Información
Área Desarrollo.

Av. José Santiago Castillo y Juan Tanca Marengo Km. 1.8  
Telf.: + (593 4) 2682050 - 2682230 Ext: 1186
Fax.: 2682065 - 2682242  
Guayaquil – Ecuador

Daniel Páez
Tecnología de Información
Área Desarrollo.

Av. José Santiago Castillo y Juan Tanca Marengo Km. 1.8  
Telf.: + (593 4) 2682050 - 2682230 Ext: 1186
Fax.: 2682065 - 2682242  
Guayaquil – Ecuador

Quagga-users mailing list
Viktor Khomyuk | 10 Dec 05:35 2014

[quagga-users 13846] OSPF sometimes does not send hello packets

Hello! I have strange phenomenon - sometimes quagga does not send hello 

tcpdump from local mashine:
09:19:05.973296 IP > OSPFv2, Hello, length 52
09:19:06.974023 IP > OSPFv2, Hello, length 52
09:19:07.974384 IP > OSPFv2, Hello, length 52
                                                             <- no packets
09:19:11.865022 IP > OSPFv2, Hello, length 52

There are no messages in dmesg or syslog before 09:19:11. CPU idle=99%

We are using ubuntu 10 (kernel 3.14.23) and quagga

How I can find the reason?
Wei Wen | 5 Dec 14:26 2014

[quagga-users 13843] Quagga does not work properly


I have Quagga installed following an easy instruction from the website:
Below is the detail of the steps:

1. Install source
#apt-get install quagga quagga-doc

2. Edit file daemons
#mcedit /etc/quagga/daemons

3. Change status no / yes on suitable routing protocol
zebra=yes //Compulsory “yes”

bgp=no //Border gateway protocol
ospf=no //OSPF with IPV4
ospf6d=no //OSPF with IPv6
ripd=yes //RIP versi 2
ripngd=no //RIP with IPV6

4. Make file configuration zebra.conf and ripd.conf, by copy from sample file configuration
#cp /usr/share/doc/quagga/examples/zebra.conf.sample /etc/quagga/zebra.conf
#cp /usr/share/doc/quagga/examples/ospfd.conf.sample /etc/quagga/ospfd.conf

5. Set password [optional]
#echo “password password” > /etc/quagga/zebra.conf

6. Change ownership user and group from files configuration in /etc/quagga, and change access permission
#chown quagga.quaggavty /etc/quagga/*.conf
#chmod 640 /etc/quagga/*.conf

7. Start daemons
#/etc/init.d/quagga start

8. Connect quagga
#telnet localhost zebra

The problem is in step 7. After I typed it in, all I could see are "y" on the first column of my terminal. And the "y" continues to stream in, so I could see the blinking "y" at the left bottom. I continued with step 8, but that didn't work. 

Since this incidence, I couldn't turn off my computer properly. I tried to remove quagga with command line (apt-get remove) and it didn't work. I tried to remove it using software update center, and that didn't work either. Could someone point out what was wrong with my installation and how can I remove quagga? Thank you very much.


Wei Wen

Quagga-users mailing list
Joe Moog | 4 Dec 23:12 2014

[quagga-users 13842] BGP multihome setup not failing to second provider

I’m hoping somebody can give me a push in the right direction. Using quagga in a BPG multihome setup with
two internet providers, both providing full routes. The config (-type cisco) is quite simplistic,
pasted here (slightly redacted):

router bgp 6XXXX
 no synchronization
 bgp router-id
 network mask
 neighbor a.b.c.d remote-as XXX
 neighbor a.b.c.d description ISP1
 neighbor a.b.c.d update-source vlan11
 neighbor a.b.c.d send-community both
 neighbor a.b.c.d prefix-list local out
 neighbor w.x.y.z remote-as YYY
 neighbor w.x.y.z description ISP2
 neighbor w.x.y.z update-source vlan12
 neighbor w.x.y.z send-community both
 neighbor w.x.y.z prefix-list local out
 no auto-summary
ip prefix-list local seq 10 permit

With one or the other provider configured alone, inbound traffic to the hosts succeeds. However, in the
presence of both providers, routing seems to want to attach itself to ISP2 (preferred as the closest,
since they actually have equipment in our building) and will not re-route itself to ISP1 if ISP2 goes away,
causing attempted connections to our hosts to fail. This was tested by placing ISP2 in shutdown state
further upstream where it enters our network, simulating a carrier failure. Output of “sh ip bgp” and
“sh ip route” shows that the second provider has taken over full population of the tables, but actual
routing just doesn’t follow. (To date, I’ve only been testing inbound connectivity.) Naturally,
shutting ISP1 down in a similar fashion has no effect since BGP has preferred ISP2 for our inbound connectivity.

On the quagga host (FreeBSD), there are a few specific routes for RFC1918 networks that point at our core.
(Not including the network we advertise to BGP, that address detail was substituted for the purpose of
this exchange.) The quagga host has no default gateway configured, which as I understand it is
appropriate for what I am trying to accomplish. If this is in error, I’d love to know for certain.

If anybody has a suggestion or recommendation, I welcome the feedback. I have tried a lot of different
settings all with similar outcomes. I’m hoping that by putting other sets of more experienced eyes on
this that anything I might have missed will be quickly and easily identified. Let me know if any additional
information would be useful, and I will provide what I can.

Thanks in advance.


Quagga-users mailing list
Quagga-users <at>
Potter, Timothy BIS | 3 Dec 22:47 2014

[quagga-users 13834] OSPF6D stops replying to Hello messages

I am attempting to do some proof of concept work with OSPF6D and zebra version  I have two PCs hooked up to each other.  Each with an IPV6 address with a /128 mask so they are each on their own network.  The ospf6d.conf file looks like this:


hostname ospf6d

password zebra

log stdout

log file /etc/quagga/ospf6d.log

debug ospf6 message all


!enable password please-set-at-here


interface eth0


interface eth1


router ospf6


redistribute kernel

redistribute connected

redistribute ripng

redistribute static

interface eth0 area

interface eth1 area


line vty



The other pc has a router-id of


The zebra.conf file looks like this:

! -*- zebra -*-


! zebra sample configuration file


! $Id: zebra.conf.sample,v 1.1 2002/12/13 20:15:30 paul Exp $


hostname Router

password zebra

enable password zebra


! Interface's description.

interface eth0


no ipv6 nd suppress-ra


interface eth1


no ipv6 nd suppress-ra


!interface lo

! description test of desc.


!interface sit0

! multicast               


! Static default route sample.


!ip route


log file /home/root/zebra.log    


line vty                         




When I boot the machines everything comes up fine and I can do a ping6 from one machine to the other with no issues.  But after about 2 minutes one of my machines stops acknowledging the hello messages from the other machine.  Eventually it times out and purges all the routes in its routing table.  The ping6 then stops working.


The machine that has this problem is an am335x-evm board from TI.  I am using their SDK 7 which provides Linux version 3.12.  When I did this using their SDK 6 which was Linux version 3.2 I did not have this problem.  The routing tables were built just fine and stayed that way.


Any thoughts on what may be wrong?






Quagga-users mailing list
OCEANET - Cédric BASSAGET | 2 Dec 16:18 2014

[quagga-users 13831] 32bit ASn in community ?


When doing a route-map which sets a community, my 32bit ASn is converted 
to "924"
Just want to know if this is a normal behavior :

wanguard(config)# route-map xxx permit 10
wanguard(config-route-map)# set community 179532:666

wanguard# sh route-map xxx
route-map xxx, permit, sequence 10
   Match clauses:
   Set clauses:
     community 924:666
   Call clause:
     Exit routemap


Quagga-users mailing list
Quagga-users <at>
Zhang, Wei X | 2 Dec 08:00 2014

[quagga-users 13830] Conneting a Sub network to a bigger network via Ipv6 using ripngd

Hi all,

I am trying to connect a sub network to my PC’s by using ripngd.

Let me explain myself by showing below.



My Sub network ( prefixed aaaa::/64 )------( connect to )----- PC( With quagga ) ----( connect to )----- A normal ipv6 network( prefixed bbbb::/64 )


I was try to set ripngd up and broadcast the PC as gateway to aaaa::/64,

And then other PCs can connect aaaa::1 normally.


But I was fail to make it work, Is there a tutorial to make it work?


I am new to Ipv6 and quagga, sorry for the disturb, and thanks for helping.


Best Regards.



Quagga-users mailing list
Srinivas K | 27 Nov 14:59 2014

[quagga-users 13829] Conditional "default-originate" in BGP

At present Quagga  BGP's conditional default-originate route-map matches route-map against BGP routing table But not zebra routing table (unlike Cisco).
Is there any patch available for this?. Same for OSPF also

Quagga-users mailing list