From bms at incunabulum.net Fri May 1 08:46:30 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 01 May 2009 16:46:30 +0100 Subject: [Xorp-users] Help ! In-Reply-To: References: Message-ID: <49FB1956.7060308@incunabulum.net> Hi there, I'm assuming you're working with the XORP 1.6 release sources, and are trying to configure RIPv2 on two interfaces. JESUS RAMIREZ wrote: > > Hello eveybody ... > Good Morning, I'm rookie in this ,but I think it's very interesting > and I want to learn, so I have a lot a questions for you. > > If I want to configure 2 network cards ( Ethernet ) were can i > found the configuration ? I read the manual and i dont konw how can i > get into the code? or do i need to download from your web and then > paste it, because i need to configure de rip too....if i need to > download any file were can i found ? and how can i paste it ? > > I don't know if any of you guys can Help me..... please : ( > Probably the best thing for you to do is to look at this file (asuming you have downloaded xorp-1.6.tar.gz and untarred it) :- cd xorp-1.6 cd rtrmgr/config and then view the file "rip.boot" in your favorite text editor. The configuration for RIP in that file contains two interfaces dc0 and dc1. You will probably want to edit this for use on whatever operating system you're trying to run XORP on. In the case of Linux, you can just use search and replace to change dc0 to eth0, dc1 to eth1, and then fill out your IP addresses. After that, configuring RIP should just be a matter of following the instructions in the User Manual to give RIP the settings you need for your network. thanks, BMS From bms at incunabulum.net Fri May 1 08:53:53 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 01 May 2009 16:53:53 +0100 Subject: [Xorp-users] OSPF4 neighbor shown in only one machine not on the other two In-Reply-To: <7fafac67bc7482e3827f3a7171fc1bf8.squirrel@gpo.iitb.ac.in> References: <7fafac67bc7482e3827f3a7171fc1bf8.squirrel@gpo.iitb.ac.in> Message-ID: <49FB1B11.4020508@incunabulum.net> Hi, Your configuration file looks OK, which suggests the problem might be with the VLAN driver support in Linux, or how XORP is trying to use it. It would be helpful to see a packet capture on the affected machines, I believe there are no outstanding bugs for the VLAN support and it has been tested as working before on generic Linux. It is possible there's a difference between your SUSE machines and your Fedora machines in that behaviour. Can you be sure they are patched to the correct switch ports in your network? You don't mention if the machines successfully synchronized their OSPF databases. Please send the output of the 'show ospf4 neighbor' commands when reporting problems with OSPF such as this, as we need to know what states they're in. You can use Wireshark (www.wireshark.org) or tcpdump (www.tcpdump.org), both should be available in Fedora as yum packages. Be sure to time what's happening, and try to preserve all XORP logs from your session. Please don't send the traces/log files here or to the list -- instead, attach them to the bug report which Syed has asked you to open. bhavin81 at iitb.ac.in wrote: > Hello Friends, > > I have tried to configure ospf4 in xorp. I have configured > 3 machines for this purpose. When i use the command show ospf4 > neighbor on one machine(fedora 8) out of three then it shows neighbor > correclty but when i use the same command on other two machines (suse > 10.3) then it does not show neighbors on the those machines. thanks, BMS From bms at incunabulum.net Fri May 1 09:06:20 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 01 May 2009 17:06:20 +0100 Subject: [Xorp-users] Appeal for suggestions re: automated support dump tool Message-ID: <49FB1DFC.8080902@incunabulum.net> More often than not, technical questions get asked here about XORP operation or use, which would probably benefit from a general debugging information tool -- as a lot of the information which gets asked for, when problems are raised, is repetitive and common to all bug reports. Rather than request this information every time, which makes it more difficult to give support, it might be better if we could provide functionality in an automated way. What we'd really appreciate from folks is suggestions for such a tool. The basics would be: OS version, OS network interface config, PCI info, followed of course by the XORP config dumps for each routing process. Such a tool could get rolled into the Router Manager. However, such a tool is probably best written in a modern, common scripting language. Python is particularly good for tasks lie this. Please share your suggestions and feedback. thanks, BMS From bms at incunabulum.net Fri May 1 09:17:53 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 01 May 2009 17:17:53 +0100 Subject: [Xorp-users] virtual link In-Reply-To: <7d5b33e3.e29f54.49f95da8.93a59@o2.pl> References: <7d5b33e3.e29f54.49f95da8.93a59@o2.pl> Message-ID: <49FB20B1.1040002@incunabulum.net> Tomek wrote: > Hi, > > I've got a problem with xorp configuration. > Actually in my case situation looks like this: > > R3 <-> R5 <-> R4 <-> R6 > > where: > > R3 is an ABR of area 0 (left side) and area 1 (right side) > R5 is internal of area 1 > R4 is ABR of area 1(left side) and area 3 (right side) > R6 internal of area 3 > > everything works fine until i want to configure virtual link between R3 and R4 > I added virtual-link section to area 0.0.0.0 in ospf4 section > ... * Which version of XORP are you using? * Which platform are you using XORP on? > area 0.0.0.0 { > virtual-link 192.168.0.40 { > transit-area: 0.0.0.1 > } > .... > and tried to add virtual-link to R4 without succeed because of error: > Command failed Virtual link must be in area 0.0.0.0 > > my question is how to configure virtual link at R4 ? > This could be a problem in vlinks traversing area borders, or more likely, a problem with the lookup by router ID. I'd have to check this code more fully before I could respond in detail about that. The configuration steps you posted look correct, assuming that you are configuring XORP up on R3.which is an ABR. But is area 1 a transit area or a stub area? An NSSA area is *not* valid for virtual links, they can only be configured across transit areas. More general questions: Does R3 learn about R4 without the virtual-link config? If so, which area(s) does R3 learn that R4 is in? Does R5 see the same topology (neighbours) as R3 and R4? thanks, BMS From bms at incunabulum.net Fri May 1 09:27:46 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 01 May 2009 17:27:46 +0100 Subject: [Xorp-users] not getting an alias In-Reply-To: <200904302145.28680.irving.lp@feestudiantes.cujae.edu.cu> References: <200904302145.28680.irving.lp@feestudiantes.cujae.edu.cu> Message-ID: <49FB2302.3040007@incunabulum.net> Hi, Can you post your full configuration file, and tell us which kernel version are you using, so that others may reproduce? Irving Leonard wrote: > I am so newbe in xorp but I read (breafly) the manual and it says that every > vif can take a lot of addresses but for me it doesn't make an alias. I have a > vif and give it 2 addresses in the same subnet, because I need to have 2 IPs > but it only get's the first. I'm using ubuntu hardy 8.04.2 > regards > irving thanks, BMS From irving.lp at feestudiantes.cujae.edu.cu Fri May 1 16:44:24 2009 From: irving.lp at feestudiantes.cujae.edu.cu (Irving Leonard) Date: Fri, 1 May 2009 19:44:24 -0400 Subject: [Xorp-users] not getting an alias In-Reply-To: <49FB2302.3040007@incunabulum.net> References: <200904302145.28680.irving.lp@feestudiantes.cujae.edu.cu> <49FB2302.3040007@incunabulum.net> Message-ID: <200905011944.25100.irving.lp@feestudiantes.cujae.edu.cu> On Viernes 01 Mayo 2009 12:27:46 PM Bruce Simpson escribi?: > Hi, > > Can you post your full configuration file, and tell us which kernel > version are you using, so that others may reproduce? I use 4 fast ethernet NICs and combined in a bond wich name is bond0; the network config file (/etc/network/interfaces) creates several vlans (including bond0.601) but only the devices (not configured with ips or up). The commented section down is the configuration of the bond (not vlan) interface that is in the /etc/network/interfaces. #uname -a Linux poseidon 2.6.24-23-server #1 SMP Thu Feb 5 15:34:58 UTC 2009 i686 GNU/Linux #my example config: interfaces { restore-original-config-on-shutdown: true # interface bond0 { # description: "El Bond" # disable: false # vif bond0 { # disable: false # address 127.1.0.1 { # prefix-length: 16 # disable: false # } # } # } interface bond0.604 { description: "Administraci?n Red" disable: false vif bond0 { disable: false address 10.8.177.129 { prefix-length: 25 disable: false } address 10.8.177.130 { prefix-length: 25 disable: false } } } } regards irving p.d.: I think is a better idea just to reply to the list. VI Conferencia Internacional de Energ?a Renovable, Ahorro de Energ?a y Educaci?n Energ?tica 9 - 12 de Junio 2009, Palacio de las Convenciones ...Por una cultura energ?tica sustentable www.ciercuba.com From bhavin81 at iitb.ac.in Sun May 3 02:18:39 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Sun, 3 May 2009 14:48:39 +0530 (IST) Subject: [Xorp-users] querry: Why Dropping packet from foreign network 10.10.20.0/24? Message-ID: <1a86605a2654d1f3cc85924d5bb52892.squirrel@gpo.iitb.ac.in> Hello friends, I have installed xorp on two suse 10.3 machines. Both the machines are connected to same switch. I have configured one machine with the following interface address: 10.10.10.10 and other machine with the adddress 10.10.10.20. Now i am getting the following message when i run ./xorp_rtrmgr on machine 1: Dropping packet from foreign network 10.10.20.0/24 and the following message on machine2: Dropping packet from foreign network 10.10.10.0/24 My configuration file is as below on machine1: interfaces { interface eth0 { description: "Ethernet Interface" disable: false vif eth0 { disable: false address 10.10.10.10 { prefix-length: 24 broadcast: 10.10.10.255 disable: false } } vif vlan1{ vlan{ vlan-id:1 } address 10.10.10.20{ prefix-length:24 broadcast: 10.10.10.255 disable: false } } } } fea { unicast-forwarding4 { disable: false } } protocols { ospf4 { router-id: 10.10.10.10 area 0.0.0.0 { interface eth0 { vif eth0{ address 10.10.10.10 { } } } } } } The following configuration file on machine2: interfaces { interface eth0 { description: "Ethernet Interface" disable: false vif eth0 { disable: false address 10.10.20.10 { prefix-length: 24 broadcast: 10.10.20.255 disable: false } } } } fea { unicast-forwarding4 { disable: false } } protocols { ospf4 { router-id: 10.10.20.10 area 0.0.0.0 { interface eth0 { vif eth0{ address 10.10.20.10 { } } } } } } Thanks in advance for your valuable help. Regards, Bhavin, Comm. Student. From chuyram28 at msn.com Sun May 3 09:02:44 2009 From: chuyram28 at msn.com (JESUS RAMIREZ) Date: Sun, 3 May 2009 10:02:44 -0600 Subject: [Xorp-users] Xorp-users Digest, Vol 38, Issue 1 In-Reply-To: References: Message-ID: Hello bruce, thank you for your help.... Ok, let me explain to you with more details.. I have installed Red Hat OS (linux)this is my router , and i tried to connect my router(xorp) with 2 computers (windows), in my 2 computers i assigned satatic addresses and i download and modify the archive that you tell me : 1st computer address 10.10.10.70 2nd computer address 10.10.10.90 for "cd rtrmgr/config" ? means that i need to burn it ?? or download and modify too. as soon as i modify the file , i do it in notepad , how do i need to save it ( rip.boot.txt ? ) and when i need to loaded how can i load de file ? or do i need to open in text editor into red hot? or do i need a software to open/edit flies w/ext .boot ? xorpsh configure edit load rip.boot ? ( shows "failed to open config file : rip.boot ) I dont'n know where i need to save it, and doesn't matter what extension is ( txt , doc ) ( in my computer shows discard0 instead dc0 ) RIP interfaces { interface eth0 { vif eth0 { address 10.10.10.70 { prefix-length: 24 } } } interface eth1 { vif eth1 { address 10.10.10.90 { prefix-length: 24 } } } } fea { unicast-forwarding4 { disable: false } } policy { policy-statement export-connected { term 100 { from { protocol: "connected" } } } } protocols { rip { /* Connected interfaces are advertised if explicitly exported */ export: "export-connected" interface eth0 { vif eth0 { address 10.10.10.70 { disable: false /* authentication { simple-password: "password"; md5 @: u32 { <------------- It is necessary this ? password: "password"; start-time: "2006-01-01.12:00" end-time: "2007-01-01.12:00" } } */ } } } interface eth1 { vif eth1 { address 10.10.10.90 { disable: false } } } } } > From: xorp-users-request at xorp.org > Subject: Xorp-users Digest, Vol 38, Issue 1 > To: xorp-users at xorp.org > Date: Fri, 1 May 2009 12:00:05 -0700 > > Send Xorp-users mailing list submissions to > xorp-users at xorp.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > or, via email, send a message with subject or body 'help' to > xorp-users-request at xorp.org > > You can reach the person managing the list at > xorp-users-owner at xorp.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Xorp-users digest..." > > > Today's Topics: > > 1. Re: virtual link (Nithya Chandrakesan) > 2. not getting an alias (Irving Leonard) > 3. Re: Help ! (Bruce Simpson) > 4. Re: OSPF4 neighbor shown in only one machine not on the other > two (Bruce Simpson) > 5. Appeal for suggestions re: automated support dump tool > (Bruce Simpson) > 6. Re: virtual link (Bruce Simpson) > 7. Re: not getting an alias (Bruce Simpson) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 30 Apr 2009 12:01:32 -0700 > From: Nithya Chandrakesan > Subject: Re: [Xorp-users] virtual link > To: Tomek > Cc: xorp-users at xorp.org > Message-ID: > <1dfe63a80904301201n73c25767m653fe0f90ddc5935 at mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi Tom, > > Could you provide us your config? Also let us know which version of XORP you > were trying. > > Thanks, > Nithya. > > On Thu, Apr 30, 2009 at 1:13 AM, Tomek wrote: > > > Hi, > > > > I've got a problem with xorp configuration. > > Actually in my case situation looks like this: > > > > R3 <-> R5 <-> R4 <-> R6 > > > > where: > > > > R3 is an ABR of area 0 (left side) and area 1 (right side) > > R5 is internal of area 1 > > R4 is ABR of area 1(left side) and area 3 (right side) > > R6 internal of area 3 > > > > everything works fine until i want to configure virtual link between R3 and > > R4 > > I added virtual-link section to area 0.0.0.0 in ospf4 section > > > > area 0.0.0.0 { > > virtual-link 192.168.0.40 { > > transit-area: 0.0.0.1 > > } > > .... > > and tried to add virtual-link to R4 without succeed because of error: > > Command failed Virtual link must be in area 0.0.0.0 > > > > my question is how to configure virtual link at R4 ? > > > > R3 reports something like this > > > > root at r3> show ospf4 neighbor > > Address Interface State ID Pri Dead > > 192.168.6.50 eth3/eth3 Full 192.168.0.50 128 > > 33 > > 192.168.5.40 vlink/192.168.0.40 Init 192.168.0.40 0 > > 0 > > .... > > > > Please help, It is very important for me! > > Thank you > > > > Best regards, > > Tom > > > > > > _______________________________________________ > > Xorp-users mailing list > > Xorp-users at xorp.org > > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > > > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090430/54bde79e/attachment-0001.html > > ------------------------------ > > Message: 2 > Date: Thu, 30 Apr 2009 21:45:28 -0400 > From: Irving Leonard > Subject: [Xorp-users] not getting an alias > To: xorp-users at xorp.org > Message-ID: <200904302145.28680.irving.lp at feestudiantes.cujae.edu.cu> > Content-Type: text/plain; charset="us-ascii" > > I am so newbe in xorp but I read (breafly) the manual and it says that every > vif can take a lot of addresses but for me it doesn't make an alias. I have a > vif and give it 2 addresses in the same subnet, because I need to have 2 IPs > but it only get's the first. I'm using ubuntu hardy 8.04.2 > regards > irving > VI Conferencia Internacional de Energ?a Renovable, Ahorro de Energ?a y Educaci?n Energ?tica > 9 - 12 de Junio 2009, Palacio de las Convenciones > ...Por una cultura energ?tica sustentable > www.ciercuba.com > > > > ------------------------------ > > Message: 3 > Date: Fri, 01 May 2009 16:46:30 +0100 > From: Bruce Simpson > Subject: Re: [Xorp-users] Help ! > To: JESUS RAMIREZ > Cc: xorp-users at xorp.org > Message-ID: <49FB1956.7060308 at incunabulum.net> > Content-Type: text/plain; charset=windows-1252; format=flowed > > Hi there, > > I'm assuming you're working with the XORP 1.6 release sources, and are > trying to configure RIPv2 on two interfaces. > > JESUS RAMIREZ wrote: > > > > Hello eveybody ... > > Good Morning, I'm rookie in this ,but I think it's very interesting > > and I want to learn, so I have a lot a questions for you. > > > > If I want to configure 2 network cards ( Ethernet ) were can i > > found the configuration ? I read the manual and i dont konw how can i > > get into the code? or do i need to download from your web and then > > paste it, because i need to configure de rip too....if i need to > > download any file were can i found ? and how can i paste it ? > > > > I don't know if any of you guys can Help me..... please : ( > > > > Probably the best thing for you to do is to look at this file (asuming > you have downloaded xorp-1.6.tar.gz and untarred it) :- > > cd xorp-1.6 > cd rtrmgr/config > > and then view the file "rip.boot" in your favorite text editor. > > The configuration for RIP in that file contains two interfaces dc0 and > dc1. You will probably want to edit this for use on whatever operating > system you're trying to run XORP on. In the case of Linux, you can just > use search and replace to change dc0 to eth0, dc1 to eth1, and then fill > out your IP addresses. > > After that, configuring RIP should just be a matter of following the > instructions in the User Manual to give RIP the settings you need for > your network. > > thanks, > BMS > > > > ------------------------------ > > Message: 4 > Date: Fri, 01 May 2009 16:53:53 +0100 > From: Bruce Simpson > Subject: Re: [Xorp-users] OSPF4 neighbor shown in only one machine not > on the other two > To: bhavin81 at iitb.ac.in > Cc: xorp-users at xorp.org > Message-ID: <49FB1B11.4020508 at incunabulum.net> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi, > > Your configuration file looks OK, which suggests the problem might be > with the VLAN driver support in Linux, or how XORP is trying to use it. > > It would be helpful to see a packet capture on the affected machines, I > believe there are no outstanding bugs for the VLAN support and it has > been tested as working before on generic Linux. > > It is possible there's a difference between your SUSE machines and your > Fedora machines in that behaviour. Can you be sure they are patched to > the correct switch ports in your network? > > You don't mention if the machines successfully synchronized their OSPF > databases. Please send the output of the 'show ospf4 neighbor' commands > when reporting problems with OSPF such as this, as we need to know what > states they're in. > > You can use Wireshark (www.wireshark.org) or tcpdump (www.tcpdump.org), > both should be available in Fedora as yum packages. Be sure to time > what's happening, and try to preserve all XORP logs from your session. > > Please don't send the traces/log files here or to the list -- instead, > attach them to the bug report which Syed has asked you to open. > > bhavin81 at iitb.ac.in wrote: > > Hello Friends, > > > > I have tried to configure ospf4 in xorp. I have configured > > 3 machines for this purpose. When i use the command show ospf4 > > neighbor on one machine(fedora 8) out of three then it shows neighbor > > correclty but when i use the same command on other two machines (suse > > 10.3) then it does not show neighbors on the those machines. > > thanks, > BMS > > > > ------------------------------ > > Message: 5 > Date: Fri, 01 May 2009 17:06:20 +0100 > From: Bruce Simpson > Subject: [Xorp-users] Appeal for suggestions re: automated support > dump tool > To: xorp-users at xorp.org > Message-ID: <49FB1DFC.8080902 at incunabulum.net> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > More often than not, technical questions get asked here about XORP > operation or use, which would probably benefit from a general debugging > information tool -- as a lot of the information which gets asked for, > when problems are raised, is repetitive and common to all bug reports. > > Rather than request this information every time, which makes it more > difficult to give support, it might be better if we could provide > functionality in an automated way. What we'd really appreciate from > folks is suggestions for such a tool. > > The basics would be: OS version, OS network interface config, PCI info, > followed of course by the XORP config dumps for each routing process. > > Such a tool could get rolled into the Router Manager. However, such a > tool is probably best written in a modern, common scripting language. > Python is particularly good for tasks lie this. > > Please share your suggestions and feedback. > > thanks, > BMS > > > > ------------------------------ > > Message: 6 > Date: Fri, 01 May 2009 17:17:53 +0100 > From: Bruce Simpson > Subject: Re: [Xorp-users] virtual link > To: Tomek > Cc: xorp-users at xorp.org > Message-ID: <49FB20B1.1040002 at incunabulum.net> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Tomek wrote: > > Hi, > > > > I've got a problem with xorp configuration. > > Actually in my case situation looks like this: > > > > R3 <-> R5 <-> R4 <-> R6 > > > > where: > > > > R3 is an ABR of area 0 (left side) and area 1 (right side) > > R5 is internal of area 1 > > R4 is ABR of area 1(left side) and area 3 (right side) > > R6 internal of area 3 > > > > everything works fine until i want to configure virtual link between R3 and R4 > > I added virtual-link section to area 0.0.0.0 in ospf4 section > > > ... > > * Which version of XORP are you using? > * Which platform are you using XORP on? > > area 0.0.0.0 { > > virtual-link 192.168.0.40 { > > transit-area: 0.0.0.1 > > } > > .... > > and tried to add virtual-link to R4 without succeed because of error: > > Command failed Virtual link must be in area 0.0.0.0 > > > > my question is how to configure virtual link at R4 ? > > > > This could be a problem in vlinks traversing area borders, or more > likely, a problem with the lookup by router ID. > > I'd have to check this code more fully before I could respond in detail > about that. > > The configuration steps you posted look correct, assuming that you are > configuring XORP up on R3.which is an ABR. But is area 1 a transit area > or a stub area? An NSSA area is *not* valid for virtual links, they can > only be configured across transit areas. > > More general questions: > Does R3 learn about R4 without the virtual-link config? > If so, which area(s) does R3 learn that R4 is in? > Does R5 see the same topology (neighbours) as R3 and R4? > > thanks, > BMS > > > > ------------------------------ > > Message: 7 > Date: Fri, 01 May 2009 17:27:46 +0100 > From: Bruce Simpson > Subject: Re: [Xorp-users] not getting an alias > To: Irving Leonard > Cc: xorp-users at xorp.org > Message-ID: <49FB2302.3040007 at incunabulum.net> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi, > > Can you post your full configuration file, and tell us which kernel > version are you using, so that others may reproduce? > > Irving Leonard wrote: > > I am so newbe in xorp but I read (breafly) the manual and it says that every > > vif can take a lot of addresses but for me it doesn't make an alias. I have a > > vif and give it 2 addresses in the same subnet, because I need to have 2 IPs > > but it only get's the first. I'm using ubuntu hardy 8.04.2 > > regards > > irving > > thanks, > BMS > > > > ------------------------------ > > _______________________________________________ > Xorp-users mailing list > Xorp-users at xorp.org > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > > > End of Xorp-users Digest, Vol 38, Issue 1 > ***************************************** _________________________________________________________________ Hotmail? goes with you. http://windowslive.com/Tutorial/Hotmail/Mobile?ocid=TXT_TAGLM_WL_HM_Tutorial_Mobile1_052009 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090503/f4335e40/attachment-0001.html From bhavin81 at iitb.ac.in Sun May 3 23:44:48 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Mon, 4 May 2009 12:14:48 +0530 (IST) Subject: [Xorp-users] Can anyone tell me how to specify source in multicast tree formed with pimsm in xorp Message-ID: <6e00e6ff51de4b369064d2ea6330ca0c.squirrel@gpo.iitb.ac.in> Hello friends, Can anyone tell me how to specify source in multicast tree formed by pimsm? Nothing is mentioned of source in the pim-sm configuration file of user manual. If we want to form multicast-tree then we need to be aware of source so that we can check the multicast-tree formed is ok or not? Thanks in advance for your valuable help. Regards, Bhavin, Comm. Student. From bhavin81 at iitb.ac.in Mon May 4 00:38:39 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Mon, 4 May 2009 13:08:39 +0530 (IST) Subject: [Xorp-users] Can anyone tell me how to specify source in multicast tree formed with pimsm in xorp Message-ID: Hello friends, Can anyone tell me how to specify source in multicast tree formed by pimsm? Nothing is mentioned of source in the pim-sm configuration file of user manual. If we want to form multicast-tree then we need to be aware of source so that we can check the multicast-tree formed is ok or not? Thanks in advance for your valuable help. Regards, Bhavin, Comm. Student. From alejandro.berna at i2cat.net Mon May 4 09:57:02 2009 From: alejandro.berna at i2cat.net (Alejandro Berna) Date: Mon, 04 May 2009 18:57:02 +0200 Subject: [Xorp-users] How to change loopback interface? Message-ID: <49FF1E5E.6090306@i2cat.net> Hi, I am trying to configure OSPF changing the loopback interfaces of my XORP routers. I have not found how to do with Xorpsh therefore I have done with ifconfig linux command. But when I try to run my config.boot file with the OSPF configuration it fails because XORP still thinks that the loopback address is 127.0.0.1, here is the trace of the error: # xorp_rtrmgr [ 2009/05/04 17:53:43 ERROR xorp_rtrmgr:5389 LIBCOMM +312 comm_sock.c comm_sock_bind4 ] Error binding socket (family = 2, my_addr = 127.0.0.1, my_port = 19999): Cannot assing requested address [ 2009/05/04 17.53.43 ERROR xorp_rtrmgr:5389 RTRMGR +243 main_rtrmgr.cc run ] Cannot assign requested address: a finder may already be running. Here is my config file is: interfaces { interface eth1 { vif eth1 { address 10.129.0.3 { prefix-length: 24 } } } } fea { unicast-forwarding4 { disable: false } } protocols { ospf4 { router-id: 192.168.0.1 area 0.0.0.1 { interface eth1 { vif eth1 { address 10.129.0.3 { } } } } } } When I again put with ifconfig the loopback to 127.0.0.1 it doesn't fail. I want to define my loopback address as the router-id, 192.168.0.1, how can I change the property my address? And how can I change the loopback address with XORP? Thanks for everything, Alejandro Berna From chuyram28 at msn.com Mon May 4 10:42:12 2009 From: chuyram28 at msn.com (JESUS RAMIREZ) Date: Mon, 4 May 2009 11:42:12 -0600 Subject: [Xorp-users] Xorp-users Digest, Vol 38, Issue 1 In-Reply-To: References: Message-ID: Hello all, It's me again, well , I fixed some problems in my configuration, (let me explain whats is my objective ) i want one router (xorp with red hat so ) and i tried to connect 2 computers ( windows ) in my router and i installed 1 more eth card, but xorp doesn't recognized , in my interfaces only shows fxp0, discard0, in fxp0 i set address 10.1.1.254/24 but i don't why does't show the other eth card... and for discard0 has the address 192.0.2.1/32 for default but i dont know for what is this interface.... when i do a ping from windows machine shows "destination host unreachable" I tried to modify some files but when i tried to save it in usr/local/sbin don't let me do it. ( rip.boot ) what Iam missing ? Thank you for you help. _________________________________________________________________ Hotmail? has ever-growing storage! Don?t worry about storage limits. http://windowslive.com/Tutorial/Hotmail/Storage?ocid=TXT_TAGLM_WL_HM_Tutorial_Storage1_052009 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090504/4f1634d2/attachment.html From segv at o2.pl Mon May 4 22:38:18 2009 From: segv at o2.pl (=?UTF-8?Q?Tomek?=) Date: Tue, 05 May 2009 07:38:18 +0200 Subject: [Xorp-users] =?utf-8?q?Fwd=3A_Re=3A__virtual_link?= Message-ID: <2aa509eb.572024cc.49ffd0ca.a4820@o2.pl> Dnia 2 maja 2009 18:52 Tomek napisa?(a): > Hi, > Thank you guys for yours replies. > I use xorp ver. 1.6 at R4,R3 and the rest uses 1.5. > So R3 doesn't learn about R4 what is really strange, dunno why.. > Below I listed my config files from R3 and R4: > > ********* R3 *************** > interfaces { > restore-original-config-on-shutdown: false > interface eth1 { > vif eth1 { > address 192.168.2.30 { > prefix-length: 24 > } > } > } > interface eth2 { > vif eth2 { > address 192.168.3.30 { > prefix-length: 24 > } > } > } > interface eth3 { > vif eth3 { > address 192.168.6.30 { > prefix-length: 24 > } > } > } > } > > fea { > unicast-forwarding4 { > disable: false > forwarding-entries { > retain-on-startup: false > retain-on-shutdown: false > } > } > } > protocols { > ospf4 { > router-id: 192.168.0.30 > > area 0.0.0.1 { > interface eth3 { > vif eth3 { > address 192.168.6.30 { > interface-cost: 2 > } > } > } > } > > area 0.0.0.0 { > /* virtual-link 192.168.0.40 { > transit-area: 0.0.0.1 > } > */ > interface eth1 { > vif eth1 { > address 192.168.2.30 { > interface-cost: 3 > } > } > } > interface eth2 { > vif eth2 { > address 192.168.3.30 { > interface-cost: 1 > } > } > } > } > } > } > > ********* END *************** > > ********* R4 *************** > > interfaces { > restore-original-config-on-shutdown: false > /* interface eth1 { > description: "Ethernet Interface #1" > disable: false > default-system-config > } > */ > interface eth2 { > description: "Ethernet Interface #2" > disable: false > default-system-config > } > interface eth3 { > description: "Ethernet Interface #3" > disable: false > default-system-config > } > } > > fea { > unicast-forwarding4 { > disable: false > forwarding-entries { > retain-on-startup: false > retain-on-shutdown: false > } > } > > } > protocols { > ospf4 { > router-id: 192.168.0.40 > area 0.0.0.1 { > > /* interface eth1 { > vif eth1 { > address 192.168.4.40 { > interface-cost: 4 > } > } > } > */ > interface eth2 { > vif eth2 { > address 192.168.5.40 { > interface-cost: 6 > } > } > } > } > area 0.0.0.3 { > interface eth3 { > vif eth3 { > address 192.168.7.40 { > interface-cost: 1 > } > } > } > } > } > } > > > ********* END *************** > > ********* R5 *************** > interfaces { > restore-original-config-on-shutdown: false > interface eth1 { > description: "Ethernet Interface #1" > disable: false > default-system-config > } > interface eth2 { > description: "Ethernet Interface #2" > disable: false > default-system-config > } > > } > > fea { > unicast-forwarding4 { > disable: false > forwarding-entries { > retain-on-startup: false > retain-on-shutdown: false > } > } > > } > protocols { > ospf4 { > router-id: 192.168.0.50 > area 0.0.0.1 { > interface eth1 { > vif eth1 { > address 192.168.6.50 { > interface-cost: 2 > } > } > } > interface eth2 { > vif eth2 { > address 192.168.5.50 { > interface-cost: 6 > } > } > } > } > > } > } > > ********* END *************** > > BR > Tom > > Dnia 1 maja 2009 18:17 Bruce Simpson napisa?(a): > > > Tomek wrote: > > > Hi, > > > > > > I've got a problem with xorp configuration. > > > Actually in my case situation looks like this: > > > > > > R3 R5 R4 R6 > > > > > > where: > > > > > > R3 is an ABR of area 0 (left side) and area 1 (right side) > > > R5 is internal of area 1 > > > R4 is ABR of area 1(left side) and area 3 (right side) > > > R6 internal of area 3 > > > > > > everything works fine until i want to configure virtual link between R3 and R4 > > > I added virtual-link section to area 0.0.0.0 in ospf4 section > > > > > ... > > > > * Which version of XORP are you using? > > * Which platform are you using XORP on? > > > area 0.0.0.0 { > > > virtual-link 192.168.0.40 { > > > transit-area: 0.0.0.1 > > > } > > > .... > > > and tried to add virtual-link to R4 without succeed because of error: > > > Command failed Virtual link must be in area 0.0.0.0 > > > > > > my question is how to configure virtual link at R4 ? > > > > > > > This could be a problem in vlinks traversing area borders, or more > > likely, a problem with the lookup by router ID. > > > > I'd have to check this code more fully before I could respond in detail > > about that. > > > > The configuration steps you posted look correct, assuming that you are > > configuring XORP up on R3.which is an ABR. But is area 1 a transit area > > or a stub area? An NSSA area is *not* valid for virtual links, they can > > only be configured across transit areas. > > > > More general questions: > > Does R3 learn about R4 without the virtual-link config? > > If so, which area(s) does R3 learn that R4 is in? > > Does R5 see the same topology (neighbours) as R3 and R4? > > > > thanks, > > BMS > > > From naresh_raga at yahoo.co.in Tue May 5 03:58:03 2009 From: naresh_raga at yahoo.co.in (naresh raga) Date: Tue, 5 May 2009 16:28:03 +0530 (IST) Subject: [Xorp-users] Facing problems while building xorp on Debian sarge 2.4.26 kernel Message-ID: <969701.98745.qm@web94814.mail.in2.yahoo.com> Hello, My aim is to build xorp on ts-7300 ARM kit.The OS on the kit is Debian sarge 2.4.26 and I am using xorp-1.6 version.I am unable to build on it.I have installed all softwares required for Debian sarge as said in BUILD_NOTES.Is it a problem with the kernel?Do xorp-1.6 requires 2.6 kernel or 2.4 is enough?Do I need to use older version of xorp in this case? Own a website.Get an unlimited package.Pay next to nothing.*Go to http://in.business.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090505/2a557ed0/attachment.html From naresh_raga at yahoo.co.in Wed May 6 14:10:25 2009 From: naresh_raga at yahoo.co.in (naresh raga) Date: Thu, 7 May 2009 02:40:25 +0530 (IST) Subject: [Xorp-users] How to implement NAT on xorp Message-ID: <292183.76814.qm@web94814.mail.in2.yahoo.com> Hai, I have installed xorp on a system with two NICs(eth2 and eth3).Internet connection has been given to eth2.I want to get internet on a laptop connected to eth3.I am able to ping eth3(172.16.0.1) from my laptop but unable to get internet.My configuration file is as below: interfaces{ interface eth2 { ??? description: "IT LAB" ??? mac: 00:0c:f1:6d:51:d3? ??? disable: false ??? vif eth2 { ????? disable: false ????? address 10.64.25.77{ ??????? prefix-length: 16 ??????? broadcast: 10.64.255.255 ??????? disable: false ????? } ??? } ? } interface eth3 { ??? description: "PLC LAN" ??? mac:? 00:0c:f1:6d:51:d4 ??? disable: false ??? vif eth3 { ????? disable: false ????? address 172.16.0.1 { ??????? prefix-length: 24 ??????? broadcast: 172.16.0.255 ??????? disable: false ????? } ????? ??? } ? } } fea { ? unicast-forwarding4 { ??? disable: false ? } } plumbing { ? mfea4 { ??? disable: false ??? interface eth2 { ????? vif eth2 { ??????? disable: false ????? } ??? } ?interface eth3 { ????? vif eth3 { ??????? disable: false ????? } ??? } ??? ??? interface register_vif { ????? vif register_vif { ??????? /* Note: this vif should be always enabled */ ??????? disable: false ????? } ??? } ??? traceoptions { ????? flag all { ??????? disable: false ????? } ??? } ? } } protocols { ? static { ??? route 0.0.0.0/0 { ????? next-hop: 10.64.1.1 ????? metric: 1 ????? } ? mrib-route 0.0.0.0/0 { ? next-hop: 10.64.1.1 ? metric: 1 ??? } ? } } When I am trying to connect to net on my laptop(connected to eth3),I am unable to connect.But I can ping to eth3.What could be the problem?What more I have to add to this file for implementing getting net on my laptop? Thanks in advance, Naresh ?????????????????????????????? Bollywood news, movie reviews, film trailers and more! Go to http://in.movies.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090507/c8319b44/attachment.html From naresh_raga at yahoo.co.in Thu May 7 02:53:34 2009 From: naresh_raga at yahoo.co.in (naresh raga) Date: Thu, 7 May 2009 15:23:34 +0530 (IST) Subject: [Xorp-users] Fw: How to implement NAT on xorp Message-ID: <280624.13454.qm@web94815.mail.in2.yahoo.com> --- On Thu, 7/5/09, naresh raga wrote: From: naresh raga Subject: [Xorp-users] How to implement NAT on xorp To: Xorp-users at xorp.org Date: Thursday, 7 May, 2009, 2:40 AM Hai, I have installed xorp on a system(PC) with two NICs(eth2 and eth3).Internet connection has been given to eth2.I want to get internet on a laptop connected to eth3.I am able to ping eth3(172.16.0.1) from my laptop but unable to get internet on it. Inernet-------|eth2|PC|eth3|-----------laptop(I need net on this laptop). My configuration file is as below: interfaces{ interface eth2 { ??? description: "IT LAB" ??? mac: 00:0c:f1:6d:51:d3? ??? disable: false ??? vif eth2 { ????? disable: false ????? address 10.64.25.77{ //active net connection IP ??????? prefix-length: 16 ??????? broadcast: 10.64.255.255 ??????? disable: false ????? } ??? } ? } interface eth3 { ??? description: "PLC LAN" ??? mac:? 00:0c:f1:6d:51:d4 ??? disable: false ??? vif eth3 { ????? disable: false ????? address 172.16.0.1 { ??????? prefix-length: 24 ??????? broadcast: 172.16.0.255 ??????? disable: false ????? } ????? ??? } ? } } fea { ? unicast-forwarding4 { ??? disable: false ? } } plumbing { ? mfea4 { ??? disable: false ??? interface eth2 { ????? vif eth2 { ??????? disable: false ????? } ??? } ?interface eth3 { ????? vif eth3 { ??????? disable: false ????? } ??? } ??? ??? interface register_vif { ????? vif register_vif { ??????? /* Note: this vif should be always enabled */ ??????? disable: false ????? } ??? } ??? traceoptions { ????? flag all { ??????? disable: false ????? } ??? } ? } } protocols { ? static { ??? route 0.0.0.0/0 { ????? next-hop: 10.64.1.1 ????? metric: 1 ????? } ? mrib-route 0.0.0.0/0 { ? next-hop: 10.64.1.1 ? metric: 1 ??? } ? } } Here 10.64.1.1 is my lab main router. When I am trying to connect to net on my laptop(connected to eth3),I am unable to connect.But I can ping to eth3.What could be the problem?What more I have to add to this file for implementing getting net on my laptop?Do I need to implement IBGP protocol? Please help me out as this is part of my minor project.. Thanks in advance, Naresh ?????????????????????????????? Now surf faster and smarter ! Check out the new Firefox 3 - Yahoo! Edition * Click here! -----Inline Attachment Follows----- _______________________________________________ Xorp-users mailing list Xorp-users at xorp.org http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users Now surf faster and smarter ! Check out the new Firefox 3 - Yahoo! Edition http://downloads.yahoo.com/in/firefox/?fr=om_email_firefox -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090507/5b8a14fa/attachment.html From jhay at meraka.org.za Thu May 7 05:37:00 2009 From: jhay at meraka.org.za (John Hay) Date: Thu, 7 May 2009 14:37:00 +0200 Subject: [Xorp-users] rip on freebsd Message-ID: <20090507123700.GA43697@zibbi.meraka.csir.co.za> Hi, I am new to xorp, so maybe I just have a config problem. I am using FreeBSD amd64 7.2-RELEASE. I am using xorp-1.6 from the 7.2 packages. My target is to get RIP, RIPNG, BGP and PIM-SM going to replace quagga that can do the first three but not pim-sm. So I thought I would start with RIP, but seems that there must be a race or something. I am sure I once saw xorp actually listening on port 520 and exchanging RIP packets, but then I did a .../xorp stop and start and now it does not exchange RIP packets anymore. According to the various show commands inside xorpsh, it looks like it xorp thinks it is working, but tcpdump show nothing being transmitted. I have started with an empty conf file and then used xorpsh to do the config and saved that. The config + xorp_rtrmgr.log and sockstat.out is on http://people.freebsd.org/~jhay/xorp/ Anybody with any ideas? Thanks. John -- John Hay -- John.Hay at meraka.csir.co.za / jhay at FreeBSD.org From bms at incunabulum.net Thu May 7 07:49:51 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 15:49:51 +0100 Subject: [Xorp-users] Facing problems while building xorp on Debian sarge 2.4.26 kernel In-Reply-To: <969701.98745.qm@web94814.mail.in2.yahoo.com> References: <969701.98745.qm@web94814.mail.in2.yahoo.com> Message-ID: <4A02F50F.1010500@incunabulum.net> naresh raga wrote: > Hello, > My aim is to build xorp on ts-7300 ARM kit.The OS on the kit is Debian > sarge 2.4.26 and I am using xorp-1.6 version.I am unable to build on > it.I have installed all softwares required for Debian sarge as said in > BUILD_NOTES.Is it a problem with the kernel?Do xorp-1.6 requires 2.6 > kernel or 2.4 is enough?Do I need to use older version of xorp in this > case? > Hi, I can't speak for Debian specifically. Are you attempting to cross-compile XORP? If so, what kind of build host do you have? Can you post the output of the configure command and the config.log to this list? It is quite possible to have all requisite packages installed on the host system, but are you installing these on the build host itself, or in a staging area for your cross-compilation? I highly advise having a copy of the Autoconf Book by Tom Tromey et al when debugging these kinds of issues. Very generally speaking, a Linux 2.6 kernel is preferable to Linux 2.4 for a number of reasons which I won't go into here for brevity. thanks, BMS P.S. Please note that I cannot respond personally to your queries, so please direct replies to the mailing list. Thank you. From bms at incunabulum.net Thu May 7 07:54:54 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 15:54:54 +0100 Subject: [Xorp-users] rip on freebsd In-Reply-To: <20090507123700.GA43697@zibbi.meraka.csir.co.za> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> Message-ID: <4A02F63E.9050208@incunabulum.net> John Hay wrote: > ... > I have started with an empty conf file and then used xorpsh to do > the config and saved that. The config + xorp_rtrmgr.log and > sockstat.out is on http://people.freebsd.org/~jhay/xorp/ > Your config file looks fine. I will hazard a wild guess: try adding a default route, using the XORP static routes module, to your config, and see if this condition is still present. I did a LOT of multicast refactoring in FreeBSD. I don't believe 7.2 should have the problem I inadvertently introduced in the 6.x train with default interface lookup, but I would like to be able to rule it out -- AFAIK it did not bite XORP, and I haven't seen that problem in the 7.x train since. The next place I'd look is ktrace-ing the FEA; use ktrace(8) to pull FEA activity at runtime, preferably as soon as it starts to run. A shell script with kill -0 to detect the process has started is fine, it's not too racy. The libcomm message involved could probably be made more useful. Also pull sockstat and check that the socket(s) are in fact being created. Multicast sockets shouldn't be bound to an interface, but they do need to be bound to the port (INADDR_ANY, 520). thanks, BMS From bms at incunabulum.net Thu May 7 07:57:56 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 15:57:56 +0100 Subject: [Xorp-users] How to implement NAT on xorp In-Reply-To: <292183.76814.qm@web94814.mail.in2.yahoo.com> References: <292183.76814.qm@web94814.mail.in2.yahoo.com> Message-ID: <4A02F6F4.4080709@incunabulum.net> naresh raga wrote: > Hai, > I have installed xorp on a system with two NICs(eth2 and > eth3).Internet connection has been given to eth2.I want to get > internet on a laptop connected to eth3.I am able to ping > eth3(172.16.0.1) from my laptop but unable to get internet.My > configuration file is as below: > ... > When I am trying to connect to net on my laptop(connected to eth3),I > am unable to connect.But I can ping to eth3.What could be the > problem?What more I have to add to this file for implementing getting > net on my laptop? > Your config file looks fine. However, to the best of my knowledge, we don't implement NAT in XORP. You will need to configure your system's NAT capability outside of XORP for now, using whatever capabilities your platform for running XORP on has. thanks, BMS From bms at incunabulum.net Thu May 7 08:04:17 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 16:04:17 +0100 Subject: [Xorp-users] How to change loopback interface? In-Reply-To: <49FF1E5E.6090306@i2cat.net> References: <49FF1E5E.6090306@i2cat.net> Message-ID: <4A02F871.7010007@incunabulum.net> Alejandro Berna wrote: > Hi, I am trying to configure OSPF changing the loopback interfaces of my > XORP routers. I have not found how to do with Xorpsh therefore I have > done with ifconfig linux command. But when I try to run my config.boot > file with the OSPF configuration it fails because XORP still thinks that > the loopback address is 127.0.0.1, here is the trace of the error: > ... > > When I again put with ifconfig the loopback to 127.0.0.1 it doesn't > fail. I want to define my loopback address as the router-id, > 192.168.0.1, how can I change the property my address? And how can I > change the loopback address with XORP? Thanks for everything, > I can't understand why you want to configure your loopback interface with an RFC 1918 prefix, this is going to break a lot of networking software, not just XORP :-) You need to do one of these following things: 1. configure a separate loopback interface instance using your operating system's capabilities to do this, and configure it in the router manager, *OR* :- 2. configure *both* the 127.0.0.1 address and another address on the default loopback interface. This is usually named lo or lo0. It is best not to remove 127.0.0.1 from most IPv4 capable hosts. XORP in particular assumes the existence of 127.0.0.1 and uses TCP sockets bound to this interface address as the default IPC mechanism. It is just one of many pieces of software which do this, so please don't remove this address. To the best of my knowledge, XORP's OSPF implementation doesn't need the loopback address to be configured with the router ID; as far as I know the router ID is an opaque identifier. It is used for winning DR elections, sure, but XORP doesn't need to see it on the loopback interface. Of course, the normative reference is the code. cheers, BMS From bms at incunabulum.net Thu May 7 08:12:36 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 16:12:36 +0100 Subject: [Xorp-users] Xorp-users Digest, Vol 38, Issue 1 In-Reply-To: References: Message-ID: <4A02FA64.9080002@incunabulum.net> JESUS RAMIREZ wrote: > Hello all, It's me again, well , I fixed some problems in my > configuration, (let me explain whats is my objective ) i want one > router (xorp with red hat so ) and i tried to connect 2 computers ( > windows ) in my router and i installed 1 more eth card, but xorp > doesn't recognized , in my interfaces only shows fxp0, discard0, in > fxp0 i set address 10.1.1.254/24 but i don't why does't show the other > eth card... and for discard0 has the address 192.0.2.1/32 for default > but i dont know for what is this interface.... when i do a ping from > windows machine shows "destination host unreachable" It sounds like you are trying to use FreeBSD interface names on a Red Hat system. Is this correct? As explained in previous replies you will need to edit the rip.boot example configuration file for your system, and consult the user manual for advice on how to do this. Are both your Windows machine and the XORP router configured to be on the same LAN, are they plugged in, patched etc? > > I tried to modify some files but when i tried to save it in > usr/local/sbin don't let me do it. ( rip.boot ) Have you checked your file permissions, is this a writable filesystem? Which command(s) did you try to use in order to save the file? thanks, BMS P.S. Please direct support queries to this mailing list as I cannot respond personally. Thank you. From bms at incunabulum.net Thu May 7 08:14:14 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 16:14:14 +0100 Subject: [Xorp-users] Fwd: Re: virtual link In-Reply-To: <2aa509eb.572024cc.49ffd0ca.a4820@o2.pl> References: <2aa509eb.572024cc.49ffd0ca.a4820@o2.pl> Message-ID: <4A02FAC6.2060301@incunabulum.net> Tomek wrote: > Dnia 2 maja 2009 18:52 Tomek napisa?(a): > >> Hi, >> Thank you guys for yours replies. >> I use xorp ver. 1.6 at R4,R3 and the rest uses 1.5. >> So R3 doesn't learn about R4 what is really strange, dunno why.. >> Below I listed my config files from R3 and R4: >> A number of bugs related to OSPF and virtual links were found and fixed in the 1.5 to 1.6 development cycle. Are you able to update your XORP 1.5 systems to the 1.6 version and confirm if the problem still exists? Thanks, BMS P.S. Please direct support queries to the list as I cannot respond personally. Thank you. From bms at incunabulum.net Thu May 7 08:21:27 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 16:21:27 +0100 Subject: [Xorp-users] querry: Why Dropping packet from foreign network 10.10.20.0/24? In-Reply-To: <1a86605a2654d1f3cc85924d5bb52892.squirrel@gpo.iitb.ac.in> References: <1a86605a2654d1f3cc85924d5bb52892.squirrel@gpo.iitb.ac.in> Message-ID: <4A02FC77.8030501@incunabulum.net> bhavin81 at iitb.ac.in wrote: > Hello friends, > > I have installed xorp on two suse 10.3 machines. Both the machines > are connected to same switch. I have configured one machine with the > following interface address: 10.10.10.10 and other machine with the > adddress > 10.10.10.20. Now i am getting the following message when i run > ./xorp_rtrmgr on machine 1: > Dropping packet from foreign network 10.10.20.0/24 > and the following message on machine2: > Dropping packet from foreign network 10.10.10.0/24 > The machines are clearly configured to be in different IPv4 subnets, therefore they will not form an OSPF adjacency. So, XORP is behaving according to design here. Is this intentional on your part? The fact that the traffic gets through at all on your second machine, without VLAN encapsulation, isn't really relevant, and I can neither prove nor disprove that, not knowing the VLAN switch config. So I'm assuming that you did in fact configure the VLAN encapsulation to be stripped on the port on the switch which your second machine is connected to. If you want these two OSPF routers to form an adjacency, you must configure them to be in the same IPv4 subnet. thanks, BMS From bms at incunabulum.net Thu May 7 08:24:40 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 16:24:40 +0100 Subject: [Xorp-users] not getting an alias In-Reply-To: <200905011944.25100.irving.lp@feestudiantes.cujae.edu.cu> References: <200904302145.28680.irving.lp@feestudiantes.cujae.edu.cu> <49FB2302.3040007@incunabulum.net> <200905011944.25100.irving.lp@feestudiantes.cujae.edu.cu> Message-ID: <4A02FD38.5060201@incunabulum.net> Irving Leonard wrote: > On Viernes 01 Mayo 2009 12:27:46 PM Bruce Simpson escribi?: > >> Hi, >> >> Can you post your full configuration file, and tell us which kernel >> version are you using, so that others may reproduce? >> > > I use 4 fast ethernet NICs and combined in a bond wich name is bond0; the > network config file (/etc/network/interfaces) creates several vlans (including > bond0.601) but only the devices (not configured with ips or up). The commented > section down is the configuration of the bond (not vlan) interface that is in > the /etc/network/interfaces. > Can you post logs of what happens when you try to do this? Your configuration file looks OK, and having more than one address on an interface is perfectly OK by XORP, but I'm not sure I understand much about the network topology involved, or what XORP's actual failure mode is from your explanation, so if you could clarify further, it would be easier to help you. thanks, BMS From bms at incunabulum.net Thu May 7 08:26:57 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 16:26:57 +0100 Subject: [Xorp-users] Problem with Xorp passing BSR messages In-Reply-To: <1B62E6082D9F034394DB2FBAC03D1A3845F33F8CEF@lrccd-exch05.LRCCD.ad.losrios.edu> References: <1B62E6082D9F034394DB2FBAC03D1A3845F33F8CEF@lrccd-exch05.LRCCD.ad.losrios.edu> Message-ID: <4A02FDC1.1090903@incunabulum.net> Hi Mark, Sorry for my late reply... Williams, Mark wrote: > I will be deploying Xorp on Linux systems soon to handle multicast routing with PIM-SM. Everything has worked well, but I have had problems with bootstrap router messages. In my testing, I used two Alcatel-Lucent routers and an Xorp system. Each router had two interfaces, and were in a chain in Alcatel - Xorp - Alcatel order. > > When the Xorp box was configured as the candidate BSR, the Alcatel routers both picked up the Xorp box as the BSR after the election time passed. > It wouldn't surprise me if there were cobwebs in the Cand-BSR mechanisms, but as Syed has said, it would need to be reproduced in a lab situation to better understand the problem and how to fix. thanks, BMS From alejandro.berna at i2cat.net Thu May 7 08:43:36 2009 From: alejandro.berna at i2cat.net (Alejandro Berna) Date: Thu, 07 May 2009 17:43:36 +0200 Subject: [Xorp-users] How to change loopback interface? In-Reply-To: <4A02F871.7010007@incunabulum.net> References: <49FF1E5E.6090306@i2cat.net> <4A02F871.7010007@incunabulum.net> Message-ID: <4A0301A8.7070500@i2cat.net> Thanks Bruce, I found finally some previous conversation where it was explained (with less detail than you gave) more or less the loopback problem. I'm just performing some tests, nothing important, this is because I'm using RFC 1918 prefix. I am using Ubuntu server so I performed: #ip address add 192.168.0.1/24 dev lo Then I keep 127.0.0.1 as the local loopback address. I am developing some code to configure OSPF in a virtual machine with XORP and I wanted to automate all the process writing the .boot file or with xorpsh, not using Ubuntu cli. But I guess it's not possible, anyway it's not a big problem so my code performs a ssh connection to the Ubuntu machine. Anyway thanks a lot for the reply, Alejandro Bruce Simpson escribi?: > Alejandro Berna wrote: >> Hi, I am trying to configure OSPF changing the loopback interfaces of >> my XORP routers. I have not found how to do with Xorpsh therefore I >> have done with ifconfig linux command. But when I try to run my >> config.boot file with the OSPF configuration it fails because XORP >> still thinks that the loopback address is 127.0.0.1, here is the >> trace of the error: >> ... >> >> When I again put with ifconfig the loopback to 127.0.0.1 it doesn't >> fail. I want to define my loopback address as the router-id, >> 192.168.0.1, how can I change the property my address? And how can I >> change the loopback address with XORP? Thanks for everything, >> > > I can't understand why you want to configure your loopback interface > with an RFC 1918 prefix, this is going to break a lot of networking > software, not just XORP :-) > > You need to do one of these following things: > > 1. configure a separate loopback interface instance using your > operating system's capabilities to do this, and configure it in the > router manager, *OR* :- > 2. configure *both* the 127.0.0.1 address and another address on the > default loopback interface. This is usually named lo or lo0. > > It is best not to remove 127.0.0.1 from most IPv4 capable hosts. XORP > in particular assumes the existence of 127.0.0.1 and uses TCP sockets > bound to this interface address as the default IPC mechanism. It is > just one of many pieces of software which do this, so please don't > remove this address. > > To the best of my knowledge, XORP's OSPF implementation doesn't need > the loopback address to be configured with the router ID; as far as I > know the router ID is an opaque identifier. It is used for winning DR > elections, sure, but XORP doesn't need to see it on the loopback > interface. Of course, the normative reference is the code. > > cheers, > BMS > > > -- Alejandro Berna Juan Network Technologies Cluster (CTX) Fundaci? i2Cat, Internet i Innovaci? Digital a Catalunya C/ Gran Capit? 2 - 4, Nexus I building, 2nd floor, office 203 08034 Barcelona, Catalonia, Spain P: +34 93 553 2522 F: +34 93 553 2520 From yindong1982 at gmail.com Thu May 7 08:46:26 2009 From: yindong1982 at gmail.com (dong yin) Date: Thu, 7 May 2009 11:46:26 -0400 Subject: [Xorp-users] XORP 1.6 support Click 1.6.0 ? Message-ID: Hi all, Dose XORP 1.6 support Click v1.6.0? When I run XORP 1.6 to control the remote Click forwarding data plane (click 1.6.0), errors happened. ///////////////////////////////////////////////////// [ 2009/05/07 10:35:50 INFO xorp_rtrmgr:4428 RTRMGR +101 module_manager.cc execute ] Executing module: firewall (fea/xorp_fea) [ 2009/05/07 10:35:54 INFO xorp_rtrmgr:4428 RTRMGR +101 module_manager.cc execute ] Executing module: fea (fea/xorp_fea) [ 2009/05/07 10:35:59 ERROR xorp_fea:4429 FEA +710 ifconfig_set_click.cc click_config_generator_done ] Failed to write the Click configuration: User-level Click command error: 520-Write handler 'hotconfig' error: 520-:1: '_xorp_ip4 :: Strip' input 0 unused 520-:3: '_xorp_arpt :: Tee' input 0 unused 520 Router could not be initialized! ///////////////////////////////////////////////////// My config file is below: ///////////////////////////////////////////////////// interfaces { interface eth1 { description: "control interface" disable: false mac: 00:1b:21:2c:a5:06 default-system-config } interface eth2 { description: "data interface" disable: false mac: 00:04:76:2f:24:91 default-system-config /* vif eth2 { disable: false address 10.1.1.1 { prefix-length: 24 } }*/ } } fea { unicast-forwarding4 { disable: true } click { disable: false kernel-click { disable: true } user-click { disable: false command-file: "/usr/local/bin/click" /* command-extra-arguments: "-R"*/ control-address: 10.0.0.2 control-socket-port: 13000 /*startup-config-file: "/root/xorptest/router-user.click" *user-click-config-generator-file: *"/usr/local/xorp/fea/xorp_fea_click_config_generator" */ } } } protocols { static { route 4.4.4.0/24 { next-hop: 5.5.5.1 metric: 1 } route 5.5.5.0/24 { next-hop: 4.4.4.1 metric: 1 } } } ////////////////////////////////////////////////////// And my click config file is : ///////////////////////////////////////////////////// AddressInfo ( net2 192.168.2.0, net3 192.168.3.0 ) ControlSocket("TCP", 13000); c2 :: Classifier(12/0806 20/0001, 12/0806 20/0002, 12/0800, -); FromDevice(eth1) -> [0]c2; out2 :: Queue(1024) -> ToDevice(eth1); ar2 :: ARPResponder(eth1:ip eth1:eth) -> out2; aq2 :: ARPQuerier(eth1:ip, eth1:eth) -> out2; c2[0] -> ar2; c2[1] -> [1]aq2; c2[3] -> Discard; _xorp_rt4 :: RangeIPLookup(net2/24 0, net3/24 1, 0.0.0.0/0 2); c2[2] -> Strip(14) -> CheckIPHeader2 -> [0]_xorp_rt4; _xorp_rt4[0] -> [0]aq2; _xorp_rt4[1] -> [0]aq2; _xorp_rt4[2] -> Discard; /////////////////////////////////////////////////////////////// Thanks a lot. Cheers, Dong -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090507/ad23306f/attachment-0001.html From irving.lp at feestudiantes.cujae.edu.cu Thu May 7 08:55:30 2009 From: irving.lp at feestudiantes.cujae.edu.cu (Irving Leonard) Date: Thu, 7 May 2009 11:55:30 -0400 Subject: [Xorp-users] not getting an alias In-Reply-To: <200905011944.25100.irving.lp@feestudiantes.cujae.edu.cu> References: <200904302145.28680.irving.lp@feestudiantes.cujae.edu.cu> <49FB2302.3040007@incunabulum.net> <200905011944.25100.irving.lp@feestudiantes.cujae.edu.cu> Message-ID: <200905071155.31078.irving.lp@feestudiantes.cujae.edu.cu> On Viernes 01 Mayo 2009 07:44:24 PM Irving Leonard escribi?: > On Viernes 01 Mayo 2009 12:27:46 PM Bruce Simpson escribi?: > > Hi, > > > > Can you post your full configuration file, and tell us which kernel > > version are you using, so that others may reproduce? > > #uname -a > Linux poseidon 2.6.24-23-server #1 SMP Thu Feb 5 15:34:58 UTC 2009 i686 > GNU/Linux > ubuntu server 8.04.2 the problem is that xorp doesn't show alias in ifconfig, even when they are active, checked with an nmap ping scan from another machine. But the bigger problem is that I can't see them either from the xorpsh shell with the "show interfaces", I have 31 vifs (vlans) in one interface, and it only shows the first one, also have an alias in the other interface and only see one address in it (this is with xorpsh). In ifconfig it shows one address for each interface, including vlans. -- regards irving VI Conferencia Internacional de Energ?a Renovable, Ahorro de Energ?a y Educaci?n Energ?tica 9 - 12 de Junio 2009, Palacio de las Convenciones ...Por una cultura energ?tica sustentable www.ciercuba.com From syed.khalid at xorp.net Thu May 7 09:16:31 2009 From: syed.khalid at xorp.net (Syed Khalid) Date: Thu, 7 May 2009 09:16:31 -0700 Subject: [Xorp-users] Problem with Xorp passing BSR messages In-Reply-To: <4A02FDC1.1090903@incunabulum.net> References: <1B62E6082D9F034394DB2FBAC03D1A3845F33F8CEF@lrccd-exch05.LRCCD.ad.losrios.edu> <4A02FDC1.1090903@incunabulum.net> Message-ID: Bruce This bug was closed as it appeared to be an Alcatel bug (bug 1071). Syed On Thu, May 7, 2009 at 8:26 AM, Bruce Simpson wrote: > Hi Mark, > > Sorry for my late reply... > > Williams, Mark wrote: > > I will be deploying Xorp on Linux systems soon to handle multicast > routing with PIM-SM. Everything has worked well, but I have had problems > with bootstrap router messages. In my testing, I used two Alcatel-Lucent > routers and an Xorp system. Each router had two interfaces, and were in a > chain in Alcatel - Xorp - Alcatel order. > > > > When the Xorp box was configured as the candidate BSR, the Alcatel > routers both picked up the Xorp box as the BSR after the election time > passed. > > > > It wouldn't surprise me if there were cobwebs in the Cand-BSR > mechanisms, but as Syed has said, it would need to be reproduced in a > lab situation to better understand the problem and how to fix. > > thanks, > BMS > > _______________________________________________ > Xorp-users mailing list > Xorp-users at xorp.org > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090507/6151ad25/attachment.html From bms at incunabulum.net Thu May 7 09:29:00 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 17:29:00 +0100 Subject: [Xorp-users] XORP 1.6 support Click 1.6.0 ? In-Reply-To: References: Message-ID: <4A030C4C.40400@incunabulum.net> dong yin wrote: > Hi all, > > Dose XORP 1.6 support Click v1.6.0? When I run XORP 1.6 to control the > remote Click forwarding data plane (click 1.6.0), errors happened. Short answer: don't know. Looks like some of the element behaviour has changed. I see you are using userland Click, there is probably going to be less churn there. Long answer: Click is a separate project from XORP and the development focus hasn't really included Click. It would be great to have someone who is able to support and maintain the Click integration on an ongoing basis, however I don't think XORP, Inc. has the resources to support the Click integration just at the moment. Click itself is positioned more as a research tool, if folk started using Click in production deployments then I daresay this would change more quickly. For Click itself I can only really speak for the FreeBSD kernel support, which I tried to keep up-to-date (based on Marko Zec's work) but this has bit-rotted somewhat. The patch I gave Eddie Kohler a while back has helped but I think others have had to pitch in and help out since then. thanks, BMS From bms at incunabulum.net Thu May 7 09:31:34 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 17:31:34 +0100 Subject: [Xorp-users] How to implement NAT on xorp In-Reply-To: <879161.74432.qm@web94802.mail.in2.yahoo.com> References: <879161.74432.qm@web94802.mail.in2.yahoo.com> Message-ID: <4A030CE6.8000701@incunabulum.net> naresh raga wrote: > hello Bruce, > What I mean is that the configuration file enough to send a packet > coming from my laptop(address 172.16.0.3 say) will go out from eth2 > to internet with an address of 10.64.25.77 or not? > Both are RFC 1918 addresses. There shouldn't be a problem reaching RFC 1918 prefixes inside your own network(s) if you route them. There should be a 'connected' route for the prefix your laptop is on, as specified in your config file. Does the laptop have a default route pointing to the XORP router? If you are trying to reach the Internet generally, you will have to set up NAT externally from XORP on the machine you're running it on as described in previous reply. thanks, BMS From naresh_raga at yahoo.co.in Thu May 7 09:26:43 2009 From: naresh_raga at yahoo.co.in (naresh raga) Date: Thu, 7 May 2009 21:56:43 +0530 (IST) Subject: [Xorp-users] How to implement NAT on xorp Message-ID: <879161.74432.qm@web94802.mail.in2.yahoo.com> hello Bruce, What I mean is that? the configuration file enough to send a packet coming from my laptop(address 172.16.0.3 say) will go out? from eth2 to internet with an address of 10.64.25.77 or not? Internet-------|eth2|PC|eth3|-----------laptop(I need net on this laptop). With that configuration I am unable to get net on my laptop which is in the network of eth3(PLC LAN).Please can you suggest what should be added to get net on my laptop. Thanks in advance, Naresh. --- On Thu, 7/5/09, Bruce Simpson wrote: From: Bruce Simpson Subject: Re: [Xorp-users] How to implement NAT on xorp To: "naresh raga" Cc: Xorp-users at xorp.org Date: Thursday, 7 May, 2009, 8:27 PM naresh raga wrote: > Hai, > I have installed xorp on a system with two NICs(eth2 and eth3).Internet connection has been given to eth2.I want to get internet on a laptop connected to eth3.I am able to ping eth3(172.16.0.1) from my laptop but unable to get internet.My configuration file is as below: > ... > When I am trying to connect to net on my laptop(connected to eth3),I am unable to connect.But I can ping to eth3.What could be the problem?What more I have to add to this file for implementing getting net on my laptop? > Your config file looks fine. However, to the best of my knowledge, we don't implement NAT in XORP. You will need to configure your system's NAT capability outside of XORP for now, using whatever capabilities your platform for running XORP on has. thanks, BMS Now surf faster and smarter ! Check out the new Firefox 3 - Yahoo! Edition http://downloads.yahoo.com/in/firefox/?fr=om_email_firefox -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090507/ba77fa17/attachment.html From bms at incunabulum.net Thu May 7 09:36:52 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 07 May 2009 17:36:52 +0100 Subject: [Xorp-users] not getting an alias In-Reply-To: <200905071155.31078.irving.lp@feestudiantes.cujae.edu.cu> References: <200904302145.28680.irving.lp@feestudiantes.cujae.edu.cu> <49FB2302.3040007@incunabulum.net> <200905011944.25100.irving.lp@feestudiantes.cujae.edu.cu> <200905071155.31078.irving.lp@feestudiantes.cujae.edu.cu> Message-ID: <4A030E24.3030605@incunabulum.net> Irving Leonard wrote: > ... > the problem is that xorp doesn't show alias in ifconfig, even when they are > active, checked with an nmap ping scan from another machine. But the bigger > problem is that I can't see them either from the xorpsh shell with the "show > interfaces", I have 31 vifs (vlans) in one interface, and it only shows the > first one, also have an alias in the other interface and only see one address > in it (this is with xorpsh). In ifconfig it shows one address for each > interface, including vlans. What is the actual output of 'show interfaces' in xorpsh, and what do you expect to see? To the best of my knowledge the XORP FEA's vlan mechanism is known good and working, the complication here may be the introduction of the bond driver. It's entirely possible that this bond driver does something new that XORP does not expect. I'm afraid I don't understand how the Linux 'bond' driver works -- my OS background is predominantly BSD/Solaris, so I am more up to date with innovations there, but from your description it just sounds like 802.3ad / EtherChannel style trunking. Before we can establish if this is a problem with the code going forward, we need to see more detailed information, i.e. logs of the behaviour. A diagram might be really helpful here, I'm not immediately able to visualize your network topology based on the information provided so far. thanks, BMS From jhay at meraka.org.za Thu May 7 10:42:22 2009 From: jhay at meraka.org.za (John Hay) Date: Thu, 7 May 2009 19:42:22 +0200 Subject: [Xorp-users] rip on freebsd In-Reply-To: <4A02F63E.9050208@incunabulum.net> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> Message-ID: <20090507174221.GA60973@zibbi.meraka.csir.co.za> Hi Bruce, On Thu, May 07, 2009 at 03:54:54PM +0100, Bruce Simpson wrote: > John Hay wrote: > >... > >I have started with an empty conf file and then used xorpsh to do > >the config and saved that. The config + xorp_rtrmgr.log and > >sockstat.out is on http://people.freebsd.org/~jhay/xorp/ > > > > Your config file looks fine. > > I will hazard a wild guess: try adding a default route, using the XORP > static routes module, to your config, and see if this condition is still > present. The wild guess was a good one. I stopped xorp, added a default route and then started xorp again and rip was working. I did it a few times and it started everytime. I also tried with adding a static default route in the xorp config, it does add the route to the kernel, but rip does not work and according to sockstat it does not listen on port 520. Maybe it happens to late? So what now? > > I did a LOT of multicast refactoring in FreeBSD. I don't believe 7.2 > should have the problem I inadvertently introduced in the 6.x train with > default interface lookup, but I would like to be able to rule it out -- > AFAIK it did not bite XORP, and I haven't seen that problem in the 7.x > train since. > > The next place I'd look is ktrace-ing the FEA; use ktrace(8) to pull FEA > activity at runtime, preferably as soon as it starts to run. A shell > script with kill -0 to detect the process has started is fine, it's not > too racy. The libcomm message involved could probably be made more useful. > > Also pull sockstat and check that the socket(s) are in fact being > created. Multicast sockets shouldn't be bound to an interface, but they > do need to be bound to the port (INADDR_ANY, 520). Thanks John -- John Hay -- John.Hay at meraka.csir.co.za / jhay at FreeBSD.org From alejandro.berna at i2cat.net Fri May 8 05:44:28 2009 From: alejandro.berna at i2cat.net (Alejandro Berna) Date: Fri, 08 May 2009 14:44:28 +0200 Subject: [Xorp-users] Problems with IBGP Message-ID: <4A04292C.3060104@i2cat.net> Hi all, I am trying to configure two ASs (65002 and 65000). Every AS for internal routing has configured OSPFv4 and it works properly. Between the two ASBR routers I have configured EBGP exporting the OSPF routes to each one. The peer interfaces of the EBGP are not configured in the internal OSPFv4 protocol of every AS. The two ASBR routers learn all the networks (the external and the internals) and know how to reach to every one. Now I need to send the networks learned thanks to EBGP to the rest of the internal routers of each AS. I need to do it with IBGP (for my objectives I cannot import these routes to the OSPF database). I configure IBGP inside every AS but I find a problem; they learn the routes but the next-hop of every external AS network is the peer interface of the EBGP (these interfaces are not in the OSPF database therefore the internal routers doesn't know how to reach them). I've tried using loopback address for the IBGP peers but with this solution the internal routers doesn't learn any external network. This is my testbed: The two peer interfaces for EBGP are Router 2 eth1 and Router 5 eth3. As I commented before these 2 interfaces are not included in the internal AS OSPF configuration. In order to expose my doubt I show Router 1 and Router 2 configuration as they were the only 2 routers that have IBGP configuration (Router 3 excluded for the IBGP configuration): The configuration file for Router 2 is: interfaces { interface eth1 { vif eth1 { address 10.10.10.10{ prefix-length: 24 disable: false } } } interface eth2 { vif eth2 { address 10.129.0.2 { prefix-length: 24 disable: false } } } interface eth3 { vif eth3 { address 10.1.10.2 { prefix-length: 24 disable: false } } } } fea { unicast-forwarding4 { disable: false } } policy { policy-statement export-routes { term 100 { from { protocol: "connected" } } term 101 { from { protocol: "ospf4" } } term 102 { from { protocol: "bgp" } } } } protocols { ospf4 { router-id: 192.168.0.2 area 0.0.0.0 { interface eth2 { vif eth2 { address 10.129.0.2 { } } } interface eth3 { vif eth3 { address 10.1.10.2 { } } } } } bgp { bgp-id: 10.10.10.10 local-as: 65002 export: "export-routes" peer 10.10.10.20 { local-ip: 10.10.10.10 as: 65000 next-hop: 10.10.10.10 } peer 10.129.0.3 { local-ip: 10.129.0.2 as: 65002 next-hop: 10.129.0.2 } } } And the configuration file for Router 1 is: interfaces { interface eth1 { vif eth1 { address 10.129.0.3 { prefix-length: 24 disable: false } } } interface eth2 { vif eth2 { address 10.9.7.3 { prefix-length: 24 disable: false } } } } fea { unicast-forwarding4 { disable: false } } protocols { ospf4 { router-id: 192.168.0.1 area 0.0.0.1 { interface eth1 { vif eth1 { address 10.129.0.3 { } } } interface eth2 { vif eth2 { address 10.9.7.3 { } } } } } bgp { bgp-id: 10.129.0.3 local-as: 65002 peer 10.129.0.2 { local-ip: 10.129.0.3 as: 65002 next-hop: 10.129.0.3 } } } The bgp route table for Router 2 is: Prefix Nexthop Peer AS Path *> 10.1.10.0/24 10.1.10.2 0.0.0.0 i *> 10.9.7.0/24 10.129.0.3 0.0.0.0 i *> 10.10.10.0/24 10.10.10.10 0.0.0.0 i *> 10.129.0.0/24 10.129.0.2 0.0.0.0 i *> 10.3.7.0/24 10.10.10.20 10.10.10.20 65000 i *> 10.3.9.0/24 10.10.10.20 10.10.10.20 65000 i *> 10.9.9.0/24 10.10.10.20 10.10.10.20 65000 i * 10.10.10.0/24 10.10.10.20 10.10.10.20 65000 i And the bgp route table for Router 1 is: Prefix Nexthop Peer AS Path *> 10.1.10.0/24 10.1.10.2 10.10.10.10 i * 10.3.7.0/24 10.10.10.20 10.10.10.10 65000 i * 10.3.9.0/24 10.10.10.20 10.10.10.10 65000 i *> 10.9.7.0/24 10.129.0.2 10.10.10.10 i * 10.9.9.0/24 10.10.10.20 10.10.10.10 65000 i * 10.10.10.0/24 10.10.10.10 10.10.10.10 i *> 10.129.0.0/24 10.129.0.2 10.10.10.10 i As I commented Router 2 is able to reach any network of both ASs. Router 1 is only able to reach the networks inside the AS because it doesn't know how to reach 10.10.10.0/24 network. How can I fix this? I specified the IBGP next hop for Router 1 as the interface 10.129.0.2 but in the bgp route it appears 10.10.10.20, what am I doing wrong? I hope somebody can help me, thanks a lot, Alejandro Berna From alejandro.berna at i2cat.net Fri May 8 06:03:03 2009 From: alejandro.berna at i2cat.net (Alejandro Berna) Date: Fri, 08 May 2009 15:03:03 +0200 Subject: [Xorp-users] Problems with IBGP In-Reply-To: <4A04292C.3060104@i2cat.net> References: <4A04292C.3060104@i2cat.net> Message-ID: <4A042D87.80106@i2cat.net> Hi again, as I've seen the bgp routes and my testbed map are not good shown in my last mail. I try to fix this: The bgp route table for Router 2 is: Prefix Nexthop Peer AS Path *> 10.1.10.0/24 10.1.10.2 0.0.0.0 i *> 10.9.7.0/24 10.129.0.3 0.0.0.0 i *> 10.10.10.0/24 10.10.10.10 0.0.0.0 i *> 10.129.0.0/24 10.129.0.2 0.0.0.0 i *> 10.3.7.0/24 10.10.10.20 10.10.10.20 65000 i *> 10.3.9.0/24 10.10.10.20 10.10.10.20 65000 i *> 10.9.9.0/24 10.10.10.20 10.10.10.20 65000 i * 10.10.10.0/24 10.10.10.20 10.10.10.20 65000 i And the bgp route table for Router 1 is: Prefix Nexthop Peer AS Path *> 10.1.10.0/24 10.1.10.2 10.10.10.10 i * 10.3.7.0/24 10.10.10.20 10.10.10.10 65000 i * 10.3.9.0/24 10.10.10.20 10.10.10.10 65000 i *> 10.9.7.0/24 10.129.0.2 10.10.10.10 i * 10.9.9.0/24 10.10.10.20 10.10.10.10 65000 i * 10.10.10.0/24 10.10.10.10 10.10.10.10 i *> 10.129.0.0/24 10.129.0.2 10.10.10.10 i Also I attach my testbed picture, sorry for the spamming and the inconvenients, Alejandro -------------- next part -------------- A non-text attachment was scrubbed... Name: testMap.gif Type: image/gif Size: 25731 bytes Desc: not available Url : http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090508/a370b550/attachment-0001.gif From jpark at cs.ucla.edu Fri May 8 07:53:27 2009 From: jpark at cs.ucla.edu (Jonathan Park) Date: Fri, 8 May 2009 07:53:27 -0700 Subject: [Xorp-users] route flap damping triggers on eBGP duplicate updates? Message-ID: Hi Xorp users, I was looking at XORP code and found that route flap damping can be triggered by eBGP duplicate updates (i.e. all attributes are the same), giving them the same penalty value of 1000. Is that true? If so, is it done intentionally? Thanks, Jonathan From jhay at meraka.org.za Sat May 9 07:22:40 2009 From: jhay at meraka.org.za (John Hay) Date: Sat, 9 May 2009 16:22:40 +0200 Subject: [Xorp-users] more rip on freebsd Message-ID: <20090509142240.GA19119@zibbi.meraka.csir.co.za> Hi, I am using FreeBSD amd64 7.2-RELEASE. I am using xorp-1.6 from the 7.2 packages. I notice that xorp rip announce the connected subnets with a metric of 0. Should that not be 1? Most other RIP daemons will ignore routes with a metric of 0. Reading RFC 2453 section 3.5 it sounds like 0 cannot be used. It should be a value of 1 to 15. My config is still the same and available at: http://people.freebsd.org/~jhay/xorp/xorp.conf John -- John Hay -- John.Hay at meraka.csir.co.za / jhay at FreeBSD.org From syed.khalid at xorp.net Sat May 9 11:17:08 2009 From: syed.khalid at xorp.net (Syed Khalid) Date: Sat, 9 May 2009 11:17:08 -0700 Subject: [Xorp-users] Problems with IBGP In-Reply-To: <4A042D87.80106@i2cat.net> References: <4A04292C.3060104@i2cat.net> <4A042D87.80106@i2cat.net> Message-ID: Hello Could you open a bug for this and we will reproduce it? Syed On Fri, May 8, 2009 at 6:03 AM, Alejandro Berna wrote: > Hi again, as I've seen the bgp routes and my testbed map are not good shown > in my last mail. I try to fix this: > > The bgp route table for Router 2 is: > Prefix Nexthop Peer AS Path > *> 10.1.10.0/24 10.1.10.2 0.0.0.0 i > *> 10.9.7.0/24 10.129.0.3 0.0.0.0 i > *> 10.10.10.0/24 10.10.10.10 0.0.0.0 i > *> 10.129.0.0/24 10.129.0.2 0.0.0.0 i > *> 10.3.7.0/24 10.10.10.20 10.10.10.20 65000 i > *> 10.3.9.0/24 10.10.10.20 10.10.10.20 65000 i > *> 10.9.9.0/24 10.10.10.20 10.10.10.20 65000 i > * 10.10.10.0/24 10.10.10.20 10.10.10.20 65000 i > > And the bgp route table for Router 1 is: > > > Prefix Nexthop Peer AS Path > *> 10.1.10.0/24 10.1.10.2 10.10.10.10 i > * 10.3.7.0/24 10.10.10.20 10.10.10.10 65000 i > * 10.3.9.0/24 10.10.10.20 10.10.10.10 65000 i > *> 10.9.7.0/24 10.129.0.2 10.10.10.10 i > * 10.9.9.0/24 10.10.10.20 10.10.10.10 65000 i > * 10.10.10.0/24 10.10.10.10 10.10.10.10 i > *> 10.129.0.0/24 10.129.0.2 10.10.10.10 i > > > Also I attach my testbed picture, sorry for the spamming and the > inconvenients, > > > Alejandro > > > > > _______________________________________________ > Xorp-users mailing list > Xorp-users at xorp.org > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090509/754a73ce/attachment.html From bms at incunabulum.net Sat May 9 11:42:51 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sat, 09 May 2009 19:42:51 +0100 Subject: [Xorp-users] rip on freebsd In-Reply-To: <20090507174221.GA60973@zibbi.meraka.csir.co.za> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> <20090507174221.GA60973@zibbi.meraka.csir.co.za> Message-ID: <4A05CEAB.6040605@incunabulum.net> John Hay wrote: > The wild guess was a good one. I stopped xorp, added a default route > and then started xorp again and rip was working. I did it a few times > and it started everytime. > > OK. I am trying to think what the problem could be. It's been many many months on since I touched stuff outside of XORP which could have affected this. > I also tried with adding a static default route in the xorp config, it > does add the route to the kernel, but rip does not work and according > to sockstat it does not listen on port 520. Maybe it happens to late? > > So what now? > Can you confirm that the sockets actually exist at this point? Use 'sockstat -4 | grep 520' to confirm. Can you confirm that the box has a multicast membership for RIP on the interface(s) where you expect to see them? Use 'netstat -g' in FreeBSD versions 5.x-7.x to confirm, or better still, 'ifmcstat'. This is still stabbing in the dark, I do not know why the FEA is saying the RIP sockets don't exist at this point. This is the interesting message: [ 2009/05/07 12:10:42 ERROR xorp_fea:865 LIBXORP +714 asyncio.cc complete_transfer ] Write error 51 51 is ENETUNREACH (grep 51 /usr/include/errno.h). Presumably the FEA closes the socket when it hits this error. This should be made verbose in asyncio.cc by calling strerror() or similar on POSIX platforms, patches welcome. As I mentioned in my previous reply, 'ktrace' is pretty much needed to find out exactly what the FEA is doing when this error is hit -- and where the sendto() is going. Because XORP processes are children of the Router Manager, you will need to intercept the FEA being started. You can do this by hand if you have good reaction times, or just write a script to do it. I believe I had a script somewhere to jump in and trap XORP process creation with gdb, but I'd have to hunt for it. ... Are your interfaces configured when you run XORP, or do you rely wholly on XORP to configure your interfaces? ... What is the output of 'ifconfig -a' before, after, and during XORP run time? ... Are your interfaces UP, RUNNING and MULTICAST? I usually test with msk(4) myself, and haven't seen issues like this, although I haven't done in-depth testing since the XORP 1.5 release cycle. Based on the information to date, and in the absence of reproducing the issue, my best guess is that this is a possible initialization race between the FEA and RIP modules. I know that XORP is still using the old Steve Deering era IP_ADD_MEMBERSHIP socket options for multicast, which whilst it is reasonably portable, has dire problems if you are lacking IPv4 addresses on the link(s) you want to use. There seems to be some misunderstanding about what multicast is and how it is expected to work out there, and it just plain breaks (in any stack) if certain steps aren't followed. It hasn't been an issue with XORP, because it does not currently support the equivalent of 'ip unnumbered' -- and there are still a number of places in the code which assume each interface has an IPv4 network layer address of some kind, be that private, link-scope or whatever. With IPv6, issues of the kind seen with the IPv4 basic multicasting API, and IGMP issues, just don't exist, as link local addresses are normally always available, except during DAD. thanks, BMS From bms at incunabulum.net Sat May 9 11:50:51 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sat, 09 May 2009 19:50:51 +0100 Subject: [Xorp-users] route flap damping triggers on eBGP duplicate updates? In-Reply-To: References: Message-ID: <4A05D08B.2040105@incunabulum.net> Hi Jonathan, Have you actually observed this happening, or is this purely an observation based on code inspection? Jonathan Park wrote: > Hi Xorp users, > > I was looking at XORP code and found that route flap damping can be > triggered by eBGP duplicate updates (i.e. all attributes are the same), > giving them the same penalty value of 1000. > > Is that true? If so, is it done intentionally? > It would be really helpful, when reporting this sort of problem on the list, if you could illustrate the call graph(s) involved. I find KScope is an excellent tool for doing this sort of work; it can export call graphs via GraphViz, I believe there may be text export for call trees in 1 direction also. So, 1000 is the FIXED value in damping.cc. Other than that, I haven't had time to read into this further. Can you elaborate further on how you believe this condition is being triggered? thanks, BMS From jpark at cs.ucla.edu Sat May 9 13:05:04 2009 From: jpark at cs.ucla.edu (Jonathan Park) Date: Sat, 09 May 2009 13:05:04 -0700 Subject: [Xorp-users] route flap damping triggers on eBGP duplicate updates? In-Reply-To: <4A05D08B.2040105@incunabulum.net> References: <4A05D08B.2040105@incunabulum.net> Message-ID: <1241899504.11741.109.camel@Rabbit.CS.UCLA.EDU> Thanks Bruce for the reply. On Sat, 2009-05-09 at 19:50 +0100, Bruce Simpson wrote: > Hi Jonathan, > > Have you actually observed this happening, or is this purely an > observation based on code inspection? > It was based on code inspection. Duplicating this needed some non-trivial test setup (requiring me to replay duplicate bgp updates). So I looked at the code, and it looked like it is giving the penalty. I was just wondering if someone already looked into this or what was the initial design. > Jonathan Park wrote: > > Hi Xorp users, > > > > I was looking at XORP code and found that route flap damping can be > > triggered by eBGP duplicate updates (i.e. all attributes are the same), > > giving them the same penalty value of 1000. > > > > Is that true? If so, is it done intentionally? > > > > It would be really helpful, when reporting this sort of problem on the > list, if you could illustrate the call graph(s) involved. > I find KScope is an excellent tool for doing this sort of work; it can > export call graphs via GraphViz, I believe there may be text export for > call trees in 1 direction also. > > So, 1000 is the FIXED value in damping.cc. Other than that, I haven't > had time to read into this further. Can you elaborate further on how you > believe this condition is being triggered? > I think I will do what you suggested and will get back to you with more details. Thanks, Jonathan > thanks, > BMS From jpark at cs.ucla.edu Sun May 10 00:15:56 2009 From: jpark at cs.ucla.edu (Jonathan Park) Date: Sun, 10 May 2009 00:15:56 -0700 Subject: [Xorp-users] route flap damping triggers on eBGPduplicate updates? In-Reply-To: <1241899504.11741.109.camel@Rabbit.CS.UCLA.EDU> References: <4A05D08B.2040105@incunabulum.net> <1241899504.11741.109.camel@Rabbit.CS.UCLA.EDU> Message-ID: <055CAB2D97D947E89276A7D1660524EA@minirabbit> Hi Bruce, I've just tested it, and it *does* penalize duplicates. In fact as you said, it penalizes *all* updates with penalty = 1000. Route flap damping RFC specifies that there should be a penalty for each flap. >From my observation, different implementations interpret "flap" in different ways. Some implementations penalize all updates (like XORP), and some implementation (Quagga) penalize an announcement following a withdrawal. Cisco penalizes withdrawals and updates with attribute changes but not duplicates. Juniper penalizes all updates but not duplicates. I was wondering whether this is a bug or it is by design. Thanks, Jonathan > -----Original Message----- > From: xorp-users-bounces at xorp.org [mailto:xorp-users-bounces at xorp.org] On > Behalf Of Jonathan Park > Sent: Saturday, May 09, 2009 1:05 PM > To: Bruce Simpson > Cc: xorp-users at xorp.org > Subject: Re: [Xorp-users] route flap damping triggers on eBGPduplicate > updates? > > Thanks Bruce for the reply. > > On Sat, 2009-05-09 at 19:50 +0100, Bruce Simpson wrote: > > Hi Jonathan, > > > > Have you actually observed this happening, or is this purely an > > observation based on code inspection? > > > > It was based on code inspection. Duplicating this needed some > non-trivial test setup (requiring me to replay duplicate bgp updates). > So I looked at the code, and it looked like it is giving the penalty. > I was just wondering if someone already looked into this or what was the > initial design. > > > > Jonathan Park wrote: > > > Hi Xorp users, > > > > > > I was looking at XORP code and found that route flap damping can be > > > triggered by eBGP duplicate updates (i.e. all attributes are the same), > > > giving them the same penalty value of 1000. > > > > > > Is that true? If so, is it done intentionally? > > > > > > > It would be really helpful, when reporting this sort of problem on the > > list, if you could illustrate the call graph(s) involved. > > I find KScope is an excellent tool for doing this sort of work; it can > > export call graphs via GraphViz, I believe there may be text export for > > call trees in 1 direction also. > > > > So, 1000 is the FIXED value in damping.cc. Other than that, I haven't > > had time to read into this further. Can you elaborate further on how you > > believe this condition is being triggered? > > > > I think I will do what you suggested and will get back to you with more > details. > > Thanks, > > Jonathan > > > > thanks, > > BMS > > _______________________________________________ > Xorp-users mailing list > Xorp-users at xorp.org > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users From bms at incunabulum.net Sun May 10 06:50:23 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sun, 10 May 2009 14:50:23 +0100 Subject: [Xorp-users] route flap damping triggers on eBGPduplicate updates? In-Reply-To: <055CAB2D97D947E89276A7D1660524EA@minirabbit> References: <4A05D08B.2040105@incunabulum.net> <1241899504.11741.109.camel@Rabbit.CS.UCLA.EDU> <055CAB2D97D947E89276A7D1660524EA@minirabbit> Message-ID: <4A06DB9F.90403@incunabulum.net> Jonathan Park wrote: > Route flap damping RFC specifies that there should be a penalty for each > flap. > > From my observation, different implementations interpret "flap" in different > ways. Some implementations penalize all updates (like XORP), and some > implementation (Quagga) penalize an announcement following a withdrawal. > Cisco penalizes withdrawals and updates with attribute changes but not > duplicates. Juniper penalizes all updates but not duplicates. > > I was wondering whether this is a bug or it is by design. > I haven't read that RFC, but it does sound like there is ambiguity and room for interpretation. To my mind, BGP has become something of a 'kitchen sink' protocol, I was relieved when the community finally came up with a single RFC as a normative reference. I believe route flap dampening was introduced between XORP releases 1.5 and 1.6. Off the top of my head, I believe Mark Handley wrote the code, he may know more (Cc'd). thanks, BMS From jjw0927 at 126.com Sun May 10 18:58:36 2009 From: jjw0927 at 126.com (jjw0927) Date: Mon, 11 May 2009 09:58:36 +0800 (CST) Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? Message-ID: <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> Hi all, I have installed XORP-1.6-RC and Net-SNMP-5.4.2.1. And Both of them have been compiled. I have configured the Net-snmp and XORP according the files of "${XORP}/mibs/snmpdscripts/README" and "XORP SNMP Agent 1.6.pdf". And now I want to launch Net-SNMP via the router manager, but it failed. Below is the output when I run the command "sudo ./xorp_rtrmgr -b config.boot" in terminal: [ 2009/05/11 09:27:59 INFO xorp_rtrmgr:3574 RTRMGR +249 master_conf_tree.cc execute ] Changed modules: xorp_if_mib [ 2009/05/11 09:27:59 INFO xorp_rtrmgr:3574 RTRMGR +101 module_manager.cc execute ] Executing module: xorp_if_mib (mibs/snmpdscripts/startsnmp) No log handling enabled - turning on stderr logging registered debug token xorp_if_mib_module, 1 registered debug token dlmod, 1 registered debug token usmUser, 1 dlmod: register mib dlmod: dlmod_path: /usr/local/lib/snmp/dlmod dlmod: dlmod_create_module dlmod: dlmod_load_module xorp_if_mib_module: /usr/local/src/xorp-1.6-RC/mibs/.libs/xorp_if_mib_module.so xorp_if_mib_module: XorpIfMib created xorp_if_mib_module: Initialized... Turning on AgentX master support. NET-SNMP version 5.4.2.1 dlmod: dlmod_create_module dlmod: dlmod_load_module bgp4_mib_1657: /usr/local/xorp/mibs/.libs/bgp4_mib_1657.so [ 2009/05/11 09:28:01 WARNING snmpd XrlXorpIfMibTarget ] Handling method for xorp_if_mib/0.1/load_mib failed: XrlCmdError 102 Command failed [ 2009/05/11 09:28:01 ERROR xorp_rtrmgr:3574 RTRMGR +691 master_conf_tree.cc commit_pass2_done ] Commit failed: 102 Command failed [ 2009/05/11 09:28:01 ERROR xorp_rtrmgr:3574 RTRMGR +261 master_conf_tree.cc config_done ] Configuration failed: 102 Command failed [ 2009/05/11 09:28:01 INFO xorp_rtrmgr:3574 RTRMGR +2233 task.cc run_task ] No more tasks to run [ 2009/05/11 09:28:01 INFO xorp_rtrmgr:3574 RTRMGR +176 module_manager.cc terminate ] Terminating module: xorp_if_mib [ 2009/05/11 09:28:01 INFO xorp_rtrmgr:3574 RTRMGR +199 module_manager.cc terminate ] Killing module: xorp_if_mib Received TERM or STOP signal... shutting down... No log handling enabled - turning on stderr logging xorp_if_mib_module: Unloaded... xorp_if_mib_module: XorpIfMib destroyed dlmod: Module xorp_if_mib_module unloaded [ 2009/05/11 09:28:01 ERROR xorp_rtrmgr:3574 RTRMGR +754 module_manager.cc done_cb ]Command "/usr/local/src/xorp-1.6-RC/mibs/snmpdscripts/startsnmp": terminated with signal 15. [ 2009/05/11 09:28:01 INFO xorp_rtrmgr:3574 RTRMGR +287 module_manager.cc module_exited ] Module killed during shutdown: xorp_if_mib And Below is the content of "config.boot" I used to run rtrmgr: protocols { snmp { mib-module bgp4_mib_1657 { abs-path: "/usr/local/xorp/mibs/.libs/bgp4_mib_1657.so" } } } I do not know what is wrong in my configuration, your help is appreciated for me. Thank you very much! Best Regard Jianwu Jin -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090511/b2c843c7/attachment.html From alejandro.berna at i2cat.net Fri May 8 05:56:19 2009 From: alejandro.berna at i2cat.net (Alejandro Berna) Date: Fri, 08 May 2009 14:56:19 +0200 Subject: [Xorp-users] Problems with IBGP In-Reply-To: <4A04292C.3060104@i2cat.net> References: <4A04292C.3060104@i2cat.net> Message-ID: <4A042BF3.7060704@i2cat.net> Hi again, as I've seen the bgp routes and my testbed map are not good shown in my last mail. I try to fix this: The bgp route table for Router 2 is: Prefix Nexthop Peer AS Path *> 10.1.10.0/24 10.1.10.2 0.0.0.0 i *> 10.9.7.0/24 10.129.0.3 0.0.0.0 i *> 10.10.10.0/24 10.10.10.10 0.0.0.0 i *> 10.129.0.0/24 10.129.0.2 0.0.0.0 i *> 10.3.7.0/24 10.10.10.20 10.10.10.20 65000 i *> 10.3.9.0/24 10.10.10.20 10.10.10.20 65000 i *> 10.9.9.0/24 10.10.10.20 10.10.10.20 65000 i * 10.10.10.0/24 10.10.10.20 10.10.10.20 65000 i And the bgp route table for Router 1 is: Prefix Nexthop Peer AS Path *> 10.1.10.0/24 10.1.10.2 10.10.10.10 i * 10.3.7.0/24 10.10.10.20 10.10.10.10 65000 i * 10.3.9.0/24 10.10.10.20 10.10.10.10 65000 i *> 10.9.7.0/24 10.129.0.2 10.10.10.10 i * 10.9.9.0/24 10.10.10.20 10.10.10.10 65000 i * 10.10.10.0/24 10.10.10.10 10.10.10.10 i *> 10.129.0.0/24 10.129.0.2 10.10.10.10 i Also I attach my testbed picture, sorry for the spamming and the inconvenients, Alejandro Alejandro Berna escribi?: > Hi all, > > I am trying to configure two ASs (65002 and 65000). Every AS for > internal routing has configured OSPFv4 and it works properly. Between > the two ASBR routers I have configured EBGP exporting the OSPF routes to > each one. The peer interfaces of the EBGP are not configured in the > internal OSPFv4 protocol of every AS. The two ASBR routers learn all the > networks (the external and the internals) and know how to reach to every > one. Now I need to send the networks learned thanks to EBGP to the rest > of the internal routers of each AS. I need to do it with IBGP (for my > objectives I cannot import these routes to the OSPF database). I > configure IBGP inside every AS but I find a problem; they learn the > routes but the next-hop of every external AS network is the peer > interface of the EBGP (these interfaces are not in the OSPF database > therefore the internal routers doesn't know how to reach them). I've > tried using loopback address for the IBGP peers but with this solution > the internal routers doesn't learn any external network. This is my testbed: > > > > > The two peer interfaces for EBGP are Router 2 eth1 and Router 5 eth3. As > I commented before these 2 interfaces are not included in the internal > AS OSPF configuration. In order to expose my doubt I show Router 1 and > Router 2 configuration as they were the only 2 routers that have IBGP > configuration (Router 3 excluded for the IBGP configuration): > > The configuration file for Router 2 is: > > interfaces { > interface eth1 { > vif eth1 { > address 10.10.10.10{ > prefix-length: 24 > disable: false > } > } > } > interface eth2 { > vif eth2 { > address 10.129.0.2 { > prefix-length: 24 > disable: false > } > } > } > interface eth3 { > vif eth3 { > address 10.1.10.2 { > prefix-length: 24 > disable: false > } > } > } > } > > fea { > unicast-forwarding4 { > disable: false > } > } > > policy { > policy-statement export-routes { > term 100 { > from { > protocol: "connected" > } > } > term 101 { > from { > protocol: "ospf4" > } > } > term 102 { > from { > protocol: "bgp" > } > } > } > } > > protocols { > ospf4 { > router-id: 192.168.0.2 > > area 0.0.0.0 { > > interface eth2 { > vif eth2 { > address 10.129.0.2 { > > } > } > } > interface eth3 { > vif eth3 { > address 10.1.10.2 { > } > } > } > } > } > bgp { > bgp-id: 10.10.10.10 > local-as: 65002 > export: "export-routes" > peer 10.10.10.20 { > local-ip: 10.10.10.10 > as: 65000 > next-hop: 10.10.10.10 > } > peer 10.129.0.3 { > local-ip: 10.129.0.2 > as: 65002 > next-hop: 10.129.0.2 > } > } > } > > > And the configuration file for Router 1 is: > > > interfaces { > interface eth1 { > vif eth1 { > address 10.129.0.3 { > prefix-length: 24 > disable: false > } > } > } > interface eth2 { > vif eth2 { > address 10.9.7.3 { > prefix-length: 24 > disable: false > } > } > } > } > > fea { > unicast-forwarding4 { > disable: false > } > } > > protocols { > ospf4 { > router-id: 192.168.0.1 > area 0.0.0.1 { > interface eth1 { > vif eth1 { > address 10.129.0.3 { > } > } > } > interface eth2 { > vif eth2 { > address 10.9.7.3 { > } > } > } > } > } > bgp { > bgp-id: 10.129.0.3 > local-as: 65002 > peer 10.129.0.2 { > local-ip: 10.129.0.3 > as: 65002 > next-hop: 10.129.0.3 > } > } > } > > The bgp route table for Router 2 is: > > Prefix Nexthop > Peer AS Path > *> 10.1.10.0/24 10.1.10.2 > 0.0.0.0 i > *> 10.9.7.0/24 10.129.0.3 > 0.0.0.0 i > *> 10.10.10.0/24 10.10.10.10 > 0.0.0.0 i > *> 10.129.0.0/24 10.129.0.2 > 0.0.0.0 i > *> 10.3.7.0/24 10.10.10.20 > 10.10.10.20 65000 i > *> 10.3.9.0/24 10.10.10.20 > 10.10.10.20 65000 i > *> 10.9.9.0/24 10.10.10.20 > 10.10.10.20 65000 i > * 10.10.10.0/24 10.10.10.20 > 10.10.10.20 65000 i > > And the bgp route table for Router 1 is: > > > Prefix Nexthop > Peer AS Path > *> 10.1.10.0/24 10.1.10.2 > 10.10.10.10 i > * 10.3.7.0/24 10.10.10.20 > 10.10.10.10 65000 i > * 10.3.9.0/24 10.10.10.20 > 10.10.10.10 65000 i > *> 10.9.7.0/24 10.129.0.2 > 10.10.10.10 i > * 10.9.9.0/24 10.10.10.20 > 10.10.10.10 65000 i > * 10.10.10.0/24 10.10.10.10 > 10.10.10.10 i > *> 10.129.0.0/24 10.129.0.2 > 10.10.10.10 i > > > As I commented Router 2 is able to reach any network of both ASs. Router > 1 is only able to reach the networks inside the AS because it doesn't > know how to reach 10.10.10.0/24 network. How can I fix this? I specified > the IBGP next hop for Router 1 as the interface 10.129.0.2 but in the > bgp route it appears 10.10.10.20, what am I doing wrong? I hope somebody > can help me, thanks a lot, > > > Alejandro Berna > > > > > _______________________________________________ > Xorp-users mailing list > Xorp-users at xorp.org > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > > -- Alejandro Berna Juan Network Technologies Cluster (CTX) Fundaci? i2Cat, Internet i Innovaci? Digital a Catalunya C/ Gran Capit? 2 - 4, Nexus I building, 2nd floor, office 203 08034 Barcelona, Catalonia, Spain P: +34 93 553 2522 F: +34 93 553 2520 -------------- next part -------------- A non-text attachment was scrubbed... Name: testMap.png Type: image/png Size: 57733 bytes Desc: not available Url : http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090508/2a2002c3/attachment-0001.bin From jhay at meraka.org.za Mon May 11 13:42:49 2009 From: jhay at meraka.org.za (John Hay) Date: Mon, 11 May 2009 22:42:49 +0200 Subject: [Xorp-users] rip on freebsd In-Reply-To: <4A05CEAB.6040605@incunabulum.net> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> <20090507174221.GA60973@zibbi.meraka.csir.co.za> <4A05CEAB.6040605@incunabulum.net> Message-ID: <20090511204249.GA334@zibbi.meraka.csir.co.za> On Sat, May 09, 2009 at 07:42:51PM +0100, Bruce Simpson wrote: > John Hay wrote: > >The wild guess was a good one. I stopped xorp, added a default route > >and then started xorp again and rip was working. I did it a few times > >and it started everytime. > > > > > > OK. I am trying to think what the problem could be. It's been many many > months on since I touched stuff outside of XORP which could have > affected this. > > > >I also tried with adding a static default route in the xorp config, it > >does add the route to the kernel, but rip does not work and according > >to sockstat it does not listen on port 520. Maybe it happens to late? > > > >So what now? > > > > Can you confirm that the sockets actually exist at this point? > Use 'sockstat -4 | grep 520' to confirm. If I there is a default route when I start xorp, the sockets exist. If there is not a default route when i start xorp, the socket does not exist. > > Can you confirm that the box has a multicast membership for RIP on the > interface(s) where you expect to see them? > Use 'netstat -g' in FreeBSD versions 5.x-7.x to confirm, or better > still, 'ifmcstat'. This is still stabbing in the dark, I do not know > why the FEA is saying the RIP sockets don't exist at this point. The group 224.0.0.9 does not show up if I start xorp without a default route already installed. > > This is the interesting message: > > [ 2009/05/07 12:10:42 ERROR xorp_fea:865 LIBXORP +714 asyncio.cc > complete_transfer ] Write error 51 > > 51 is ENETUNREACH (grep 51 /usr/include/errno.h). Presumably the FEA closes > the socket when it hits this error. This should be made verbose in > asyncio.cc by calling strerror() or similar on POSIX platforms, patches > welcome. > > As I mentioned in my previous reply, 'ktrace' is pretty much needed to > find out exactly what the FEA is doing when this error is hit -- and where > the sendto() is going. Because XORP processes are children of the Router > Manager, you will need to intercept the FEA being started. You can do this > by hand if you have good reaction times, or just write a script to do it. I > believe I had a script somewhere to jump in and trap XORP process creation > with gdb, but I'd have to hunt for it. I have not tried to ktrace it yet. > ... Are your interfaces configured when you run XORP, or do you rely > wholly on XORP to configure your interfaces? The interfaces are configured before I run xorp. > ... What is the output of 'ifconfig -a' before, after, and during XORP > run time? # ifconfig -a msk0: flags=8843 metric 1 mtu 1500 options=11a ether 00:21:85:57:f9:0c inet 146.64.5.3 netmask 0xffffff00 broadcast 146.64.5.255 media: Ethernet autoselect (100baseTX ) status: active msk1: flags=8843 metric 1 mtu 1500 options=11a ether 00:21:85:57:f9:0d inet 10.250.250.1 netmask 0xffffff00 broadcast 10.250.250.255 media: Ethernet autoselect (100baseTX ) status: active lo0: flags=8049 metric 0 mtu 16384 inet6 fe80::1%lo0 prefixlen 64 scopeid 0x3 inet6 ::1 prefixlen 128 inet 127.0.0.1 netmask 0xff000000 # It does not seem to change during or after. > ... Are your interfaces UP, RUNNING and MULTICAST? Yes > I usually test with msk(4) myself, and haven't seen issues like this, > although I haven't done in-depth testing since the XORP 1.5 release cycle. > > Based on the information to date, and in the absence of reproducing the > issue, my best guess is that this is a possible initialization race > between the FEA and RIP modules. So if you start xorp/rip on a freebsd box with interfaces configured, but without routes allready installed, it works? > I know that XORP is still using the old Steve Deering era > IP_ADD_MEMBERSHIP socket options for multicast, which whilst it is > reasonably portable, has dire problems if you are lacking IPv4 addresses > on the link(s) you want to use. There seems to be some misunderstanding > about what multicast is and how it is expected to work out there, and it > just plain breaks (in any stack) if certain steps aren't followed. > > It hasn't been an issue with XORP, because it does not currently support > the equivalent of 'ip unnumbered' -- and there are still a number of > places in the code which assume each interface has an IPv4 network layer > address of some kind, be that private, link-scope or whatever. > > With IPv6, issues of the kind seen with the IPv4 basic multicasting API, > and IGMP issues, just don't exist, as link local addresses are normally > always available, except during DAD. > > thanks, > BMS John -- John Hay -- John.Hay at meraka.csir.co.za / jhay at FreeBSD.org From bms at incunabulum.net Mon May 11 22:02:47 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 12 May 2009 06:02:47 +0100 Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? In-Reply-To: <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> References: <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> Message-ID: <4A0902F7.8030707@incunabulum.net> Hi, I'm not fully up to date with how SNMP support in XORP may have changed since the 1.5 release. I understand XORP has since cut over to using shared libraries in the build by default. However I have a few questions: * Which OS platform are you running XORP on? I need to know this because the runtime dynamic linker behaviour differs across platforms. * Are you running XORP after it has been installed using 'gmake install', or are you running XORP from a source tree checkout? I need to know this because it looks as though you are attempting to use XORP's SNMP MIB support without installing XORP. jjw0927 wrote: > Hi all, > > I have installed XORP-1.6-RC and Net-SNMP-5.4.2.1. And Both of them > have been compiled. > I have configured the Net-snmp and XORP according the files of > "${XORP}/mibs/snmpdscripts/README" and "XORP SNMP Agent 1.6.pdf". > > And now I want to launch Net-SNMP via the router manager, but it > failed. Below is the output when I run the command "sudo ./xorp_rtrmgr > -b config.boot" in terminal: Based on the log output and configuration file in your post, it looks as though you are attempting to load the MIB module into the Net-SNMP daemon process without using 'gmake install' first to install XORP on your system. I don't believe this is a use case we can support, because shared objects found in a .libs directory need to be re-linked by libtool before installation -- they should not be used directly. So it is highly likely that the MIB module you are attempting to load into Net-SNMP has missing runtime library dependencies. To confirm this, you'd need to use the 'ldd' command on the /usr/local/xorp/mibs/.libs/bgp4_mib_1657.so file. However, not all platforms support listing shared library dependencies directly, which is why I need to know what platform you are running XORP on when reporting problems of this kind, it is essential information. thanks, BMS From bms at incunabulum.net Mon May 11 22:10:31 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 12 May 2009 06:10:31 +0100 Subject: [Xorp-users] rip on freebsd In-Reply-To: <20090511204249.GA334@zibbi.meraka.csir.co.za> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> <20090507174221.GA60973@zibbi.meraka.csir.co.za> <4A05CEAB.6040605@incunabulum.net> <20090511204249.GA334@zibbi.meraka.csir.co.za> Message-ID: <4A0904C7.6080505@incunabulum.net> John Hay wrote: > ... >> >> As I mentioned in my previous reply, 'ktrace' is pretty much needed to >> find out exactly what the FEA is doing when this error is hit -- and where >> the sendto() is going. Because XORP processes are children of the Router >> Manager, you will need to intercept the FEA being started. You can do this >> by hand if you have good reaction times, or just write a script to do it. I >> believe I had a script somewhere to jump in and trap XORP process creation >> with gdb, but I'd have to hunt for it. >> > > I have not tried to ktrace it yet. > I will dig up this script and modify it to use ktrace instead of gdb; it is essential that the problem gets pinned down to a particular set of socket syscalls. The ENETUNREACH error on a socket can only come from a very limited set of call paths in the FreeBSD kernel. > So if you start xorp/rip on a freebsd box with interfaces configured, but > without routes allready installed, it works? > I haven't tested this directly since the 1.5 release. My desktop machine has a single msk(4) interface and has 7.2-RELEASE installed from source, which should make it possible to exercise it once I have some time to reproduce the issue. thanks, BMS From jhay at meraka.org.za Mon May 11 23:28:20 2009 From: jhay at meraka.org.za (John Hay) Date: Tue, 12 May 2009 08:28:20 +0200 Subject: [Xorp-users] rip and different policies for different interfaces Message-ID: <20090512062820.GA37448@zibbi.meraka.csir.co.za> Hi, I have been reading some and trying stuff, but I seem to be stuck a bit. I have found that it is possible to configure stuff that give an error when you commit it and then something seems to get stuck and it is impossible to get rid of it except by restarting xorp. :-) Is it possible to have different "export" policies on different interfaces for rip? I have two cases. A static default route that I want to propagate on some interfaces, but not on others. Some routes that I learn, I do not want to propagate on a specific interface. I had a look at "to" inside "term", neighbor is almost, but with rip multicast is used, so it does not quite fit. I guess I'm looking for the functionality of the quagga router rip ... distribute-list prefix csir-prot out em1 distribute-list prefix remote-prot out vlan1 John -- John Hay -- John.Hay at meraka.csir.co.za / jhay at FreeBSD.org From bms at incunabulum.net Tue May 12 00:26:06 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 12 May 2009 08:26:06 +0100 Subject: [Xorp-users] rip on freebsd In-Reply-To: <4A0904C7.6080505@incunabulum.net> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> <20090507174221.GA60973@zibbi.meraka.csir.co.za> <4A05CEAB.6040605@incunabulum.net> <20090511204249.GA334@zibbi.meraka.csir.co.za> <4A0904C7.6080505@incunabulum.net> Message-ID: <4A09248E.20004@incunabulum.net> Bruce Simpson wrote: > I will dig up this script and modify it to use ktrace instead of gdb; it > is essential that the problem gets pinned down to a particular set of > socket syscalls. > Here's that script. This is intended for trapping the creation of a XORP child process and hooking GDB up to it, I think I wrote it to debug a DoS in the XRL input path, it is probably in Bugzilla somewhere also. As written here, it is intended for use under Gentoo Linux. To modify it to ktrace a FreeBSD process, do the following: * Uncomment the 'if' conditional inside the 'until' loop, and uncomment the 'last=!?' line. * Uncomment the line containing 'killall -s -0'. This just uses 'killall' to find out the process ID of whatever process name you pass to the script. * Comment out the line containing 'pidof' and uncomment the line above that (pidof is linux-specific, and we need the last field output which is the process ID). * Comment out the 'exec' line(s) at the end, and just rewrite that to do 'exec ktrace -f /tmp/xorp_fea.ktrace -p $pid'. * Comment 'sleep 0.2', uncomment 'sleep 1' -- FreeBSD's sleep is strictly POSIX and won't sleep for less than a second. Pass xorp_fea as the command line parameter to this script. Run at the same time as when you run xorp_rtrmgr to start XORP in another shell. This can be racy, remember this shell will only wake up every second to check if the process is now there. Using 'kdump -E -r -f /tmp/xorp_fea.ktrace' should shed more light on what syscalls the FEA is invoking when the multicast socket failure is encountered. kdump knows how to decode some setsockopt() calls in a human-readable format, but not all. Alternatively, you could just ktrace xorp_rtrmgr and all children; this will not race xorp_fea's creation, but will produce LOTS of output. You could also hack the FEA's template file to run xorp_fea under ktrace directly, which is how I usually ran valgrind to find various QA issues in XORP. thanks, BMS -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: trapgdb.sh Url: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090512/65a219f1/attachment.ksh From bms at incunabulum.net Tue May 12 00:56:16 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 12 May 2009 08:56:16 +0100 Subject: [Xorp-users] rip and different policies for different interfaces In-Reply-To: <20090512062820.GA37448@zibbi.meraka.csir.co.za> References: <20090512062820.GA37448@zibbi.meraka.csir.co.za> Message-ID: <4A092BA0.9090409@incunabulum.net> John Hay wrote: > Is it possible to have different "export" policies on different interfaces > for rip? > I don't think the RIP policy hooks are capable of doing this at the moment without modifications, see below. > I had a look at "to" inside "term", neighbor is almost, but with rip > multicast is used, so it does not quite fit. > The root of the problem is that policies are global, and do not have per-interface granularity as such. RIP peers can't be statically configured. As such, XORP dynamically creates RIP peers whenever it discovers them on-link; RIPv2 uses link-scope multicasts as you know. Some background: XORP's RIP implementation only implements RIPv2. RIPv1 is a seriously brain-dead protocol, although folk can and do still use it in situations where legacy vendor kit only speaks RIPv1. The hooks in the FEA exist now for doing limited broadcast in a portable way, largely due to the work on OLSR, which also discovers its peers dynamically. Compare RIP policy with OLSR policy, which has a few additional useful tags: http://cvsweb.xorp.org/cgi-bin/cvsweb.cgi/xorp/contrib/olsr/policy_varrw.hh?rev=1.5 ... OLSR's policy support implements the VAR_ORIGINATOR tag, which allows you to implement import policies (only learn routes from OLSR nodes with the given router-ID). That is not quite the same thing, however, OLSR would probably also benefit from policy filters on outgoing interfaces also; think of the case where XORP is configured as an OLSR border node and is redistributing e.g. OSPF routes into OLSR HNA messages, very similar conceptually to RIP redist. Unfortunately, we don't have an output interface match condition implemented. That would cleanly solve your problem in the example you give. I believe the 'neighbor' match condition is only implemented for BGP; it doesn't make sense for OSPF, as LSAs are flooded to all OSPF routers in an area. Each interface in xorp_rip is represented by a instance of the C++ class Port. Whilst xorp_rip does invoke the policy filters on a per-Port (i.e. per-interface) basis, we currently don't have a policy tag in RIP itself for the outgoing interface. The machinery in RIP itself is there for doing this, but the ifname/vifname are not currently set, see here: http://cvsweb.xorp.org/cgi-bin/cvsweb.cgi/xorp/rip/output_table.cc?rev=1.21 thanks, BMS From jhay at meraka.org.za Tue May 12 02:58:55 2009 From: jhay at meraka.org.za (John Hay) Date: Tue, 12 May 2009 11:58:55 +0200 Subject: [Xorp-users] rip on freebsd In-Reply-To: <4A09248E.20004@incunabulum.net> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> <20090507174221.GA60973@zibbi.meraka.csir.co.za> <4A05CEAB.6040605@incunabulum.net> <20090511204249.GA334@zibbi.meraka.csir.co.za> <4A0904C7.6080505@incunabulum.net> <4A09248E.20004@incunabulum.net> Message-ID: <20090512095855.GA44296@zibbi.meraka.csir.co.za> On Tue, May 12, 2009 at 08:26:06AM +0100, Bruce Simpson wrote: > Bruce Simpson wrote: > >I will dig up this script and modify it to use ktrace instead of gdb; it > >is essential that the problem gets pinned down to a particular set of > >socket syscalls. > > > > Here's that script. This is intended for trapping the creation of a XORP > child process and hooking GDB up to it, I think I wrote it to debug a > DoS in the XRL input path, it is probably in Bugzilla somewhere also. As > written here, it is intended for use under Gentoo Linux. ... > > Using 'kdump -E -r -f /tmp/xorp_fea.ktrace' should shed more light on > what syscalls the FEA is invoking when the multicast socket failure is > encountered. kdump knows how to decode some setsockopt() calls in a > human-readable format, but not all. > > Alternatively, you could just ktrace xorp_rtrmgr and all children; this > will not race xorp_fea's creation, but will produce LOTS of output. You > could also hack the FEA's template file to run xorp_fea under ktrace > directly, which is how I usually ran valgrind to find various QA issues > in XORP. I tried an easier (for me) way. I moved xorp_fea to xorp_fea-bin and replaced it with a script that just did: exec ktrace -f /tmp/xorp_fea.ktrace /usr/local/fea/xorp_fea-bin $* I have put xorp.conf, xorp_fea.ktrace.gz, xorp_fea.kdump.gz and xorp_rtrmgr.log on http://people.freebsd.org/~jhay/xorp/ BTW I found that I do not need a default route for rip to work, "route add 224.0.0.0/4 -interface lo0" is enough. John -- John Hay -- John.Hay at meraka.csir.co.za / jhay at FreeBSD.org From bms at incunabulum.net Tue May 12 03:59:55 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 12 May 2009 11:59:55 +0100 Subject: [Xorp-users] rip on freebsd In-Reply-To: <20090512095855.GA44296@zibbi.meraka.csir.co.za> References: <20090507123700.GA43697@zibbi.meraka.csir.co.za> <4A02F63E.9050208@incunabulum.net> <20090507174221.GA60973@zibbi.meraka.csir.co.za> <4A05CEAB.6040605@incunabulum.net> <20090511204249.GA334@zibbi.meraka.csir.co.za> <4A0904C7.6080505@incunabulum.net> <4A09248E.20004@incunabulum.net> <20090512095855.GA44296@zibbi.meraka.csir.co.za> Message-ID: <4A0956AB.2040209@incunabulum.net> John Hay wrote: > I have put xorp.conf, xorp_fea.ktrace.gz, xorp_fea.kdump.gz and > xorp_rtrmgr.log on http://people.freebsd.org/~jhay/xorp/ > Thanks, I'll try to get around to looking at this soon. > BTW I found that I do not need a default route for rip to work, > "route add 224.0.0.0/4 -interface lo0" is enough. > Ah, the old Solaris-ish workaround. Strictly speaking none of these workarounds should be needed, if they are, then there are problems looming deeper which need to be looked at. Thanks, BMS From bms at incunabulum.net Tue May 12 07:59:36 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 12 May 2009 15:59:36 +0100 Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? In-Reply-To: <4003451.889121242135788177.JavaMail.coremail@bj126app91.126.com> References: <4A0902F7.8030707@incunabulum.net> <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> <4003451.889121242135788177.JavaMail.coremail@bj126app91.126.com> Message-ID: <4A098ED8.7050908@incunabulum.net> Hello, Thanks for your feedback. jjw0927 wrote: > Hi, > Sorry, below is something I want to add to the last mail I sent to you: > > Following is the output after I run the command "ldd > /usr/local/xorp/mibs/.libs/bgp4_mib_1657.so" in the terminal: > linux-gate.so.1 => (0x00110000) > libnetsnmpxorp.so => > /usr/local/src/xorp-1.6-RC/mibs/.libs/libnetsnmpxorp.so (0x00169000) ... Most likely the cause of failure is that the libnetsnmpxorp.so component cannot be found at runtime by Net-SNMP, as this appears to be referenced from a source directory. As per my original reply, you will need to use 'gmake install' to install XORP before attempting to use the Net-SNMP MIB support. P.S. Please Cc: all responses regarding support to the xorp-users mailing list as I cannot respond personally. Thank you. From jjw0927 at 126.com Thu May 14 06:22:06 2009 From: jjw0927 at 126.com (jjw0927) Date: Thu, 14 May 2009 21:22:06 +0800 (CST) Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? In-Reply-To: <4A09BAF9.4070107@incunabulum.net> References: <4A09BAF9.4070107@incunabulum.net> <4A0902F7.8030707@incunabulum.net> <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> <1581966.871441242135232109.JavaMail.coremail@bj126app84.126.com> Message-ID: <19660925.854201242307326701.JavaMail.coremail@bj126app87.126.com> ?2009-05-13 02:07:53?"Bruce Simpson" ??? >jjw0927?wrote: >>?Hi, >>?Thanks?for?your?help! >> >>?The?platform?I?am?running?XORP?on?is?Fedora?9. >>?I?run?the?command?"gmake?install"?in?the?terminal,?but?it?also?has? >>?errors?in?the?output.?Below?is?the?output?I?got?when?I?run?the?command? >>?"sudo?./xorp_rtrmgr?-b?config.boot"after?ran?the?command?"gmake?install": >> >>?... >>?dlmod:?dlmod_create_module >>?dlmod:?dlmod_load_module?xorp_if_mib_module:? >>?/usr/local/src/xorp-1.6-RC/mibs/.libs/xorp_if_mib_module.so > > >You?need?to?point?the?XORP?SNMP?config?at?the?location?of?the? >*installed*?MIB?.so?files. > >thanks, >BMS > >P.S.?Please?Cc:?all?correspondence?to?the?mailing?list?in?future?as?I? >*cannot*?respond?personally?to?support?requests. > Sorry, I can not understand what you said. Can you tell me some detailed processes about "how to point?the?XORP?SNMP?config?at?the?location?of?the? *installed*?MIB?.so?files" ? Thank you very much, and I am looking forward for your reply. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090514/a78007b7/attachment.html From bms at incunabulum.net Thu May 14 22:42:29 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 15 May 2009 06:42:29 +0100 Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? In-Reply-To: <19660925.854201242307326701.JavaMail.coremail@bj126app87.126.com> References: <4A09BAF9.4070107@incunabulum.net> <4A0902F7.8030707@incunabulum.net> <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> <1581966.871441242135232109.JavaMail.coremail@bj126app84.126.com> <19660925.854201242307326701.JavaMail.coremail@bj126app87.126.com> Message-ID: <4A0D00C5.70207@incunabulum.net> jjw0927 wrote: > > Sorry, I can not understand what you said. > Can you tell me some detailed processes about "how to point the XORP SNMP config at the location of the > *installed* MIB .so files" ? > You need to change the location of the bgp4_mib_1657.so file in your XORP configuration file to point to where the XORP 'gmake install' installation step has actually installed it. Shared libraries found in .libs directories need to be relinked by libtool before they may be used, using them directly cannot be supported. thanks, BMS From jjw0927 at 126.com Fri May 15 08:02:49 2009 From: jjw0927 at 126.com (jjw0927) Date: Fri, 15 May 2009 23:02:49 +0800 (CST) Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? In-Reply-To: <4A0D00C5.70207@incunabulum.net> References: <4A0D00C5.70207@incunabulum.net> <4A09BAF9.4070107@incunabulum.net> <4A0902F7.8030707@incunabulum.net> <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> <1581966.871441242135232109.JavaMail.coremail@bj126app84.126.com> <19660925.854201242307326701.JavaMail.coremail@bj126app87.126.com> Message-ID: <21942634.397691242399769931.JavaMail.coremail@bj126app78.126.com> ?2009-05-15 13:42:29?"Bruce Simpson" ??? >jjw0927 wrote: >> >> Sorry, I can not understand what you said. >> Can you tell me some detailed processes about "how to point the XORP SNMP config at the location of the >> *installed* MIB .so files" ? >> > >You need to change the location of the bgp4_mib_1657.so file in your >XORP configuration file to point to where the XORP 'gmake install' >installation step has actually installed it. > >Shared libraries found in .libs directories need to be relinked by >libtool before they may be used, using them directly cannot be supported. > >thanks, >BMS Hi, I have changed the location of the bgp4_mib_1657.so file in my XORP configuration file(*.boot file) to point to where the XORP `gmake install` installed it(in /usr/local/xorp/mibs by default). But it can not take effective. And I also can not launch Net-SNMP via XORP. However, if I use the ospf_mib_1850.so file, it can work and Net-SNMP has been launched by XORP. I do not know how to use the libtool file which is found in the $XORP directory(the same as Makefile and .configure) to relink the shared libraries in .libs directory. Can you tell me the command and the parameter? Thank you very much! And I am looking forward to your reply! Best wishes Jin JianWu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090515/122dc901/attachment.html From m.j.goldstone at isc.keele.ac.uk Fri May 15 10:16:34 2009 From: m.j.goldstone at isc.keele.ac.uk (Martin Goldstone) Date: Fri, 15 May 2009 18:16:34 +0100 Subject: [Xorp-users] Multicast Routing on Windows Message-ID: <4A0DA372.6000308@isc.keele.ac.uk> Hi all, I'm currently having an issue involving multicast, OSPF and Windows Server 2003. We have a software deployment system (Windows only, I'm afraid) that uses multicast, and I need to use OSPF on the servers in order to provide some form of fail-over (in the form of a service address, either added as an additional address to the main network connection, or a vmware network card that goes nowhere). I've started out by looking at Routing and Remote Access, which does the OSPF part of things fine (after a little tweak to our Cisco routers), however as the box now has 2 interfaces, and the software deployment system is bound to the service address, multicast fails as the traffic is not routed between the 2 interfaces. Sadly, Windows lacks any multicast routing protocols (it has IGMP, but no support for PIM, which is what all the Cisco routers speak here). As I understand it, that would be fine if we just had a receiver on the Windows end, but it doesn't seem to work with the sender being there. So, after getting rather frustrated with Windows, I decided to take a look at Xorp. Originally, I wanted to just rip Rounting and Remote Access out, and get Xorp to do the OSPF stuff in the first instance. This proved to be difficult, as I was unable to get 1.6 or 1.5 to start at all (first warning about a route via unknown interface, then they complain about an assertion (vifp != null, I believe) failing, then they quit), but I managed to get 1.4 started, but failed to get OSPF working in the way I needed it to (it seems it was working to a degree, as I could see it as a neighbour on the Cisco kit, and I could see the routing table from OSPF in Xorp, but for some reason, the route to the service address wasn't being exposed, thus leaving it inaccessible). From this point I decided to go back to Routing and Remote Access, and had it running doing OSPF while I tried to convince Xorp to just do the multicast stuff. Sadly all 3 versions complained about not having an RTMv2 pipe available. I found the source code for the relevant dll in contrib/win32/xorprtm, convinced it to compile (using the suggested Visual C++ Toolkit 2003, and the Windows Server 2003 R2 Platform SDK, after having to add some #include lines to bsdroute.h), and put that in, at which point I was able to get Xorp running again. It's at this point I hit what seems to be my final roadblock, and sadly it wasn't in the release notes (though I've since found it in the build notes): Multicast routing is not supported on Windows. So at the moment, I'm stuck. I'm wondering if it's possible to just get Xorp to handle the PIM stuff on it's own, though to be honest, I don't know enough about multicast to know if that would be sufficient. Does anyone have any ideas about a solution to this mess? Aside from using a different operating system, of course, as I'm stuck with Windows on this one. Thanks in advance, Martin -- Martin Goldstone Keele University, Keele, IT Systems Administrator Staffordshire, United Kingdom, ST5 5BG Finance & IT Telephone: +44 1782 734457 "I feel a great disturbance in the network, as if millions of processes cried out in terror and were suddenly silenced" Boyd Duffee -------------- next part -------------- A non-text attachment was scrubbed... Name: m_j_goldstone.vcf Type: text/x-vcard Size: 261 bytes Desc: not available Url : http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090515/d420d346/attachment.vcf From bms at incunabulum.net Fri May 15 11:22:40 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 15 May 2009 19:22:40 +0100 Subject: [Xorp-users] Multicast Routing on Windows In-Reply-To: <4A0DA372.6000308@isc.keele.ac.uk> References: <4A0DA372.6000308@isc.keele.ac.uk> Message-ID: <4A0DB2F0.1020905@incunabulum.net> Martin Goldstone wrote: > ... > From this point I decided to go back to Routing and Remote Access, and > had it running doing OSPF while I tried to convince Xorp to just do > the multicast stuff. Sadly all 3 versions complained about not having > an RTMv2 pipe available. ... > It's at this point I hit what seems to be my final roadblock, and > sadly it wasn't in the release notes (though I've since found it in > the build notes): Multicast routing is not supported on Windows. Correct... and that should be in the release notes, given that a binary build was distributed. But it is largely academic given the lack of ongoing Windows support. No organisation has had sufficient interest to make use of it, as far as I know, since Microsoft's original interest. > > So at the moment, I'm stuck. I'm wondering if it's possible to just > get Xorp to handle the PIM stuff on it's own, though to be honest, I > don't know enough about multicast to know if that would be sufficient. > > Does anyone have any ideas about a solution to this mess? Aside from > using a different operating system, of course, as I'm stuck with > Windows on this one. I'm afraid you're out of luck. There are a number of things in the Windows port which couldn't be finished, either due to lack of time or lack of information. Multicast forwarding is just very different in the Windows environment. If a business case could be made for reopening Windows support on an ongoing basis, then I daresay something could be done about the lack of multicast support. However, this seems unlikely, given that Windows is not part of XORP, Inc.'s development focus at all. thanks, BMS From bms at incunabulum.net Fri May 15 11:26:15 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 15 May 2009 19:26:15 +0100 Subject: [Xorp-users] How to Launching Net-SNMP via the router manager? In-Reply-To: <21942634.397691242399769931.JavaMail.coremail@bj126app78.126.com> References: <4A0D00C5.70207@incunabulum.net> <4A09BAF9.4070107@incunabulum.net> <4A0902F7.8030707@incunabulum.net> <2062880.75411242007116773.JavaMail.coremail@bj126app87.126.com> <1581966.871441242135232109.JavaMail.coremail@bj126app84.126.com> <19660925.854201242307326701.JavaMail.coremail@bj126app87.126.com> <21942634.397691242399769931.JavaMail.coremail@bj126app78.126.com> Message-ID: <4A0DB3C7.9030501@incunabulum.net> jjw0927 wrote: > > I have changed the location of the bgp4_mib_1657.so file in my XORP > configuration file(**.boot* file) to point to where the XORP `gmake > install` installed it(in */usr/local/xorp/mibs* by default). But it > can not take effective. And I also can not launch Net-SNMP via XORP. You will need to determine why the shared library is not loaded. You may need to resort to using other techniques such as system call tracing (e.g. strace). > > I do not know how to use the *libtool* file which is found in the > $XORP directory(the same as Makefile and .configure) to relink the > shared libraries in .libs directory. Can you tell me the command and > the parameter? The point is that this is done for you by the install target in the GNU makefiles already, and you should not need to do it yourself. From bhavin81 at iitb.ac.in Tue May 19 05:10:59 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Tue, 19 May 2009 17:40:59 +0530 Subject: [Xorp-users] Need Help for pimsm multicasting join or any message modification Message-ID: <6eeb262e51a42742e7c508f0bc5254ac.squirrel@gpo.iitb.ac.in> Hello friends, I want to do some modification in standard pim-sm multicasting. I want to modify either join or some other message such that whenever any user wants to join any group it(user) specifies some sort of choice by some flag which is part of pim-sm message field. Is there any field in the message which i can modify means any optional field? If yes then which code out of pim-sm in xorp i need to modify? Regards, Bhavin, Comm. student. From bms at incunabulum.net Tue May 19 11:51:40 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 19 May 2009 19:51:40 +0100 Subject: [Xorp-users] Need Help for pimsm multicasting join or any message modification In-Reply-To: <6eeb262e51a42742e7c508f0bc5254ac.squirrel@gpo.iitb.ac.in> References: <6eeb262e51a42742e7c508f0bc5254ac.squirrel@gpo.iitb.ac.in> Message-ID: <4A12FFBC.30703@incunabulum.net> bhavin81 at iitb.ac.in wrote: > Hello friends, > > I want to do some modification in standard pim-sm multicasting. > I want to modify either join or some other message such that whenever > any user wants to join any group it(user) specifies some sort of choice by > some flag which is part of pim-sm message field. OK... > Is there any field in > the message which i can modify means any optional field? If yes then which > code out of pim-sm in xorp i need to modify? Not really, unless you modify the reserved bits in the PIM-SM fixed header (which are supposed to be ignored on receive). A better bet would be the source address tuple, where a number of bits are already allocated in the first 18 bits of the tuple. Pick up a copy of RFC 2362 (in PostScript format preferably, I cannot seem to find the original) or refer to Interdomain Multicast Routing (Edwards/Giuliano/Wright, Addison-Wesley). See PimJpHeader::network_send() here, where encoding actually happens, work your way back up the call graph: http://cvsweb.xorp.org/cgi-bin/cvsweb.cgi/xorp/pim/pim_proto_join_prune_message.cc?rev=1.33 P.S. I would advise testing very carefully if you plan to run this modification in a production network with standards-compliant equipment. thanks, BMS From bms at incunabulum.net Tue May 19 11:54:02 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 19 May 2009 19:54:02 +0100 Subject: [Xorp-users] Need Help for pimsm multicasting join or any message modification In-Reply-To: <4A12FFBC.30703@incunabulum.net> References: <6eeb262e51a42742e7c508f0bc5254ac.squirrel@gpo.iitb.ac.in> <4A12FFBC.30703@incunabulum.net> Message-ID: <4A13004A.7010507@incunabulum.net> Bruce Simpson wrote: >> Is there any field in >> the message which i can modify means any optional field? If yes then >> which >> code out of pim-sm in xorp i need to modify? > > Not really, unless you modify the reserved bits in the PIM-SM fixed > header (which are supposed to be ignored on receive). Actually you probably want to look at this RFC, which XORP PIM-SM does not yet implement as there has been no demand for it yet: http://www.rfc-editor.org/rfc/rfc5384.txt patches welcome! thanks, BMS From william at losrios.edu Tue May 19 14:46:00 2009 From: william at losrios.edu (Williams, Mark) Date: Tue, 19 May 2009 14:46:00 -0700 Subject: [Xorp-users] Xorp behavior when underlying physical interfaces are administratively down Message-ID: <1B62E6082D9F034394DB2FBAC03D1A3860E40C90B7@lrccd-exch05.LRCCD.ad.losrios.edu> I am planning on deploying Xorp on a standby router, which normally has all of its interfaces cabled into a switch but administratively down on Linux system. What will Xorp do in such a situation? I plan on using OSPF, and I want to make sure that I don't get any IP address conflicts or problems with OSPF routes. -Mark Williams -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090519/8a5fcbfd/attachment.html From bms at incunabulum.net Thu May 21 06:46:40 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 21 May 2009 14:46:40 +0100 Subject: [Xorp-users] Xorp behavior when underlying physical interfaces are administratively down In-Reply-To: <1B62E6082D9F034394DB2FBAC03D1A3860E40C90B7@lrccd-exch05.LRCCD.ad.losrios.edu> References: <1B62E6082D9F034394DB2FBAC03D1A3860E40C90B7@lrccd-exch05.LRCCD.ad.losrios.edu> Message-ID: <4A155B40.8020803@incunabulum.net> Williams, Mark wrote: > > I am planning on deploying Xorp on a standby router, which normally > has all of its interfaces cabled into a switch but administratively > down on Linux system. > > What will Xorp do in such a situation? I plan on using OSPF, and I > want to make sure that I don?t get any IP address conflicts or > problems with OSPF routes. > So this is not a 'hot standby' in the VRRP / CARP / HSRP sense of the word, but a 'warm standby' ? I believe this should work, however, the OSPF process will remain idle most of the time until you bring the interfaces up. I assume you are using some form of out-of-band access (e.g. serial, or a management LAN) to administer your standby router. It would probably be best to do this using the Router Manager CLI, i.e. configure the interfaces as 'disabled: true' in the interfaces {} block, and use configuration mode to bring them up manually when failing over to your 'warm standby' router. Of course your router will not have any OSPF state as all your interfaces will be down -- it will have to come up and learn that from other OSPF routers when you manually fail over to it. It might be better to leave it running with an artificially high interface cost, or consider using the new VRRP feature in XORP 1.6. thanks, BMS From bhavin81 at iitb.ac.in Fri May 22 21:26:25 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Sat, 23 May 2009 09:56:25 +0530 Subject: [Xorp-users] basic querry: if xorp makes a pc router then what about node mentioned in pimsm test suite? Message-ID: <9f51a1c4c6fa5adf1f29a3ef96a67967.squirrel@gpo.iitb.ac.in> Hello friends, I have got one basic querry. By using xorp we make a pc act like a router. Then in the case of multicasting how are we going to involve nodes as mentioned in test suite of pim-sm? Tell me whether following statement is correct or not?If we involve nodes then and only then igmp join or prune will be send? If yes then how can we send igmp join from any node(pc) to xorp router? Regards, Bhavin, Comm. Student. From luckystud55 at gmail.com Fri May 22 22:14:17 2009 From: luckystud55 at gmail.com (Lucky Stud) Date: Sat, 23 May 2009 01:14:17 -0400 Subject: [Xorp-users] XORP - Linux - Multicast setup help Message-ID: <957b38290905222214k1aa0ecd3k6d13392db72fe9b8@mail.gmail.com> Hi, I am trying to setup the xorp on a linux (RHEL4) machine with 3 interfaces which routes the traffic for two subnets (setup below). eth0 is directly connected to the a cisco router. I need to enable multicast routing in order to be able to use norton ghost and multicast to image the machines in the 192.0/26 and 192.64/26 subnets and also to receive a multicast video feed transmission from internet. cisco router is setup properly as I can use multicast on all other subnets that are directly connected to the cisco router. The two subnets behind the linux router are live ip and not natted. I am not sure how to setup the static-rps and/or the bootstrap for the pimsm4 directive to get it to work properly. which interface needs to be the BSR or RP or just use the static rps. Any advise is greatly appreciated. I downloaded the latest version of xorp from official site, compiled and installed it. Thanks. [muulticast receivers] 192.0/26 ---------- [192.1 (eth1) \ | XORP-linux router- 67.141 (eth0) ]-----------------------[ 67.129 Cisco Router ]-----------> out / multicast sender [muulticast receivers] 192.64/26---------- [192.65 (eth2) / route -n 192.0 0.0.0.0 255.255.255.192 U 0 0 0 eth1 192.64 0.0.0.0 255.255.255.192 U 0 0 0 eth2 67.128 0.0.0.0 255.255.255.192 U 0 0 0 eth0 0.0.0.0 142.204.67.129 0.0.0.0 UG 0 0 0 eth0 config.boot interfaces { restore-original-config-on-shutdown: true; interface eth0 { default-system-config } interface eth1 { default-system-config } interface eth2 { default-system-config } } fea { unicast-forwarding4 { disable: false } unicast-forwarding6 { disable: true } } plumbing { mfea4 { disable: false interface eth0 { vif eth0 { } } interface eth1{ vif eth1 { } } interface eth2{ vif eth2 { } } interface register_vif { vif register_vif { } } traceoptions { flag all { disable: false } } } } protocols { igmp { interface eth0 { vif eth0 { } } interface eth1 { vif eth1 { } } interface eth2 { vif eth2 { } } traceoptions { flag all { disable: false } } } } protocols { pimsm4 { disable: false interface eth0 { vif eth0 { } } interface eth1 { vif eth1 { } } interface eth2 { vif eth2 { } } interface register_vif { vif register_vif { } } /* static-rps { rp 142.204.67.129 { group-prefix 224.0.0.0/4 { } } } */ bootstrap { disable: false cand-bsr { scope-zone 224.0.0.0/4 { /* is-scope-zone: false */ cand-bsr-by-vif-name: "eth0" /* cand-bsr-by-vif-addr: 10.10.10.10 */ /* bsr-priority: 1 */ /* hash-mask-len: 30 */ } } cand-rp { group-prefix 224.0.0.0/4 { /* is-scope-zone: false */ cand-rp-by-vif-name: "eth0" /* cand-rp-by-vif-addr: 10.10.10.10 */ /* rp-priority: 192 */ /* rp-holdtime: 150 */ } } } traceoptions { flag all { disable: false } } } } protocols { fib2mrib { } } -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090523/586b70d5/attachment.html From bms at incunabulum.net Sat May 23 06:51:18 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sat, 23 May 2009 14:51:18 +0100 Subject: [Xorp-users] Need Help for pimsm multicasting join or any message modification In-Reply-To: References: <6eeb262e51a42742e7c508f0bc5254ac.squirrel@gpo.iitb.ac.in> <4A12FFBC.30703@incunabulum.net> <4A13004A.7010507@incunabulum.net> Message-ID: <4A17FF56.5030604@incunabulum.net> bhavin81 at iitb.ac.in wrote: > ... > I have read the rfc mentioned in your message but i think i do not > need that much big modification of different type encoding as such. > Your view please? > In any event, you'll need to modify PIM itself to propagate your information up the shared tree. Given that there's an RFC published for extending PIM with attributes now, it's slightly more likely that vendors will adopt the RFC just because it exists -- although that obviously doesn't guarantee that anyone will implement the RFC at all. There is a learning curve for the XORP PIM-SM implementation, and the time involved in that learning curve is probably high enough that it is worth implementing PIM attributes anyway. You certainly don't have to, that is just my opinion -- but be aware that PIM is a difficult protocol to extend, for the reasons I gave in my original reply, which is why the RFC authors spent time drafting an RFC and putting it through the rfc-editor meat grinder. P.S. Please Cc: questions/messages about XORP to the xorp-users list as I cannot respond personally. Thank you. From bms at incunabulum.net Sat May 23 06:52:34 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sat, 23 May 2009 14:52:34 +0100 Subject: [Xorp-users] basic querry: if xorp makes a pc router then what about node mentioned in pimsm test suite? In-Reply-To: <9f51a1c4c6fa5adf1f29a3ef96a67967.squirrel@gpo.iitb.ac.in> References: <9f51a1c4c6fa5adf1f29a3ef96a67967.squirrel@gpo.iitb.ac.in> Message-ID: <4A17FFA2.9080903@incunabulum.net> bhavin81 at iitb.ac.in wrote: > Hello friends, > > I have got one basic querry. By using xorp we make a pc > act like a router. Then in the case of multicasting how are we > going to involve nodes as mentioned in test suite of pim-sm? > Tell me whether following statement is correct or not?If we involve nodes > then and only then igmp join or prune will be send? If yes then how > can we send igmp join from any node(pc) to xorp router? > I believe the use of the term Node in the PIM-SM code is not directly related to the notion of a XORP router as a 'node'. thanks, BMS From bhavin81 at iitb.ac.in Sat May 23 18:31:06 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Sun, 24 May 2009 07:01:06 +0530 Subject: [Xorp-users] Basic querry: PIM TEST SUITE Message-ID: <08c3afc427832cd3e7df8e1593dc49e6.squirrel@gpo.iitb.ac.in> Hello friends, I have got one basic querry. By using xorp we make a pc act like a router. Then in the case of multicasting how are we going to involve nodes as mentioned in test suite of pim-sm? Tell me whether following statement is correct or not?If we involve nodes then and only then igmp join or prune will be send? If yes then how can we send igmp join from any node(pc) to xorp router? What i mean to say is: PC1 ----> XORPROUTER1(XORP RUNNING ON PC2)-----> XORPROUTER2(XORP RUNNING ON PC3)----->XORPROUTER3(XORP RUNNING ON PC4)WHICH may be RP. The querry is how pc1 is going to send igmp join message to XORPROUTER1. Are we going to run Xorp on it and run only up to IGMP i.e. PIM-SM is not enabled on that interface or Do we need separte mechanism for sending igmp join messages to XORPROUTER1? The other querry is : pim-test suite: Diagram 3.3 Rx1 ---LAN1--------> TR1 ------LAN2----->TR2 ---LAN3---------> RUT-----LAN4------> S1 In above diagram Rx1 is a receiver and S1 is a sender but how are we going to configure them to send any data? whether Rx1 is a separte machine attached to TR1(test router1) via some lan interface? Regards, Bhavin, Comm. Student. From eqmaker at gmail.com Tue May 26 22:57:31 2009 From: eqmaker at gmail.com (Andrei Kim) Date: Wed, 27 May 2009 14:57:31 +0900 Subject: [Xorp-users] I can not start xorp.. Message-ID: Hello all.. I'm new~ :) I installd debian 5.0.1 and xorp daemon yesterday. after all, I typed 'xorpsh' on shell (as root)... *tom/etc# xoropsh Waiting for xorp_rtrmgr.. [2009/05/27 14:40:17 ERROR xorpsh:3228 RTRMGR +91 xorpsh_main.cc wait_for_xrl_router_ready ] XrlRouter failed. No Finder? [2009/05/27 14:40:17 ERROR xorpsh:3228 RTRMGR +908 xorpsh_main.cc main] xorpsh exiting due to an init error : Failed to connect to the router manager* and.. exit to propmpt... and I tyied another... *tom\etc# xorp_rtrmgr [2009/05/27 14:43:22 ERROR xorp_rtrmgr:3229 RTRMGR +349 main_rtrmgr.cc run] rtrmgr shutting down due to an init error : Failed to open config file : /usr/lib/xorp/config.boot]* there are no 'config.boot'file at /usr/lib/xorp... How can I make this working? Thanks. Andrei. eqmaker at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090527/0d8a2a4e/attachment.html From bhavin81 at iitb.ac.in Wed May 27 04:46:38 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Wed, 27 May 2009 17:16:38 +0530 Subject: [Xorp-users] querry in pim-sm module file:pim_proto_join_prune_message.cc Message-ID: <472233dd2c59d75bbf5767c2c9a0ea6a.squirrel@gpo.iitb.ac.in> Hello friends, I have got one querry in pim-smmodulefile :pim_proto_join_prune_message.cc of xorp.The querry is located on line no.s: 246-250. The lines are: // Remove redundant entries: all (S,G,rpt)J while (! jp_group->sg_rpt()->j_list().empty()) { const IPvX& addr = *jp_group->sg_rpt()->j_list().begin(); jp_group->sg_rpt()->j_list_remove(addr); What does above lines do? Why do not we do it in (*,*,RP) table? Thanks in advance for help. Regards, Bhavin, Networking Student. From bhavin81 at iitb.ac.in Wed May 27 04:56:23 2009 From: bhavin81 at iitb.ac.in (bhavin81 at iitb.ac.in) Date: Wed, 27 May 2009 17:26:23 +0530 Subject: [Xorp-users] Reply: querry in pim-sm module file:pim_proto_join_prune_message.cc] Message-ID: <32cbe9a060c5da6c33817dc9074202f6.squirrel@gpo.iitb.ac.in> Hello friends, As per my best knowledge it tries to delete all (S,G,RPT) entries from (*,G) table but why is not known to me? Regards, Bhavin. Hello friends, I have got one querry in pim-smmodulefile :pim_proto_join_prune_message.cc of xorp.The querry is located on line no.s: 246-250. The lines are: // Remove redundant entries: all (S,G,rpt)J while (! jp_group->sg_rpt()->j_list().empty()) { const IPvX& addr = *jp_group->sg_rpt()->j_list().begin(); jp_group->sg_rpt()->j_list_remove(addr); What does above lines do? Why do not we do it in (*,*,RP) table? Thanks in advance for help. Regards, Bhavin, Networking Student. From bms at incunabulum.net Wed May 27 06:53:12 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Wed, 27 May 2009 14:53:12 +0100 Subject: [Xorp-users] I can not start xorp.. In-Reply-To: References: Message-ID: <4A1D45C8.2030404@incunabulum.net> Andrei Kim wrote: > /.../ > /m\etc# xorp_rtrmgr > [2009/05/27 14:43:22 ERROR xorp_rtrmgr:3229 RTRMGR +349 main_rtrmgr.cc > run] rtrmgr shutting down due to an init error : Failed to open config > file : /usr/lib/xorp/config.boot]/ > > there are no 'config.boot'file at /usr/lib/xorp... > > How can I make this working? You need to write a configuration file for XORP, it cannot run without a configuration. For examples see the rtrmgr/config directory in the source tree. thanks, BMS From petar.bojovic at paxy.in.rs Wed May 27 12:11:38 2009 From: petar.bojovic at paxy.in.rs (Petar Bojovic) Date: Wed, 27 May 2009 21:11:38 +0200 Subject: [Xorp-users] ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! Message-ID: <2945798e0905271211p30ff567clf7182ec2bc756f3e@mail.gmail.com> Hi, I am trying to simulate similar situation as in post: http://mailman.icsi.berkeley.edu/pipermail/xorp-users/2007-February/001680.html I am using Voyage Linux platform based on Ubuntu. Have compiled and installed XORP 1.6 from source, and set same config file as in mentioned post. config.boot interfaces { interface eth0 { description: "upstream interface" disable: false default-system-config } interface eth1{ description: "downstream interface" disable: false default-system-config } } fea { unicast-forwarding4 { disable: false } } plumbing { mfea4 { disable: false interface eth0 { vif eth0 { disable: false } } interface eth1 { vif eth1 { disable: false } } interface register_vif { vif register_vif { disable: false } } traceoptions { flag all { disable: false } } } } protocols { igmp { disable: false interface eth0 { vif eth0 { version: 3 disable: false } } interface eth1 { vif eth1 { version: 3 disable: false } } traceoptions { flag all { disable: false } } } } protocols { pimsm4 { disable: false interface eth0 { vif eth0 { disable: false } } interface eth1 { vif eth1 { disable: false } } interface register_vif { vif register_vif { disable: false } } traceoptions { flag all { disable: false } } } } protocols { fib2mrib { disable: false } ifconfig -> eth0 Link encap:Ethernet HWaddr 00:0D:B9:14:D8:00 inet addr:172.16.1.245 Bcast:172.16.1.255 Mask:255.255.255.0 inet6 addr: fe80::20d:b9ff:fe14:d800/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:20843 errors:0 dropped:0 overruns:0 frame:0 TX packets:9830 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1453533 (1.3 MiB) TX bytes:2335612 (2.2 MiB) Interrupt:10 eth1 Link encap:Ethernet HWaddr 00:0D:B9:14:D8:01 inet addr:172.16.2.245 Bcast:172.16.2.255 Mask:255.255.255.0 inet6 addr: fe80::20d:b9ff:fe14:d801/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1485 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:106289 (103.7 KiB) TX bytes:468 (468.0 b) Interrupt:12 Base address:0x4000 After running xorp_rtrmgr, I get next output: [ 2009/05/27 19:06:42 INFO xorp_rtrmgr:10011 RTRMGR +239 master_conf_tree.cc execute ] Changed modules: interfaces, firewall, fea, mfea4, rib, fib2mrib, igmp, pimsm4 [ 2009/05/27 19:06:42 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: interfaces (fea/xorp_fea) [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] MFEA enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI started [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] MFEA enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI started [ 2009/05/27 19:06:44 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: firewall (fea/xorp_fea) [ 2009/05/27 19:06:48 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: fea (fea/xorp_fea) [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: mfea4 (fea/xorp_fea) [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface added: Vif[eth0] pif_index: 2 vif_index: 0 addr: 172.16.1.245 subnet: 172.16.1.0/24broadcast: 172.16.1.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface added: Vif[eth1] pif_index: 3 vif_index: 1 addr: 172.16.2.245 subnet: 172.16.2.0/24broadcast: 172.16.2.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] MFEA started [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface enabled Vif[eth0] pif_index: 2 vif_index: 0 addr: 172.16.1.245 subnet: 172.16.1.0/24broadcast: 172.16.1.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 DOWN IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface started: Vif[eth0] pif_index: 2 vif_index: 0 addr: 172.16.1.245 subnet: 172.16.1.0/24broadcast: 172.16.1.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 UP IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface added: Vif[register_vif] pif_index: 2 vif_index: 2 addr: 172.16.1.245 subnet: 172.16.1.245/32 broadcast: 172.16.1.245 peer: 0.0.0.0 Flags: PIM_REGISTER UNDERLYING_VIF_UP MTU: 1500 [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface enabled Vif[eth1] pif_index: 3 vif_index: 1 addr: 172.16.2.245 subnet: 172.16.2.0/24broadcast: 172.16.2.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 DOWN IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface started: Vif[eth1] pif_index: 3 vif_index: 1 addr: 172.16.2.245 subnet: 172.16.2.0/24broadcast: 172.16.2.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 UP IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface enabled Vif[register_vif] pif_index: 2 vif_index: 2 addr: 172.16.1.245 subnet: 172.16.1.245/32 broadcast: 172.16.1.245 peer: 0.0.0.0 Flags: PIM_REGISTER UNDERLYING_VIF_UP MTU: 1500 DOWN IPv4 ENABLED [ 2009/05/27 19:06:54 ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument [ 2009/05/27 19:06:54 ERROR xorp_fea:10013 MFEA +1186 mfea_node.cc start_vif ] Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 WARNING xorp_fea XrlMfeaTarget ] Handling method for mfea/0.1/start_vif failed: XrlCmdError 102 Command failed Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 ERROR xorp_rtrmgr:10011 RTRMGR +681 master_conf_tree.cc commit_pass2_done ] Commit failed: 102 Command failed Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 ERROR xorp_rtrmgr:10011 RTRMGR +251 master_conf_tree.cc config_done ] Configuration failed: 102 Command failed Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +2228 task.cc run_task ] No more tasks to run [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: fea [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: firewall [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: interfaces [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: mfea4 [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +194 module_manager.cc terminate ] Killing module: mfea4 [ 2009/05/27 19:06:54 ERROR xorp_rtrmgr:10011 RTRMGR +747 module_manager.cc done_cb ] Command "/usr/lib/xorp/fea/xorp_fea": terminated with signal 15. [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +282 module_manager.cc module_exited ] Module killed during shutdown: mfea4 I guest that main problem is: ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! Have tried some older version, same as binary versions, but always same problem. Same problem came up when I test this configuration on FC9. Does anybody have a clue what could make this problem ? Thank you folks ! Petar Bojovic -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090527/c3fc52a8/attachment.html From petar.bojovic at paxy.in.rs Wed May 27 13:14:24 2009 From: petar.bojovic at paxy.in.rs (Petar Bojovic) Date: Wed, 27 May 2009 22:14:24 +0200 Subject: [Xorp-users] ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! In-Reply-To: <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DAD@lrccd-exch05.LRCCD.ad.losrios.edu> References: <2945798e0905271211p30ff567clf7182ec2bc756f3e@mail.gmail.com> <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DAD@lrccd-exch05.LRCCD.ad.losrios.edu> Message-ID: <2945798e0905271314j476f875bx9aefadb9fd013d36@mail.gmail.com> Thank you for your answer. There are outputs: voyage:/usr/local/xorp/bin# cat /proc/net/ip_mr_cache Group Origin Iif Pkts Bytes Wrong Oifs voyage:/usr/local/xorp/bin# cat /proc/net/ip_mr_vif Interface BytesIn PktsIn BytesOut PktsOut Flags Local Remote What could be mine next steps to solve this issue ? Thank you Mark ! -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090527/970b765f/attachment.html From petar.bojovic at paxy.in.rs Thu May 28 03:53:18 2009 From: petar.bojovic at paxy.in.rs (Petar Bojovic) Date: Thu, 28 May 2009 12:53:18 +0200 Subject: [Xorp-users] ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! In-Reply-To: <2945798e0905271314j476f875bx9aefadb9fd013d36@mail.gmail.com> References: <2945798e0905271211p30ff567clf7182ec2bc756f3e@mail.gmail.com> <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DAD@lrccd-exch05.LRCCD.ad.losrios.edu> <2945798e0905271314j476f875bx9aefadb9fd013d36@mail.gmail.com> Message-ID: <2945798e0905280353v6c1dfa81h1b06afeeb810dce8@mail.gmail.com> Thank you Mark, After recompiling kernel with next options, there is no more this error. Have set next config in kernel configuration file: CONFIG_IP_MULTICAST=y CONFIG_IP_MROUTE=y CONFIG_IP_PIMSM_V1=y CONFIG_IP_PIMSM_V2=y CONFIG_NET_IPIP=y On Wed, May 27, 2009 at 10:14 PM, Petar Bojovic wrote: > Thank you for your answer. > There are outputs: > > voyage:/usr/local/xorp/bin# cat /proc/net/ip_mr_cache > Group Origin Iif Pkts Bytes Wrong Oifs > > voyage:/usr/local/xorp/bin# cat /proc/net/ip_mr_vif > Interface BytesIn PktsIn BytesOut PktsOut Flags Local Remote > > What could be mine next steps to solve this issue ? > > Thank you Mark ! > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090528/3f24322d/attachment-0001.html From bms at incunabulum.net Thu May 28 07:52:23 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 28 May 2009 15:52:23 +0100 Subject: [Xorp-users] Basic querry: PIM TEST SUITE In-Reply-To: <08c3afc427832cd3e7df8e1593dc49e6.squirrel@gpo.iitb.ac.in> References: <08c3afc427832cd3e7df8e1593dc49e6.squirrel@gpo.iitb.ac.in> Message-ID: <4A1EA527.2000201@incunabulum.net> bhavin81 at iitb.ac.in wrote: > Hello friends, > > I have got one basic querry. By using xorp we make a pc > act like a router. Then in the case of multicasting how are we > going to involve nodes as mentioned in test suite of pim-sm? > OK, there is no existing test suite in the source tree that I can find for XORP's PIM-SM implementation, so I'm assuming from your subject that you intend to create one. Excellent, thank you. I had some trouble understanding this from the context of your message. > Tell me whether following statement is correct or not?If we involve nodes > then and only then igmp join or prune will be send? If yes then how > can we send igmp join from any node(pc) to xorp router? > > What i mean to say is: > PC1 ----> XORPROUTER1(XORP RUNNING ON PC2)-----> XORPROUTER2(XORP > RUNNING ON PC3)----->XORPROUTER3(XORP RUNNING ON PC4)WHICH may be RP. > > The querry is how pc1 is going to send igmp join message to XORPROUTER1. > Are we going to run Xorp on it and run only up to IGMP i.e. PIM-SM is not > enabled on that interface or Do we need separte mechanism for sending igmp > join messages to XORPROUTER1? > You need IGMP in any IPv4 PIM deployment where hosts (not routers) are involved, as it is the means by which PIM routers learn about the hosts. XORP itself only implements the router portion of the IGMP protocol. The host portion of the IGMP protocol is usually part of the operating system on an end-station. It is possible to simulate it fairly easily, e.g. using a toolkit such as Packet Construction Set (pcs.sourceforge.net) which is a Python-based, object-oriented library for working with network protocols. For IGMP to work, you need to simulate a multicast-capable link. It doesn't matter if this link is Ethernet, Point-to-Point, or NBMA; what matters is that you can send IGMP traffic over it to the next-hop multicast router. Whilst multicast forwarding can happen on a completely different set of routers from normal IPv4/IPv6 unicast forwarding, multicast routers *do* need to know the unicast routing topology for PIM's internal uRPF checks to work. These are a necessary part of the PIM protocol to prevent multicast forwarding loops. You don't specify how you intend to implement the simulation of each network node in the topology, so I assume you intend run a virtual instance of an operating system. In this case, the host part of the IGMP protocol will be implemented by whichever OS you use (e.g. FreeBSD, Linux etc). I highly recommend moving to IGMPv3 for this. IGMPv2 is a very different protocol, and IGMPv1 may be regarded as historical. IGMPv3 has a number of features in it which potentially reduce interrupt load on multicast routers, as well as having a degree of robustness over lossy links through retransmissions. It is also necessary if you want to implement the Source-Specific Multicast (SSM) model, rather than the old Any-Source Multicast (ASM) model. I also recommend you find a copy of the book "Interdomain Multicast Routing", as pointed out in previous replies, as it explains all of the above (i.e. how IPv4 multicast infrastructure works) in great technical detail. thanks, BMS From bms at incunabulum.net Thu May 28 08:07:51 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 28 May 2009 16:07:51 +0100 Subject: [Xorp-users] Reply: querry in pim-sm module file:pim_proto_join_prune_message.cc] In-Reply-To: <32cbe9a060c5da6c33817dc9074202f6.squirrel@gpo.iitb.ac.in> References: <32cbe9a060c5da6c33817dc9074202f6.squirrel@gpo.iitb.ac.in> Message-ID: <4A1EA8C7.6050805@incunabulum.net> bhavin81 at iitb.ac.in wrote: > Hello friends, > > As per my best knowledge it tries to delete all (S,G,RPT) entries > from (*,G) table but why is not known to me? > The best person to direct questions to about the PIM implementation is probably the original author, Pavlin Radoslavov. However, he is quite a busy guy. Reading around that code (which I'm not directly familiar with and took a little while) it seems that the code you're discussing is correct. If the WC bit is set on a pending PIM join-prune, it means the group is in (*,G) (any-source) mode, therefore it makes no sense for the sources to be sent for that group to a PIM-SM peer in the join-prune message. thanks, BMS From bms at incunabulum.net Thu May 28 08:28:34 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 28 May 2009 16:28:34 +0100 Subject: [Xorp-users] XORP - Linux - Multicast setup help In-Reply-To: <957b38290905222214k1aa0ecd3k6d13392db72fe9b8@mail.gmail.com> References: <957b38290905222214k1aa0ecd3k6d13392db72fe9b8@mail.gmail.com> Message-ID: <4A1EADA2.6070309@incunabulum.net> Hello, Sorry for the delay in replying to your message. Lucky Stud wrote: > Hi, > > I am trying to setup the xorp on a linux (RHEL4) machine with 3 > interfaces which routes the traffic for two subnets (setup below). > eth0 is directly connected to the a cisco router. I need to enable > multicast routing in order to be able to use norton ghost and > multicast to image the machines in the 192.0/26 and 192.64/26 subnets > and also to receive a multicast video feed transmission from internet. > cisco router is setup properly as I can use multicast on all other > subnets that are directly connected to the cisco router. The two > subnets behind the linux router are live ip and not natted. > > I am not sure how to setup the static-rps and/or the bootstrap for the > pimsm4 directive to get it to work properly. which interface needs to > be the BSR or RP or just use the static rps. I'm assuming that you don't already have any other multicast routers deployed in your network, possibly apart from the Cisco device. Is it configured as an RP? For a 3 armed router with two stub networks, the placement of the RP is not really going to be an issue, however PIM-SM does need one in order to operate. I can foresee that you may have trouble with your video transmission requirement -- you don't specify if your upstream ISP runs PIM-SM already, or if you are able to peer PIM-SM with them. XORP does not currently support PIM-DM, and it is generally not widely deployed. If your ISP only speaks IGMP on customer networks, you will most likely have to install a third-party IGMP proxy, as XORP does not currently support IGMP proxy mode. If your ISP does however speak PIM-SM on customer networks, then you will have to contact them and ask them for the RP configuration details, assuming they do not support the PIM bootstrap mechanism. There is also auto-rp, however, that is somewhat proprietary in nature, see Pavlin's old post on the subject: http://www.mail-archive.com/xorp-hackers at icir.org/msg00001.html ...if your Cisco device has the necessary IOS feature set, then you might just have to use that instead. In your situation, I would configure your XORP router to be a static RP using the configuration you've posted, which looks as though it should work properly, although obviously I haven't tested it. Your zone scope should be OK for a stub/customer network, but I wouldn't use that configuration in a carrier network. If you are having problems with this config, please post the error messages and/or describe the behaviour you see. Currently there is a known issue with the MFEA that each network port must have active Ethernet link pulse in order to work. There is a community patch for this which currently isn't in upstream source, please see the xorp-users archives. thanks, BMS From william at losrios.edu Thu May 28 08:43:21 2009 From: william at losrios.edu (Williams, Mark) Date: Thu, 28 May 2009 08:43:21 -0700 Subject: [Xorp-users] XORP - Linux - Multicast setup help In-Reply-To: <957b38290905222214k1aa0ecd3k6d13392db72fe9b8@mail.gmail.com> References: <957b38290905222214k1aa0ecd3k6d13392db72fe9b8@mail.gmail.com> Message-ID: <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DB1@lrccd-exch05.LRCCD.ad.losrios.edu> You could use static RPs or the bootstrap mechanism. If you go with static, you need to ensure that the RP set is consistent among all routers in the PIM domain. In your scenario, I would configure the Cisco as the BSR, and configure it as an RP-candidate for the ghost multicast group range 224.77.0.0/16, since the Cisco will be the DR for that group. One problem that I see in your setup is the Xorp box does not have a route to the other side of the Cisco router, where your multicast sender is. This will cause problems when the Xorp system wants to send the source-specific join message back up the multicast tree. I'm not sure if you can use kernel routes, or if you will have to use a routing protocol within Xorp. -Mark Williams ________________________________ From: xorp-users-bounces at xorp.org [mailto:xorp-users-bounces at xorp.org] On Behalf Of Lucky Stud Sent: Friday, May 22, 2009 22:14 To: xorp-users at xorp.org Subject: [Xorp-users] XORP - Linux - Multicast setup help Hi, I am trying to setup the xorp on a linux (RHEL4) machine with 3 interfaces which routes the traffic for two subnets (setup below). eth0 is directly connected to the a cisco router. I need to enable multicast routing in order to be able to use norton ghost and multicast to image the machines in the 192.0/26 and 192.64/26 subnets and also to receive a multicast video feed transmission from internet. cisco router is setup properly as I can use multicast on all other subnets that are directly connected to the cisco router. The two subnets behind the linux router are live ip and not natted. I am not sure how to setup the static-rps and/or the bootstrap for the pimsm4 directive to get it to work properly. which interface needs to be the BSR or RP or just use the static rps. Any advise is greatly appreciated. I downloaded the latest version of xorp from official site, compiled and installed it. Thanks. [muulticast receivers] 192.0/26 ---------- [192.1 (eth1) \ | XORP-linux router- 67.141 (eth0) ]-----------------------[ 67.129 Cisco Router ]-----------> out / multicast sender [muulticast receivers] 192.64/26---------- [192.65 (eth2) / route -n 192.0 0.0.0.0 255.255.255.192 U 0 0 0 eth1 192.64 0.0.0.0 255.255.255.192 U 0 0 0 eth2 67.128 0.0.0.0 255.255.255.192 U 0 0 0 eth0 0.0.0.0 142.204.67.129 0.0.0.0 UG 0 0 0 eth0 config.boot interfaces { restore-original-config-on-shutdown: true; interface eth0 { default-system-config } interface eth1 { default-system-config } interface eth2 { default-system-config } } fea { unicast-forwarding4 { disable: false } unicast-forwarding6 { disable: true } } plumbing { mfea4 { disable: false interface eth0 { vif eth0 { } } interface eth1{ vif eth1 { } } interface eth2{ vif eth2 { } } interface register_vif { vif register_vif { } } traceoptions { flag all { disable: false } } } } protocols { igmp { interface eth0 { vif eth0 { } } interface eth1 { vif eth1 { } } interface eth2 { vif eth2 { } } traceoptions { flag all { disable: false } } } } protocols { pimsm4 { disable: false interface eth0 { vif eth0 { } } interface eth1 { vif eth1 { } } interface eth2 { vif eth2 { } } interface register_vif { vif register_vif { } } /* static-rps { rp 142.204.67.129 { group-prefix 224.0.0.0/4 { } } } */ bootstrap { disable: false cand-bsr { scope-zone 224.0.0.0/4 { /* is-scope-zone: false */ cand-bsr-by-vif-name: "eth0" /* cand-bsr-by-vif-addr: 10.10.10.10 */ /* bsr-priority: 1 */ /* hash-mask-len: 30 */ } } cand-rp { group-prefix 224.0.0.0/4 { /* is-scope-zone: false */ cand-rp-by-vif-name: "eth0" /* cand-rp-by-vif-addr: 10.10.10.10 */ /* rp-priority: 192 */ /* rp-holdtime: 150 */ } } } traceoptions { flag all { disable: false } } } } protocols { fib2mrib { } } -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090528/cbc58bb2/attachment-0001.html From william at losrios.edu Thu May 28 15:08:00 2009 From: william at losrios.edu (Williams, Mark) Date: Thu, 28 May 2009 15:08:00 -0700 Subject: [Xorp-users] Xorp system as DR in backbone = disaster Message-ID: <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DB4@lrccd-exch05.LRCCD.ad.losrios.edu> This morning, I attempted to deploy a Xorp system configured as an OSPF ABR. One interface was in the backbone, and all others were in area 0.0.0.2. When the interface in the backbone peered with the other routers in the backbone, the Xorp system became the DR (it had the highest priority). When it became the DR, all of the other routers in the backbone lost their routes through other backbone-connected routers, including the default route, which is injected by an ASBR also in the backbone. The whole topology fell apart. Other routers in the backbone showed that they were peered with the Xorp system, and that it was the DR, but they were not picking up any routes from it. The system has an iptables rule set, but nothing in the logs indicated that any OSPF related traffic was dropped. Has anyone had similar problems with Xorp as DR, particularly if it is the backbone area? -Mark Williams -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090528/ee41c133/attachment.html From bms at incunabulum.net Fri May 29 04:51:35 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 29 May 2009 12:51:35 +0100 Subject: [Xorp-users] Xorp system as DR in backbone = disaster In-Reply-To: <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DB4@lrccd-exch05.LRCCD.ad.losrios.edu> References: <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DB4@lrccd-exch05.LRCCD.ad.losrios.edu> Message-ID: <4A1FCC47.5010506@incunabulum.net> Mark, Thanks for your mail. Williams, Mark wrote: > > This morning, I attempted to deploy a Xorp system configured as an > OSPF ABR. One interface was in the backbone, and all others were in > area 0.0.0.2. When the interface in the backbone peered with the other > routers in the backbone, the Xorp system became the DR (it had the > highest priority). > > > > When it became the DR, all of the other routers in the backbone lost > their routes through other backbone-connected routers, including the > default route, which is injected by an ASBR also in the backbone. The > whole topology fell apart. Other routers in the backbone showed that > they were peered with the Xorp system, and that it was the DR, but > they were not picking up any routes from it. The system has an > iptables rule set, but nothing in the logs indicated that any OSPF > related traffic was dropped. > > > > Has anyone had similar problems with Xorp as DR, particularly if it is > the backbone area? > Sorry to hear that. I have not encountered similar reports. Can you raise a Bugzilla report about this and attach the configuration from your XORP ABR node? thanks, BMS From luckystud55 at gmail.com Fri May 29 09:16:52 2009 From: luckystud55 at gmail.com (Lucky Stud) Date: Fri, 29 May 2009 12:16:52 -0400 Subject: [Xorp-users] XORP - Linux - Multicast setup help In-Reply-To: <4A1EADA2.6070309@incunabulum.net> References: <957b38290905222214k1aa0ecd3k6d13392db72fe9b8@mail.gmail.com> <4A1EADA2.6070309@incunabulum.net> Message-ID: <957b38290905290916x54039cd4s6dfa17b6adb69a1e@mail.gmail.com> Thanks for you help, Bruce Simpson and Mark Williams. We eventually got it to work. Since we were not running RP or BSR on our network, we used the static RP to point to our ISP's RP and it works fine. Also the xorp didn't like having the Cisco HSRP as the default gateway for the box, so we used the static route to have the mrib point to the active physical interface IP of cisco router (since we wanted the default gateway of the box to point to the HSRP). I've posted the config below. It kept coming with the warning that upstream neighbor for RP not found. I hope these information would be of help to others who might exprience the same issue now or later. JoinDesired(*,G) = true: upstream neighbor for RP for group : not found Thanks for your help again. Pim Sparse works amazingly good with xorp now but we are having difficulty with dense multicast (ghost sessions and such). protocols { static { mrib-route 0.0.0.0/0 { next-hop: metric: 1 } } } static-rps { rp { group-prefix 224.0.0.0/4 { } } } On Thu, May 28, 2009 at 11:28 AM, Bruce Simpson wrote: > Hello, > > Sorry for the delay in replying to your message. > > Lucky Stud wrote: > >> Hi, >> >> I am trying to setup the xorp on a linux (RHEL4) machine with 3 interfaces >> which routes the traffic for two subnets (setup below). eth0 is directly >> connected to the a cisco router. I need to enable multicast routing in order >> to be able to use norton ghost and multicast to image the machines in the >> 192.0/26 and 192.64/26 subnets and also to receive a multicast video feed >> transmission from internet. cisco router is setup properly as I can use >> multicast on all other subnets that are directly connected to the cisco >> router. The two subnets behind the linux router are live ip and not natted. >> >> I am not sure how to setup the static-rps and/or the bootstrap for the >> pimsm4 directive to get it to work properly. which interface needs to be the >> BSR or RP or just use the static rps. >> > > I'm assuming that you don't already have any other multicast routers > deployed in your network, possibly apart from the Cisco device. Is it > configured as an RP? > > For a 3 armed router with two stub networks, the placement of the RP is > not really going to be an issue, however PIM-SM does need one in order to > operate. I can foresee that you may have trouble with your video > transmission requirement -- you don't specify if your upstream ISP runs > PIM-SM already, or if you are able to peer PIM-SM with them. XORP does not > currently support PIM-DM, and it is generally not widely deployed. > > If your ISP only speaks IGMP on customer networks, you will most likely > have to install a third-party IGMP proxy, as XORP does not currently support > IGMP proxy mode. If your ISP does however speak PIM-SM on customer networks, > then you will have to contact them and ask them for the RP configuration > details, assuming they do not support the PIM bootstrap mechanism. > > There is also auto-rp, however, that is somewhat proprietary in nature, see > Pavlin's old post on the subject: > http://www.mail-archive.com/xorp-hackers at icir.org/msg00001.html > ...if your Cisco device has the necessary IOS feature set, then you might > just have to use that instead. > > In your situation, I would configure your XORP router to be a static RP > using the configuration you've posted, which looks as though it should work > properly, although obviously I haven't tested it. Your zone scope should be > OK for a stub/customer network, but I wouldn't use that configuration in a > carrier network. > > If you are having problems with this config, please post the error messages > and/or describe the behaviour you see. Currently there is a known issue with > the MFEA that each network port must have active Ethernet link pulse in > order to work. There is a community patch for this which currently isn't in > upstream source, please see the xorp-users archives. > > thanks, > BMS > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090529/8f16a974/attachment.html From william at losrios.edu Wed May 27 13:07:27 2009 From: william at losrios.edu (Williams, Mark) Date: Wed, 27 May 2009 13:07:27 -0700 Subject: [Xorp-users] ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! In-Reply-To: <2945798e0905271211p30ff567clf7182ec2bc756f3e@mail.gmail.com> References: <2945798e0905271211p30ff567clf7182ec2bc756f3e@mail.gmail.com> Message-ID: <1B62E6082D9F034394DB2FBAC03D1A3860E7AB6DAD@lrccd-exch05.LRCCD.ad.losrios.edu> >From the error messages, I would guess that your kernel was not configured to support multicast routing and PIM-SM v2. Do the files /proc/net/ip_mr_cache and /proc/net/ip_mr_vif exist on your system? -Mark Williams ________________________________ From: xorp-users-bounces at xorp.org [mailto:xorp-users-bounces at xorp.org] On Behalf Of Petar Bojovic Sent: Wednesday, May 27, 2009 12:12 To: xorp-users at xorp.org Subject: [Xorp-users] ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! Hi, I am trying to simulate similar situation as in post: http://mailman.icsi.berkeley.edu/pipermail/xorp-users/2007-February/001680.html I am using Voyage Linux platform based on Ubuntu. Have compiled and installed XORP 1.6 from source, and set same config file as in mentioned post. config.boot interfaces { interface eth0 { description: "upstream interface" disable: false default-system-config } interface eth1{ description: "downstream interface" disable: false default-system-config } } fea { unicast-forwarding4 { disable: false } } plumbing { mfea4 { disable: false interface eth0 { vif eth0 { disable: false } } interface eth1 { vif eth1 { disable: false } } interface register_vif { vif register_vif { disable: false } } traceoptions { flag all { disable: false } } } } protocols { igmp { disable: false interface eth0 { vif eth0 { version: 3 disable: false } } interface eth1 { vif eth1 { version: 3 disable: false } } traceoptions { flag all { disable: false } } } } protocols { pimsm4 { disable: false interface eth0 { vif eth0 { disable: false } } interface eth1 { vif eth1 { disable: false } } interface register_vif { vif register_vif { disable: false } } traceoptions { flag all { disable: false } } } } protocols { fib2mrib { disable: false } ifconfig -> eth0 Link encap:Ethernet HWaddr 00:0D:B9:14:D8:00 inet addr:172.16.1.245 Bcast:172.16.1.255 Mask:255.255.255.0 inet6 addr: fe80::20d:b9ff:fe14:d800/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:20843 errors:0 dropped:0 overruns:0 frame:0 TX packets:9830 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1453533 (1.3 MiB) TX bytes:2335612 (2.2 MiB) Interrupt:10 eth1 Link encap:Ethernet HWaddr 00:0D:B9:14:D8:01 inet addr:172.16.2.245 Bcast:172.16.2.255 Mask:255.255.255.0 inet6 addr: fe80::20d:b9ff:fe14:d801/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1485 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:106289 (103.7 KiB) TX bytes:468 (468.0 b) Interrupt:12 Base address:0x4000 After running xorp_rtrmgr, I get next output: [ 2009/05/27 19:06:42 INFO xorp_rtrmgr:10011 RTRMGR +239 master_conf_tree.cc execute ] Changed modules: interfaces, firewall, fea, mfea4, rib, fib2mrib, igmp, pimsm4 [ 2009/05/27 19:06:42 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: interfaces (fea/xorp_fea) [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] MFEA enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI started [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] MFEA enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI enabled [ 2009/05/27 19:06:43 INFO xorp_fea MFEA ] CLI started [ 2009/05/27 19:06:44 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: firewall (fea/xorp_fea) [ 2009/05/27 19:06:48 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: fea (fea/xorp_fea) [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +96 module_manager.cc execute ] Executing module: mfea4 (fea/xorp_fea) [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface added: Vif[eth0] pif_index: 2 vif_index: 0 addr: 172.16.1.245 subnet: 172.16.1.0/24 broadcast: 172.16.1.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface added: Vif[eth1] pif_index: 3 vif_index: 1 addr: 172.16.2.245 subnet: 172.16.2.0/24 broadcast: 172.16.2.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] MFEA started [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface enabled Vif[eth0] pif_index: 2 vif_index: 0 addr: 172.16.1.245 subnet: 172.16.1.0/24 broadcast: 172.16.1.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 DOWN IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface started: Vif[eth0] pif_index: 2 vif_index: 0 addr: 172.16.1.245 subnet: 172.16.1.0/24 broadcast: 172.16.1.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 UP IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface added: Vif[register_vif] pif_index: 2 vif_index: 2 addr: 172.16.1.245 subnet: 172.16.1.245/32 broadcast: 172.16.1.245 peer: 0.0.0.0 Flags: PIM_REGISTER UNDERLYING_VIF_UP MTU: 1500 [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface enabled Vif[eth1] pif_index: 3 vif_index: 1 addr: 172.16.2.245 subnet: 172.16.2.0/24 broadcast: 172.16.2.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 DOWN IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface started: Vif[eth1] pif_index: 3 vif_index: 1 addr: 172.16.2.245 subnet: 172.16.2.0/24 broadcast: 172.16.2.255 peer: 0.0.0.0 Flags: MULTICAST BROADCAST UNDERLYING_VIF_UP MTU: 1500 UP IPv4 ENABLED [ 2009/05/27 19:06:54 INFO xorp_fea MFEA ] Interface enabled Vif[register_vif] pif_index: 2 vif_index: 2 addr: 172.16.1.245 subnet: 172.16.1.245/32 broadcast: 172.16.1.245 peer: 0.0.0.0 Flags: PIM_REGISTER UNDERLYING_VIF_UP MTU: 1500 DOWN IPv4 ENABLED [ 2009/05/27 19:06:54 ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument [ 2009/05/27 19:06:54 ERROR xorp_fea:10013 MFEA +1186 mfea_node.cc start_vif ] Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 WARNING xorp_fea XrlMfeaTarget ] Handling method for mfea/0.1/start_vif failed: XrlCmdError 102 Command failed Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 ERROR xorp_rtrmgr:10011 RTRMGR +681 master_conf_tree.cc commit_pass2_done ] Commit failed: 102 Command failed Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 ERROR xorp_rtrmgr:10011 RTRMGR +251 master_conf_tree.cc config_done ] Configuration failed: 102 Command failed Cannot start vif register_vif: cannot add the multicast vif to the kernel [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +2228 task.cc run_task ] No more tasks to run [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: fea [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: firewall [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: interfaces [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +171 module_manager.cc terminate ] Terminating module: mfea4 [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +194 module_manager.cc terminate ] Killing module: mfea4 [ 2009/05/27 19:06:54 ERROR xorp_rtrmgr:10011 RTRMGR +747 module_manager.cc done_cb ] Command "/usr/lib/xorp/fea/xorp_fea": terminated with signal 15. [ 2009/05/27 19:06:54 INFO xorp_rtrmgr:10011 RTRMGR +282 module_manager.cc module_exited ] Module killed during shutdown: mfea4 I guest that main problem is: ERROR xorp_fea:10013 MFEA +1169 mfea_mrouter.cc add_multicast_vif ] setsockopt(MRT_ADD_VIF, vif register_vif) failed: Invalid argument ! Have tried some older version, same as binary versions, but always same problem. Same problem came up when I test this configuration on FC9. Does anybody have a clue what could make this problem ? Thank you folks ! Petar Bojovic -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090527/12f7cd43/attachment-0001.html From jjw0927 at 126.com Sat May 30 02:34:38 2009 From: jjw0927 at 126.com (jjw0927) Date: Sat, 30 May 2009 17:34:38 +0800 (CST) Subject: [Xorp-users] How do class SnmpEventLoop export events to the snmp agent Message-ID: <25480751.317311243676078263.JavaMail.coremail@bj126app55.126.com> Hi all, I want to know how Class SnmpEventLoop export events to the snmp agent. But after I have read the two source files(xorpevents.cc and xorpevents.hh) , I still do not know very clearly. Can you tell me how the Class SnmpEventLoop works? And I am looking forward for your replies. Best wishes! Jin -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090530/f32815ba/attachment.html From bms at incunabulum.net Sat May 30 07:48:18 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sat, 30 May 2009 15:48:18 +0100 Subject: [Xorp-users] How do class SnmpEventLoop export events to the snmp agent In-Reply-To: <25480751.317311243676078263.JavaMail.coremail@bj126app55.126.com> References: <25480751.317311243676078263.JavaMail.coremail@bj126app55.126.com> Message-ID: <4A214732.7090502@incunabulum.net> jjw0927 wrote: > Hi all, > I want to know how Class SnmpEventLoop export events to the snmp > agent. But after I have read the two source files(xorpevents.cc and > xorpevents.hh) , I still do not know very clearly. > > Can you tell me how the Class SnmpEventLoop works? And I am looking > forward for your replies It is basically a hack to run the XORP event loop from the Net-SNMP library event loop. I've been of the opinion for a while that this should be rewritten to be the other way around, but lacking motivation and/or funding to perform the task, I have not. thanks BMS From jjw0927 at 126.com Sat May 30 23:31:41 2009 From: jjw0927 at 126.com (jjw0927) Date: Sun, 31 May 2009 14:31:41 +0800 (CST) Subject: [Xorp-users] How do class SnmpEventLoop export events to the snmp agent In-Reply-To: <4A214732.7090502@incunabulum.net> References: <4A214732.7090502@incunabulum.net> <25480751.317311243676078263.JavaMail.coremail@bj126app55.126.com> Message-ID: <28742318.186831243751501992.JavaMail.coremail@bj126app33.126.com> ?2009-05-30 22:48:18?"Bruce Simpson" ??? >jjw0927 wrote: >> Hi all, >> I want to know how Class SnmpEventLoop export events to the snmp >> agent. But after I have read the two source files(xorpevents.cc and >> xorpevents.hh) , I still do not know very clearly. >> >> Can you tell me how the Class SnmpEventLoop works? And I am looking >> forward for your replies > >It is basically a hack to run the XORP event loop from the Net-SNMP >library event loop. > >I've been of the opinion for a while that this should be rewritten to be >the other way around, but lacking motivation and/or funding to perform >the task, I have not. > >thanks >BMS Thanks for Bruce's reply. But can you tell me the detail about the communication between the Class SnmpEventLoop and the mibs(such as bgp4 and ospf)? I do not know which function in the files of xorpevents.cc and xorpevents.hh to get the informantion in mibs. Best wishes Jin -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20090531/8906ad37/attachment.html From bms at incunabulum.net Sun May 31 09:06:05 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sun, 31 May 2009 17:06:05 +0100 Subject: [Xorp-users] How do class SnmpEventLoop export events to the snmp agent In-Reply-To: <28742318.186831243751501992.JavaMail.coremail@bj126app33.126.com> References: <4A214732.7090502@incunabulum.net> <25480751.317311243676078263.JavaMail.coremail@bj126app55.126.com> <28742318.186831243751501992.JavaMail.coremail@bj126app33.126.com> Message-ID: <4A22AAED.8070102@incunabulum.net> jjw0927 wrote: > ... > Thanks for Bruce's reply. > But can you tell me the detail about the *communication* between the > Class SnmpEventLoop and the mibs(such as bgp4 and ospf)? I do not know > which *function* in the files of xorpevents.cc and xorpevents.hh to > get the informantion in mibs. You probably need to gain more familiarity with Net-SNMP as a toolkit to understand that code, as the coupling is more involved than a single function. I got my Net-SNMP stripes by modifying some 802.11 Wi-Fi routers to export MIBs many years ago. It just embeds a Net-SNMP MIB module in the snmpd process, using the shared library mechanism, to retrieve the MIB information via XRL. Unfortunately this involves taking a copy of the BGP routing table data, making it unsuitable for use in the Default-Free Zone unless you have RAM to burn. The OSPF stuff is just a stub and doesn't have any functionality, ignore it. Last August I did a feasibility study / technical report for XORP, Inc. to look at the problem in detail. I've been of the opinion that AgentX should be used, as it side-steps the need to depend on the loader mechanism. I'd regard the SNMP functionality as alpha quality in terms of functionality only. One of the original goals was that XORP router processes wouldn't need any existing changes to be SNMP visible. I don't believe that's possible without changes to the XRL architecture. Net-SNMP does implement macro and function glue to facilitate splicing MIBs into the OID tree, but it's largely only suitable for simple C applications. If you have a large data structure which needs to be exported via SNMP, and it is not mapped into the same address space, you can't benefit from Net-SNMP's glue. These problems are not XORP-specific, or even Net-SNMP-specific, by any means. It's just something which needs to be revisited carefully. thanks, BMS