From mnunna0 at gmail.com Thu Nov 5 14:13:27 2009 From: mnunna0 at gmail.com (mahendra nunna) Date: Thu, 5 Nov 2009 17:13:27 -0500 Subject: [Xorp-users] does xorp replace linux routing deamons Message-ID: <6e49b4d40911051413x50960c78m81c79e210251853b@mail.gmail.com> hi When i run xorp.... will it handle all the packets i m trying to send out of my host.... ? I mean ..... does XORP send the packets out of my host instead the kernel ?.... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091105/0f055e28/attachment.html From bms at incunabulum.net Thu Nov 5 21:21:07 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Fri, 06 Nov 2009 05:21:07 +0000 Subject: [Xorp-users] [Xorp-hackers] does xorp replace linux routing deamons In-Reply-To: <6e49b4d40911051413x50960c78m81c79e210251853b@mail.gmail.com> References: <6e49b4d40911051413x50960c78m81c79e210251853b@mail.gmail.com> Message-ID: <4AF3B243.2000102@incunabulum.net> mahendra nunna wrote: > hi > > When i run xorp.... will it handle all the packets i m trying to send > out of my host.... ? I mean ..... does XORP send the packets out of my > host instead the kernel ?.... No -- the host operating system is responsible for all packet I/O. cheers, BMS From naresh_raga at yahoo.co.in Mon Nov 9 05:37:31 2009 From: naresh_raga at yahoo.co.in (naresh raga) Date: Mon, 9 Nov 2009 19:07:31 +0530 (IST) Subject: [Xorp-users] Xorp router's performance measures In-Reply-To: <4AE469D5.5050506@incunabulum.net> Message-ID: <669509.61717.qm@web94807.mail.in2.yahoo.com> Hi , What are the xorp router performance measures? How these performance measures can be measured.I mean what tools are used in measuring them. Thanks, Naresh. The INTERNET now has a personality. YOURS! See your Yahoo! Homepage. http://in.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091109/741f4386/attachment.html From bms at incunabulum.net Mon Nov 9 08:19:57 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Mon, 09 Nov 2009 16:19:57 +0000 Subject: [Xorp-users] Xorp router's performance measures In-Reply-To: <669509.61717.qm@web94807.mail.in2.yahoo.com> References: <669509.61717.qm@web94807.mail.in2.yahoo.com> Message-ID: <4AF8412D.2030508@incunabulum.net> Hi Naresh, Thanks for your question about XORP performance measurements. naresh raga wrote: > Hi , > What are the xorp router performance measures? > How these performance measures can be measured.I mean what tools are > used in measuring them. > Which aspects of the system, are you specifically interested in measuring for performance? The question on its own is quite vague. There are some simple tools within the source tree, e.g. the profiling framework, and the IPC benchmark, although these only measure very specific aspects of the system, which probably aren't particularly useful for drawing comparisons e.g. between XORP and other router control planes. For general purpose profiling, a tool like valgrind or oprofile is probably more useful, although these tools still need to be wielded appropriately, to obtain useful information about the system's behaviour. It's difficult to publish benchmarks for something like community XORP. This is especially because community XORP alone is not a top-to-bottom integrated router; it needs a host environment (e.g. BSD, Linux) and a forwarding plane implementation. It's easy to make apples-with-oranges comparisons, and these are unscientific. The host operating system or forwarding plane are going to make some difference to measurements made here, so this has to be considered when making measurements of this kind. We welcome third party contributions in this area. I understand Ben Greear is doing some performance measurements, but these are mostly specific to his own changes. thanks, BMS From syed.khalid at xorp.net Tue Nov 10 07:22:24 2009 From: syed.khalid at xorp.net (Syed Khalid) Date: Tue, 10 Nov 2009 07:22:24 -0800 Subject: [Xorp-users] Xorp router's performance measures In-Reply-To: <4AF8412D.2030508@incunabulum.net> References: <669509.61717.qm@web94807.mail.in2.yahoo.com> <4AF8412D.2030508@incunabulum.net> Message-ID: Hello Naresh I might add that vendors like Ixia, Agilent and Spirent provide tools for measuring different aspects of routing performance such as 1. Route convergence 2. Route capacity Each of these vendors have fairly detailed descriptions of the above scenarios. Syed On Mon, Nov 9, 2009 at 8:19 AM, Bruce Simpson wrote: > Hi Naresh, > > Thanks for your question about XORP performance measurements. > > naresh raga wrote: > > Hi , > > What are the xorp router performance measures? > > How these performance measures can be measured.I mean what tools are > > used in measuring them. > > > > Which aspects of the system, are you specifically interested in > measuring for performance? The question on its own is quite vague. > > There are some simple tools within the source tree, e.g. the profiling > framework, and the IPC benchmark, although these only measure very > specific aspects of the system, which probably aren't particularly > useful for drawing comparisons e.g. between XORP and other router > control planes. > > For general purpose profiling, a tool like valgrind or oprofile is > probably more useful, although these tools still need to be wielded > appropriately, to obtain useful information about the system's behaviour. > > It's difficult to publish benchmarks for something like community XORP. > This is especially because community XORP alone is not a top-to-bottom > integrated router; it needs a host environment (e.g. BSD, Linux) and a > forwarding plane implementation. > > It's easy to make apples-with-oranges comparisons, and these are > unscientific. The host operating system or forwarding plane are going to > make some difference to measurements made here, so this has to be > considered when making measurements of this kind. > > We welcome third party contributions in this area. I understand Ben > Greear is doing some performance measurements, but these are mostly > specific to his own changes. > > 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/20091110/02841d04/attachment.html From murad03_khan at yahoo.com Tue Nov 10 23:45:21 2009 From: murad03_khan at yahoo.com (Murad) Date: Tue, 10 Nov 2009 23:45:21 -0800 (PST) Subject: [Xorp-users] how to start XORP in background? Message-ID: <129888.56544.qm@web52407.mail.re2.yahoo.com> Hi ALL, Please if anybody tell me how to start xorp_rtrmgr in background? I have XORP release 1.6 and currently use below command to start xorp_rtrmgr: xorp_rtrmgr -b config.boot Thank you. Mvh, /Murad __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091110/5f0c3394/attachment.html From bms at incunabulum.net Wed Nov 11 05:34:16 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Wed, 11 Nov 2009 13:34:16 +0000 Subject: [Xorp-users] how to start XORP in background? In-Reply-To: <129888.56544.qm@web52407.mail.re2.yahoo.com> References: <129888.56544.qm@web52407.mail.re2.yahoo.com> Message-ID: <4AFABD58.2000701@incunabulum.net> Murad wrote: > Hi ALL, > > Please if anybody tell me how to start xorp_rtrmgr in background? I > have XORP release 1.6 and currently use below command to start > xorp_rtrmgr: > xorp_rtrmgr -b config.boot > There is a '-d' switch which was added to support daemon mode, however, it hasn't yet been added to the LaTeX documentation. Off the top of my head I think this switch is in 1.6. thanks, BMS From iprebeg at freebsd.org Wed Nov 11 09:52:07 2009 From: iprebeg at freebsd.org (iprebeg at freebsd.org) Date: Wed, 11 Nov 2009 18:52:07 +0100 Subject: [Xorp-users] Mcast routing on FreeBSD 7.2 and 8.0 Message-ID: <20091111175207.GA29767@valeria.zesoi.fer.hr> Hi all I'm trying to set up mcast routing in very simple scenario like this: --------------- ----------10.0.1.1---| FreeBSD-8.0 |---10.0.2.1---------- | --------------- | | | | | 10.0.1.2 10.0.2.2 | | --------------- --------------- | FreeBSD-8.0 | | FreeBSD-8.0 | --------------- --------------- So, 10.0.1.2 listens 235.0.0.1 with mcastrecv from mcast-tools and triggers IGMPv3 join for source 10.0.2.2. 10.0.2.2 uses mcastsend and generates mcasted traffic on address 235.0.0.1. Traffic is detected with tcpdump on 10.0.2.1 and IGMP join is detected on 10.0.1.1, also tcpdump. 10.0.1.1/10.0.2.1 runs xorp-1.6 (same if built with or without advanced multicast API) and kernel built with MROUTING option. The problem is traffic doesn't get from 10.0.2.1 to 10.0.1.1. Also, XORP output doesn't show any reaction on incoming IGMP joins or mcasted traffic. My config.boot is here: http://pastebin.com/m7e00583c Output is here: http://pastebin.com/m485df247 What is going on on lines 26 and 38? Why these warnings? Thanks in advance Ivor Prebeg -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: not available Url : http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091111/69bfc5b4/attachment.bin From naresh_raga at yahoo.co.in Wed Nov 11 12:27:27 2009 From: naresh_raga at yahoo.co.in (naresh raga) Date: Thu, 12 Nov 2009 01:57:27 +0530 (IST) Subject: [Xorp-users] Xorp router's performance measures In-Reply-To: <4AF8412D.2030508@incunabulum.net> Message-ID: <497853.76488.qm@web94807.mail.in2.yahoo.com> Hi all, >Which aspects of the system, are you specifically interested in measuring for >performance? The question on its own is quite vague. I am learning .So my questions can be vague.I have asked in general what are a router's performance figures.I just want to know in what all aspects a router's performance can be measured.For eg throughput,latency are two such aspects. Are there any testbenches/test beds to measure such aspects.I think to evaluate xorp performance,there should be some aspects where xorp router platform would be put to test.What are those aspects.And what are the test beds(procedures) used. ? Can anyone please guide me if there is good material in the net.I am unable to find any regarding router's performance figures and their test beds(procedures to measure them). Sorry,if you find my English bad. Thanks, Naresh. The INTERNET now has a personality. YOURS! See your Yahoo! Homepage. http://in.yahoo.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091112/b52dcd6d/attachment.html From bms at incunabulum.net Thu Nov 12 04:21:25 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 12 Nov 2009 12:21:25 +0000 Subject: [Xorp-users] Mcast routing on FreeBSD 7.2 and 8.0 In-Reply-To: <20091111175207.GA29767@valeria.zesoi.fer.hr> References: <20091111175207.GA29767@valeria.zesoi.fer.hr> Message-ID: <4AFBFDC5.8010308@incunabulum.net> iprebeg at freebsd.org wrote: > The problem is traffic doesn't get from 10.0.2.1 to 10.0.1.1. > Also, XORP output doesn't show any reaction on incoming IGMP joins or mcasted > traffic. > Please refer to my reply off-list about what needs to be done here. > My config.boot is here: http://pastebin.com/m7e00583c > Output is here: http://pastebin.com/m485df247 > > What is going on on lines 26 and 38? Why these warnings? > These are probably just warnings due to the Router Manager anticipating the startup of the IGMP and MLD components, and unrelated to the condition you're seeing. thanks, BMS From bms at incunabulum.net Thu Nov 12 04:27:11 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 12 Nov 2009 12:27:11 +0000 Subject: [Xorp-users] Xorp router's performance measures In-Reply-To: <497853.76488.qm@web94807.mail.in2.yahoo.com> References: <497853.76488.qm@web94807.mail.in2.yahoo.com> Message-ID: <4AFBFF1F.7040607@incunabulum.net> naresh raga wrote: > > > I am learning .So my questions can be vague.I have asked in > general what are a router's performance figures.I just want to > know in what all aspects a router's performance can be > measured.For eg throughput,latency are two such aspects. > A good starting point might be to devise an experiment around typical BGP usage scenarios (e.g. initial bootstrap, flap of a single upstream provider, flap of a tier 1 ISP). Both throughput of updates, and latency of propagation to neighbours, are elements of such use which could be measured. This is of course unrelated to a router's forwarding performance. The NANOG mailing list might be a better place to ask for specific suggestions for tools. thanks, BMS From ron at nscee.edu Fri Nov 13 09:40:56 2009 From: ron at nscee.edu (Ron Young) Date: Fri, 13 Nov 2009 09:40:56 -0800 Subject: [Xorp-users] help testing multicast config Message-ID: <20091113174056.27AEDFF8A2@ron2> Hi everyone: I'm brand new to xorp & multicasting and would like some ideas on how to test my xorp configuration. The configuration was gleaned from previous xorp postings. I have attached a copy of the conf file below. Xorp is running on ubuntu server 9.04. Interface eth1 is the connection to the internet (aaa.bbb.220.194/30). Interface eth2 is the connection to the internal net (aaa.bbb.118.0/24). When I run ssmping from the xorp router, I get the unicast replies but no multicast replies. Is this to be expected if there no multicast clients joined in a group? Any help would be appreciated, thanks -ron root at i2-core:/etc/xorp# cat config.boot fea { unicast-forwarding4 { disable: false } targetname: "fea" } interfaces { interface eth1 { description: "gateway interface" disable: false default-system-config } interface eth2 { description: "Internal network interface" disable: false default-system-config } targetname: "fea" } plumbing { mfea4 { disable: false interface eth2 { vif eth2 { disable: false } } interface eth1 { vif eth1 { disable: false } } interface register_vif { vif register_vif { disable: false } } traceoptions { flag { all { disable: false } } } targetname: "MFEA_4" } } protocols { igmp { interface eth1 { vif eth1 { disable: false } } interface eth2 { vif eth2 { disable: false } } traceoptions { flag { all { disable: false } } } targetname: "IGMP" disable: false } pimsm4 { disable: false interface eth1 { vif eth1 { disable: false dr-priority: 1 } } interface eth2 { vif eth2 { disable: false dr-priority: 1 } } interface register_vif { vif register_vif { disable: false dr-priority: 1 } } static-rps { rp aaa.bbb.118.1 { group-prefix 224.0.0.0/4 { rp-priority: 192 hash-mask-len: 30 } } } traceoptions { flag { all { disable: false } } } targetname: "PIMSM_4" } fib2mrib { disable: false targetname: "fib2mrib" } } root at i2-core:/etc/xorp# root at i2-core:/home/ron# ssmping flo.nrc.ca ssmping joined (S,G) = (132.246.2.20,232.43.211.234) pinging S from aaa.bbb.220.194 unicast from 132.246.2.20, seq=1 dist=11 time=125.006 ms unicast from 132.246.2.20, seq=2 dist=11 time=113.689 ms unicast from 132.246.2.20, seq=3 dist=11 time=113.698 ms unicast from 132.246.2.20, seq=4 dist=11 time=114.030 ms unicast from 132.246.2.20, seq=5 dist=11 time=113.642 ms unicast from 132.246.2.20, seq=6 dist=11 time=113.613 ms unicast from 132.246.2.20, seq=7 dist=11 time=113.637 ms unicast from 132.246.2.20, seq=8 dist=11 time=113.611 ms unicast from 132.246.2.20, seq=9 dist=11 time=113.579 ms unicast from 132.246.2.20, seq=10 dist=11 time=113.698 ms ^C --- 132.246.2.20 statistics --- 10 packets transmitted, time 9205 ms unicast: 10 packets received, 0% packet loss rtt min/avg/max/std-dev = 113.579/114.820/125.006/3.407 ms multicast: 0 packets received, 100% packet loss root at i2-core:/home/ron# =============================================================================== Ron Young, Research Support Analyst National Supercomputing Center for Energy and the Environment 4505 Maryland Parkway, Box 454028, Las Vegas, NV 89154-4028 v (702) 895-4017 / f (702) 895-4156 / email: ron.young at nscee.edu From lizhaous2000 at yahoo.com Mon Nov 16 06:59:29 2009 From: lizhaous2000 at yahoo.com (Li Zhao) Date: Mon, 16 Nov 2009 06:59:29 -0800 (PST) Subject: [Xorp-users] ospfv3 btw xorp and cisco Message-ID: <918702.65583.qm@web58708.mail.re1.yahoo.com> Has any body tried to form ospfv3 (ospf for IPv6) neighbors between xorp router and cisco router. I had experiences that IPv4 works between xorp and cisco, but when i tried to form IPv6 ospf between xorp and cisco, cisco did not receive multicast packets from xorp which already sent out hello messages. From esj at cs.fiu.edu Mon Nov 16 08:00:38 2009 From: esj at cs.fiu.edu (Eric S. Johnson) Date: Mon, 16 Nov 2009 11:00:38 -0500 Subject: [Xorp-users] vrrp in xorp 1.6 Message-ID: <20091116160038.A1898B88418@cheetah.cs.fiu.edu> Does vrrp work in xorp 1.6? I ask because I tried to use it (linux centos 5 base system) and I found first that the templates (.tp and .cmds) for vrrp are not installed by default. Simple standard configure: ./configure --prefix=/opt/router/xorp So I installed them manually, and it then seemed to understand the commands, but it didn't seem to work. I did not do any extensive debugging yet. Is this supposed to work? E From diego360xen at gmail.com Mon Nov 16 08:03:57 2009 From: diego360xen at gmail.com (=?ISO-8859-1?Q?Diego_Augusto_Pinzon_Carre=F1o?=) Date: Mon, 16 Nov 2009 11:03:57 -0500 Subject: [Xorp-users] cc1plus: warnings being treated as errors Message-ID: I'm just doing a scons nothing fancy at this point. cc1plus: warnings being treated as errors fea/data_plane/io/io_ip_socket.cc: In member function ?void IoIpSocket::proto_socket_read(XorpFd, IoEventType)?: fea/data_plane/io/io_ip_socket.cc:1614: error: ?int inet6_option_next(const cmsghdr*, uint8_t**)? is deprecated (declared at /usr/include/netinet/in.h:498) fea/data_plane/io/io_ip_socket.cc:1614: error: ?int inet6_option_next(const cmsghdr*, uint8_t**)? is deprecated (declared at /usr/include/netinet/in.h:498) fea/data_plane/io/io_ip_socket.cc: In member function ?virtual int IoIpSocket::send_packet(const std::string&, const std::string&, const IPvX&, const IPvX&, int32_t, int32_t, bool, bool, const std::vector >&, const std::vector >, std::allocator > > >&, const std::vector >&, std::string&)?: fea/data_plane/io/io_ip_socket.cc:2215: error: ?int inet6_option_space(int)? is deprecated (declared at /usr/include/netinet/in.h:487) fea/data_plane/io/io_ip_socket.cc:2215: error: ?int inet6_option_space(int)? is deprecated (declared at /usr/include/netinet/in.h:487) fea/data_plane/io/io_ip_socket.cc:2312: error: ?int inet6_option_init(void*, cmsghdr**, int)? is deprecated (declared at /usr/include/netinet/in.h:489) fea/data_plane/io/io_ip_socket.cc:2312: error: ?int inet6_option_init(void*, cmsghdr**, int)? is deprecated (declared at /usr/include/netinet/in.h:489) fea/data_plane/io/io_ip_socket.cc:2316: error: ?int inet6_option_append(cmsghdr*, const uint8_t*, int, int)? is deprecated (declared at /usr/include/netinet/in.h:492) fea/data_plane/io/io_ip_socket.cc:2316: error: ?int inet6_option_append(cmsghdr*, const uint8_t*, int, int)? is deprecated (declared at /usr/include/netinet/in.h:492) scons: *** [obj/x86_64-unknown-linux-gnu/fea/data_plane/io/io_ip_socket.o] Error 1 scons: building terminated because of errors. these is in a fedora 10 64x86 on a 32bits. may i need do a cflag configuration.? i'm using gcc-4.3.2-7(X86_64):it was revision 11575 just donwload a few hours ago from svn i will try it on virtualbox but a i686 to see if there is a issue about architecture. with the before builder gmake on fedora everything was ok at same machine but to relocale a binary openssl. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091116/2f653512/attachment.html From greearb at candelatech.com Mon Nov 16 10:52:55 2009 From: greearb at candelatech.com (Ben Greear) Date: Mon, 16 Nov 2009 10:52:55 -0800 Subject: [Xorp-users] cc1plus: warnings being treated as errors In-Reply-To: References: Message-ID: <4B019F87.1000704@candelatech.com> Diego Augusto Pinzon Carre?o wrote: > I'm just doing a scons nothing fancy at this > point. I'll take a look a bit later. I hit something similar, but I think my patch was accepted upstream, will double-check. For what it's worth, I can compile on stock FC8 and F11, 32 and 64-bit (ie, I'm not installing any special compilers...just whatever comes with 'yum install ...' Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com From greearb at candelatech.com Mon Nov 16 17:54:16 2009 From: greearb at candelatech.com (Ben Greear) Date: Mon, 16 Nov 2009 17:54:16 -0800 Subject: [Xorp-users] Build problem on F11 Message-ID: <4B020248.3070005@candelatech.com> Diego: I think this patch I posted a few months ago might help. Or, you can try my tree with has this and other patches. My tree compiles fine on F8 and F11, as long as you have libpcap devel installed "yum install libpcap-devel" should do the trick: http://www.candelatech.com/oss/xorp-ct.html Thanks, Ben -------- Original Message -------- Subject: Re: [Xorp-users] HAVE_IPV6_MULTICAST never defined in XORP-1.7 on LINUX Date: Fri, 25 Sep 2009 17:02:05 -0700 From: Ben Greear Organization: Candela Technologies To: xorp at sipxx.com CC: xorp users To change back to the original topic: Here is a patch that lets xorp compile IPV6 multicast logic on Linux, at least. I haven't tested functionality. * Properly detect RFC3542 support on Linux. (it probably exists, certainly does on F11) * Check for HAVE_IPV6_MULTICAST * Allow HAVE_IPV6_MULTICAST_ROUTING on Linux. PS. In case anyone ever wants to know, the scons test files are located in obj/[platform]/.sconf_temp Had to run strace to figure out where it was hiding these things :P Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: xorp_mcast6_scons.patch Url: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091116/91a1199d/attachment.ksh From bms at incunabulum.net Tue Nov 17 01:32:35 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 17 Nov 2009 09:32:35 +0000 Subject: [Xorp-users] vrrp in xorp 1.6 In-Reply-To: <20091116160038.A1898B88418@cheetah.cs.fiu.edu> References: <20091116160038.A1898B88418@cheetah.cs.fiu.edu> Message-ID: <4B026DB3.7090903@incunabulum.net> Eric S. Johnson wrote: > Does vrrp work in xorp 1.6? > > I ask because I tried to use it (linux centos 5 base system) > and I found first that the templates (.tp and .cmds) for vrrp > are not installed by default. > This is a known issue with the 1.6 community release. I only noticed it when updating the FreeBSD port at the time (at that particular time, I wasn't active on the XORP project for many months, so wasn't involved in the 1.6 release process). > Simple standard configure: ./configure --prefix=/opt/router/xorp > > So I installed them manually, and it then seemed to understand > the commands, but it didn't seem to work. I did not do any > extensive debugging yet. > I believe you still need to configure vrrp as for any other XORP process in the configuration file, and you will need to have pcap installed and working for it to read and write to the link layer. > Is this supposed to work? I believe the VRRP code is working, but haven't worked with it. You may wish to contact the author, Andrea Bittau, directly. He is sometimes present on these lists. Alternatively please raise a Trac ticket about the issue, and someone else may be able to look at the ticket as time permits. thanks, BMS From bms at incunabulum.net Tue Nov 17 03:03:08 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Tue, 17 Nov 2009 11:03:08 +0000 Subject: [Xorp-users] Mcast routing on FreeBSD 7.2 and 8.0 In-Reply-To: <4AFBFDC5.8010308@incunabulum.net> References: <20091111175207.GA29767@valeria.zesoi.fer.hr> <4AFBFDC5.8010308@incunabulum.net> Message-ID: <4B0282EC.6010603@incunabulum.net> Bruce Simpson wrote: > iprebeg at freebsd.org wrote: > >> The problem is traffic doesn't get from 10.0.2.1 to 10.0.1.1. >> Also, XORP output doesn't show any reaction on incoming IGMP joins or mcasted >> traffic. >> >> > > Please refer to my reply off-list about what needs to be done here. > An appropriate fix is committed to HEAD, and pulled up on the stable/8 SVN branch. The bug was found by Franz Struwig. Unfortunately this fix was too late for the FreeBSD 8.0 release, so an appropriate erratum will need to be issued by the Release Engineers there. Once again, apologies that this fix didn't make it in sooner. thanks, BMS From greearb at candelatech.com Tue Nov 17 19:07:44 2009 From: greearb at candelatech.com (Ben Greear) Date: Tue, 17 Nov 2009 19:07:44 -0800 Subject: [Xorp-users] Build problem on F11 In-Reply-To: <4B020248.3070005@candelatech.com> References: <4B020248.3070005@candelatech.com> Message-ID: <4B036500.3040601@candelatech.com> On 11/16/2009 05:54 PM, Ben Greear wrote: > Diego: I think this patch I posted a few months ago might help. Actually, the problem appears to be (on my machine at least), that the 'host-os' is 'linux-gnu', but the allconfig.py was matching on just 'linux'. I don't know enough python to make it do a looser match (case insensitive substring 'linux') would be my preference, but the patch attached fixes the problem for me. As an aside, scons doesn't detect that things are changed in allconfig.py correctly, or I would have noticed this bug earlier (I have been just using cashed values from before the os became 'linux-gnu'.) I still am not sure where the 'linux-gnu' comes from. Bruce: Please apply this or something similar. And, if you can build-test on Linux, be sure to rm -fr the obj dir first to make sure the changes are actually tested. For what it's worth, this patch on top of the generic tree makes it build for me on F11. Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: xorp_linux-gnu.patch Url: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091117/27954842/attachment.ksh From greearb at candelatech.com Tue Nov 17 19:37:27 2009 From: greearb at candelatech.com (Ben Greear) Date: Tue, 17 Nov 2009 19:37:27 -0800 Subject: [Xorp-users] Build problem on F11 In-Reply-To: <4B036500.3040601@candelatech.com> References: <4B020248.3070005@candelatech.com> <4B036500.3040601@candelatech.com> Message-ID: <4B036BF7.3090005@candelatech.com> On 11/17/2009 07:07 PM, Ben Greear wrote: > On 11/16/2009 05:54 PM, Ben Greear wrote: >> Diego: I think this patch I posted a few months ago might help. > > Actually, the problem appears to be (on my machine at least), that > the 'host-os' is 'linux-gnu', but the allconfig.py was matching on > just 'linux'. I don't know enough python to make it do a looser > match (case insensitive substring 'linux') would be my preference, > but the patch attached fixes the problem for me. > > As an aside, scons doesn't detect that things are changed in allconfig.py > correctly, or I would have noticed this bug earlier (I have been > just using cashed values from before the os became 'linux-gnu'.) Or, maybe I just hadn't merged in upstream recently enough to see the bug. I think it might have appeared rather recently in some of the scons changes to detect the OS. Guess not many people are trying to build the official tree on Linux! Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com From bms at incunabulum.net Wed Nov 18 06:07:48 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Wed, 18 Nov 2009 14:07:48 +0000 Subject: [Xorp-users] Build problem on F11 In-Reply-To: <4B036BF7.3090005@candelatech.com> References: <4B020248.3070005@candelatech.com> <4B036500.3040601@candelatech.com> <4B036BF7.3090005@candelatech.com> Message-ID: <4B03FFB4.8050407@incunabulum.net> Ben Greear wrote: > ... > Or, maybe I just hadn't merged in upstream recently enough to see > the bug. I think it might have appeared rather recently in some > of the scons changes to detect the OS. > > Guess not many people are trying to build the official tree on Linux! > It might have had something to do with the config change which JT committed during October, which imports the GNU config.guess script, and wraps its use from SCons, so that we can use target/host names identical to those of an Autotools build for cross-compilation. Committed, thanks! From globalcouchsurfer at gmail.com Mon Nov 23 04:54:20 2009 From: globalcouchsurfer at gmail.com (CouchSurfer) Date: Mon, 23 Nov 2009 12:54:20 +0000 Subject: [Xorp-users] Help with iBGP configuration Message-ID: <2148c4de0911230454i3fa6d43mff992a5e9a504122@mail.gmail.com> Hi all, I have two linux boxes running xorp. I want them to share routes, possibly using some form of ibgp, but have so far failed. I am wondering if anyone can help. Everything below are ipv4 only, and I am using xorp v 1.6 (2009/01/08). I have two routers, iA and iB. Each of them is connected to the same external AS but to different eBGP peers eA_1, eA_2 and eB_1, eB_2. Each connection with eBGP peers are established using a different interface. The eBGP peers eA_1 and eB_1advertise the same unicast destinations but via different routes (eA_1 and eB_1 respectively). Additionally, eA_2 and eB_2 are pim neighbours with different rendex-vous points eRP_a, eRP_b that know about different multicast groups with different source ip addresses. When run individually, both iA and iB can establish eBGP peerships properly and hosts having them as default gateways can join and leave multicast groups and establish unicast connections to foreign servers OK. Now, I want both of these routers to share both unicast and multicast routes learnt from their respective eBGP peers so that a hosts with with router as default gateway can join and leave multicast groups and establish unicast connections to foreign servers OK. I am not a networking expert so I am a little confused what additional protocols I may need to set up iBGP nor what type of policy I would need to set up or where they should be applied. Using the following pasted configuration, when iA is on and is a default gateway for a host, i found the following: - iA ONLY is on - the host can establish unicast connection with foreigh servers and join goups known by eRP_a; - iB ONLY is on - the host canNOT establish unicast connection with foreigh servers but can join goups known by eRP_b; - iA started BEFORE iB - the host can establish unicast connection with foreigh servers but can join goups known by eRP_b; - iA started AFTER iB - the host can establish unicast connection with foreigh servers but can join goups known by eRP_b; - In any case, whenever iB is on, all multicast group queries seem to only go through iB. Any help will be gratefully appreciated. Thanks in advance ======================================================================================== Config snipet. At present, to create iBGP peership, I do (with x, y = 1, 2, x!=y) protocols { bgp { bgp-id: ip_of_x confederation { identifies: "my_as" } peer "ip_of_y" { local-ip: "ip_of_x" as: my_as next-hop: ip_of_x ipv4-unicast: true ipv4-multiicast: false import: "import-from-ibgp" export: "export-from-ibgp" } /* info about eBPG peers */ } opsf4 { router-id: ip_of_x area 0.0.0.0 { interface "eth0" { vif "eth0" { address ip_of_x } } } } /* additional protocols: igmp, pimsm4, fib2mrib */ } policy { policy-statement "export-to-ibgp" { term 0 { from { protocol: "ospf4" } then { med: 100 accept { } } } term 1 { from { protocol: "connected" } then { med: 100 accept { } } } term 2 { from { protocol: "bgp" } then { med: 100 accept { } } } } policy-statement "import-from-ibgp" { term 0 { from { } then { localpref: 100 accept { } } } } } ================================================================================ From globalcouchsurfer at gmail.com Mon Nov 23 10:01:39 2009 From: globalcouchsurfer at gmail.com (CouchSurfer) Date: Mon, 23 Nov 2009 18:01:39 +0000 Subject: [Xorp-users] Multicast Routing with Multiple xorp routers Message-ID: <2148c4de0911231001x30319fefoc8a3880e2ea0f720@mail.gmail.com> I have two xorp routers connected directly by a L2 switch. They are configured to each have a distinct rendez-vous point RP_224 and RP_225, relating to different (non-overlapping) sets of multicast groups, say 224.7.7.x and 225.7.7.x. Both configured with IGMPv2 and PIM-SM dr-priority 1. I have a host H1 which wants to join two groups 224.7.7.y and 225.7.7.y. It has Router 1 R1 as default gateway. When R1 is on only, H1 can join group 224.7.7.y and on R1 I can see warning messages saying "JoinDesired RP for 225.7.7.y" not found, as R1 is only configured with RP_224" - as expected. Now, in the case where either I have R2 (with RP_225) is on only OR having both R1 and R2 running, H1 can join group 225.7.7.y ONLY, and I can see on R2 a load of warnings "RX PIM_HELLO from (ebgp-router-ip for RP_224) to 224.0.0.13 on vif xxx: source must be directly connected" "RX PIM_HELLO from (R1-interface--ip for ebgp-router-ip) to 224.0.0.13 on vif xxx: source must be directly connected" and on trying to join, I can see a load of warnings "JoinDesired RP for 224.7.7.y" not found, as R1 is only configured with RP_224" "RX IGMP_V2_MEMBERSHIP_REPORT from (R1-interface--ip for ebgp-router-ip) to 224.0.0.13 on vif xxx: source must be directly connected" Further more, on switching off R2, H1 then fails to join any multicasts groups and no more "JoinDesired" warning appears on R1 (as if it just stopped being a MS router) It seems that when both routers are running, R2 just take charge of all pim and igmp messages eventhough it is not configures with those interfaces nor information about the other RP! this happens whether I configure IGMP with version 3 or the order in which I start the routers. Is there a way to configure xorp so that - R2 does not take charge? - I can join multicast groups in both sets? - the two routers perhaps share RP information? Thanks in advance for your help -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091123/5c03aba4/attachment.html From miro_todorovic at yahoo.com Tue Nov 24 04:05:54 2009 From: miro_todorovic at yahoo.com (Miroslav Todorovic) Date: Tue, 24 Nov 2009 04:05:54 -0800 (PST) Subject: [Xorp-users] problem with starting xorp_rtrmgr Message-ID: <540653.32350.qm@web37008.mail.mud.yahoo.com> Hi everyone! I'm new in Xorp and encaunter a problem with starting xorp_rtrmgr. I use VMware for virtualisation and I've installed Ubuntu 9.04, kernel version 2.6.28-11,? as virtual machine.I want to try xorp on this Ubuntu virtual machine. After I download xorp-1.6, and compile it, I encouter a problem when I try to start xorp_rtrmgr. I used this file as config.boot: interfaces { ??? interface eth0 { ??????? description: "data interface" ??????? disable: false ??????? /* default-system-config */ ??????? vif eth0 { ??????????? disable: false ??????????? address 192.168.40.1 { ??????????????? prefix-length: 24 ??????????????? broadcast: 192.168.40.255 ??????????????? disable: false ??????????? } ??????? } ??? } } pe when i type /usr/local/xorp/bin/xorp_rtrmgr -b /usr/local/xorp/config.boot to start xorp_rtrmgr proces, i get massage like this: [ 2009/11/24 12:50:26? INFO xorp_rtrmgr:8533 RTRMGR +249 master_conf_tree.cc execute ] Changed modules: interfaces [ 2009/11/24 12:50:26? INFO xorp_rtrmgr:8533 RTRMGR +101 module_manager.cc execute ] Executing module: interfaces (fea/xorp_fea) [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] MFEA enabled [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI enabled [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI started [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] MFEA enabled [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI enabled [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI started [ 2009/11/24 12:50:28? INFO xorp_rtrmgr:8533 RTRMGR +2233 task.cc run_task ] No more tasks to run ^C[ 2009/11/24 12:51:25? INFO xorp_rtrmgr:8533 RTRMGR +1024 task.cc shutdown ] Shutting down module: interfaces [ 2009/11/24 12:51:25 INFO xorp_fea MFEA ] CLI stopped [ 2009/11/24 12:51:25 INFO xorp_fea MFEA ] CLI stopped [ 2009/11/24 12:51:25? INFO xorp_rtrmgr:8533 RTRMGR +280 module_manager.cc module_exited ] Module normal exit: interfaces [ 2009/11/24 12:51:26? WARNING xorp_rtrmgr:8533 XrlFinderTarget +407 ../xrl/targets/finder_base.cc handle_finder_0_2_resolve_xrl ] Handling method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command failed Target "fea" does not exist or is not enabled. [ 2009/11/24 12:51:27? INFO xorp_rtrmgr:8533 RTRMGR +2233 task.cc run_task ] No more tasks to run Could anyone give some advice? Do I need to use another config.boot file, or to change something else...? Thank you! ? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091124/c6b49aa0/attachment.html From luca.belforte at student.uclouvain.be Tue Nov 24 04:30:44 2009 From: luca.belforte at student.uclouvain.be (Luca Belforte) Date: Tue, 24 Nov 2009 13:30:44 +0100 Subject: [Xorp-users] problem with starting xorp_rtrmgr In-Reply-To: <540653.32350.qm@web37008.mail.mud.yahoo.com> References: <540653.32350.qm@web37008.mail.mud.yahoo.com> Message-ID: <4B0BD1F4.4090501@student.uclouvain.be> Hello, you need to start it in sudo mode: sudo /usr/local/xorp/bin/xorp_rtrmgr -b /usr/local/xorp/config.boot ;) Luca Miroslav Todorovic wrote: > Hi everyone! > > I'm new in Xorp and encaunter a problem with starting xorp_rtrmgr. > I use VMware for virtualisation and I've installed Ubuntu 9.04, kernel > version 2.6.28-11, as virtual machine.I want to try xorp on this > Ubuntu virtual machine. After I download xorp-1.6, and compile it, I > encouter a problem when I try to start xorp_rtrmgr. > I used this file as config.boot: > > interfaces { > interface eth0 { > description: "data interface" > disable: false > /* default-system-config */ > vif eth0 { > disable: false > address 192.168.40.1 { > prefix-length: 24 > broadcast: 192.168.40.255 > disable: false > } > } > } > } > pe > when i type > > /usr/local/xorp/bin/xorp_rtrmgr -b /usr/local/xorp/config.boot > > to start xorp_rtrmgr proces, i get massage like this: > > [ 2009/11/24 12:50:26 INFO xorp_rtrmgr:8533 RTRMGR +249 > master_conf_tree.cc execute ] Changed modules: interfaces > [ 2009/11/24 12:50:26 INFO xorp_rtrmgr:8533 RTRMGR +101 > module_manager.cc execute ] Executing module: interfaces (fea/xorp_fea) > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] MFEA enabled > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI enabled > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI started > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] MFEA enabled > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI enabled > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI started > [ 2009/11/24 12:50:28 INFO xorp_rtrmgr:8533 RTRMGR +2233 task.cc > run_task ] No more tasks to run > ^C[ 2009/11/24 12:51:25 INFO xorp_rtrmgr:8533 RTRMGR +1024 task.cc > shutdown ] Shutting down module: interfaces > [ 2009/11/24 12:51:25 INFO xorp_fea MFEA ] CLI stopped > [ 2009/11/24 12:51:25 INFO xorp_fea MFEA ] CLI stopped > [ 2009/11/24 12:51:25 INFO xorp_rtrmgr:8533 RTRMGR +280 > module_manager.cc module_exited ] Module normal exit: interfaces > [ 2009/11/24 12:51:26 WARNING xorp_rtrmgr:8533 XrlFinderTarget +407 > ../xrl/targets/finder_base.cc handle_finder_0_2_resolve_xrl ] Handling > method for finder/0.2/resolve_xrl failed: XrlCmdError 102 Command > failed Target "fea" does not exist or is not enabled. > [ 2009/11/24 12:51:27 INFO xorp_rtrmgr:8533 RTRMGR +2233 task.cc > run_task ] No more tasks to run > > Could anyone give some advice? > Do I need to use another config.boot file, or to change something else...? > Thank you! > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Xorp-users mailing list > Xorp-users at xorp.org > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-users > From luca.belforte at student.uclouvain.be Wed Nov 25 07:18:51 2009 From: luca.belforte at student.uclouvain.be (Luca Belforte) Date: Wed, 25 Nov 2009 16:18:51 +0100 Subject: [Xorp-users] IBGP received routes, nexthop not changed Message-ID: <4B0D4ADB.6050605@student.uclouvain.be> Hello all, I'm setting up a two AS network, and i use ibgp inside an AS to propagate route. I have : AS1: R1 ------ R2 | AS2: R3 R1 receive via eBGP routes from R1, and transmit it via iBGP to R2. This work, if i look on R2 i see with the command show bgp routes all the routes from R3, but as next hop i have R3. So the route are not chosen (R2 not know the subnet) How i can change the nexthop, before forwarding a eBGP route into iBGP? I put nexthop parameter, as the local-ip, but this doesn't change it. I put R1 config file, R2 and R3 are similar, only some changes in the different address to match the correct info. Thanks Luca -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: R1.boot Url: http://mailman.ICSI.Berkeley.EDU/pipermail/xorp-users/attachments/20091125/e8f323ae/attachment.ksh From bms at incunabulum.net Thu Nov 26 12:17:45 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 26 Nov 2009 20:17:45 +0000 Subject: [Xorp-users] Installed XORP image layout; rtld linker origin Message-ID: <4B0EE269.5010703@incunabulum.net> I'd propose we bite the bullet for 1.7 and adopt a similar installation layout to that of the commercial SVN tree. That is, all shared libraries go in $libdir, which is usually $DESTDIR/lib. This makes life much easier across the tree. At the moment, most of the original XORP regression test suite builds, but not the BGP harness. This is good enough for most. Fortunately, for benchmarking Thrift changes, I can get away with just using the RIB and libxipc itself, but for hard science on the critical BGP-RIB-FEA path, I'd need to go back in and fix that harness. But I am using shared library builds, mostly to save on disk space and CPU on link during hacking. Regressions don't work for shared=true; rtld can't find the XORP libraries. The tests build, but cannot be run. They aren't intended to be installed, but the RPATH still needs to get fixed. In the Automake era, libtool handled all this for us. SCons does not use libtool, which is a blessing, and also a curse. libtool will integrate with Automake's 'gmake check' facility, so Automake test primaries will be invoked with the correct RPATH. The cost of the feature is that a shell script is generated for every dynamically-linked executable. I'm working on a patch to deal with this situation now, and have just checked in the first part of it. The SCons issue tracker has more information on how RPATH is supposed to work, here: http://scons.tigris.org/issues/show_bug.cgi?id=1644 Rev 11583: %%% When allowing the user to override CFLAGS/CXXFLAGS/LINKFLAGS on the command line, don't obliterate what's already there. SCons stashes some useful defaults in there, and RPATH support breaks if you clobber LINKFLAGS in this way. This isn't a bug in SCons, rather how we use it. We need RPATH working to resolve some issues with: 1) installing shared libraries in a retargettable system location (i.e. by using $ORIGIN relative paths in ELF rtld), 2) running the regression tests in a non-cross, shared-library build. Currently, we need to relink every time we change DESTDIR. Don't forget that the RPATH variable needs to see expanded paths, as they are passed to the linker unescaped. For $ORIGIN, each SConscript will need to compute its own relative paths, depending on which directory layout we're using in the stage tree or installation tree. Use Python 2.6 os.path.relpath() for this. While we're here, allow the user to override the AutoTest() builder's default behaviour to ignore errors during a 'scons check' run, by passing 'ignore_check_errors=true' on the command line. %%% cheers, BMS From bms at incunabulum.net Thu Nov 26 12:46:36 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Thu, 26 Nov 2009 20:46:36 +0000 Subject: [Xorp-users] Installed XORP image layout; rtld linker origin In-Reply-To: <4B0EE269.5010703@incunabulum.net> References: <4B0EE269.5010703@incunabulum.net> Message-ID: <4B0EE92C.2000707@incunabulum.net> Woops, this was intended for xorp-hackers at . Feedback on dealing with the installation layout would be great, though. One of the problems with the FreeBSD packaging for XORP, is that the whole shooting match gets installed as a set of subdirectories, most of which don't fit well with the BSD hier(7) layout. The Debian/Fedora packages are affected by this also. There are a few things in the installed tree which are only really relevant to folk actually extending or hacking the code, rather than deploying it. One of these things which got identified was the XRL database. Right now, the Router Manager won't run properly unless the *.xrls files exist in $DESTDIR/xrl/targets/. This is only used for validating each XRL method exists, and is in the XRL target spec, as it's sent to a router process. In a production deployment, this isn't really needed. Ben Greear came up with a patch to eliminate this behaviour, which has yet to be integrated. thanks, BMS From luca.belforte at student.uclouvain.be Thu Nov 26 14:09:33 2009 From: luca.belforte at student.uclouvain.be (Luca Belforte) Date: Thu, 26 Nov 2009 23:09:33 +0100 Subject: [Xorp-users] problem with starting xorp_rtrmgr In-Reply-To: <160044.6683.qm@web37005.mail.mud.yahoo.com> References: <4B0D2740.4080906@student.uclouvain.be> <160044.6683.qm@web37005.mail.mud.yahoo.com> Message-ID: <836838690911261409u34156c2x9a14f6760340a0e1@mail.gmail.com> hello, you need to check the click part (or disable it if you don't want to use it) Luca Belforte On Thu, Nov 26, 2009 at 9:04 PM, Miroslav Todorovic < miro_todorovic at yahoo.com> wrote: > Hello, > > You were right, there was missing fea part. I've changed the config.boot > file, and now used click.boot file from the rtrmgr/config, as a config.boot > file. But I encounter anther problem now. Output message is following: > > > [ 2009/11/26 19:20:40 INFO xorp_rtrmgr:4063 RTRMGR +249 > master_conf_tree.cc execute ] Changed modules: interfaces, firewall, fea > [ 2009/11/26 19:20:40 INFO xorp_rtrmgr:4063 RTRMGR +101 module_manager.cc > execute ] Executing module: interfaces (fea/xorp_fea) > [ 2009/11/26 19:20:41 INFO xorp_fea MFEA ] MFEA enabled > [ 2009/11/26 19:20:41 INFO xorp_fea MFEA ] CLI enabled > [ 2009/11/26 19:20:41 INFO xorp_fea MFEA ] CLI started > [ 2009/11/26 19:20:41 INFO xorp_fea MFEA ] MFEA enabled > [ 2009/11/26 19:20:41 INFO xorp_fea MFEA ] CLI enabled > [ 2009/11/26 19:20:41 INFO xorp_fea MFEA ] CLI started > [ 2009/11/26 19:20:42 INFO xorp_rtrmgr:4063 RTRMGR +101 module_manager.cc > execute ] Executing module: firewall (fea/xorp_fea) > [ 2009/11/26 19:20:46 INFO xorp_rtrmgr:4063 RTRMGR +101 module_manager.cc > execute ] Executing module: fea (fea/xorp_fea) > [ 2009/11/26 19:20:50 ERROR xorp_fea:4064 LIBCOMM +610 comm_sock.c > comm_sock_connect4 ] Error connecting socket (family = 2, remote_addr = > 127.0.0.1, remote_port = 13000): Connection refused > [ 2009/11/26 19:20:50 WARNING xorp_fea FEA ] Could not open user-level > Click socket: Connection refused. Trying again... > [ 2009/11/26 19:20:50 ERROR xorp_fea:4064 LIBCOMM +610 comm_sock.c > comm_sock_connect4 ] Error connecting socket (family = 2, remote_addr = > 127.0.0.1, remote_port = 13000): Connection refused > [ 2009/11/26 19:20:50 WARNING xorp_fea FEA ] Could not open user-level > Click socket: Connection refused. Trying again... > [ 2009/11/26 19:20:51 ERROR xorp_fea:4064 LIBCOMM +610 comm_sock.c > comm_sock_connect4 ] Error connecting socket (family = 2, remote_addr = > 127.0.0.1, remote_port = 13000): Connection refused > [ 2009/11/26 19:20:51 WARNING xorp_fea FEA ] Could not open user-level > Click socket: Connection refused. Trying again... > [ 2009/11/26 19:20:51 ERROR xorp_fea:4064 LIBCOMM +610 comm_sock.c > comm_sock_connect4 ] Error connecting socket (family = 2, remote_addr = > 127.0.0.1, remote_port = 13000): Connection refused > [ 2009/11/26 19:20:51 WARNING xorp_fea XrlFeaTarget ] Handling method for > fea_click/0.1/start_click failed: XrlCmdError 102 Command failed Could not > open user-level Click socket: Connection refused > [ 2009/11/26 19:20:51 ERROR xorp_rtrmgr:4063 RTRMGR +691 > master_conf_tree.cc commit_pass2_done ] Commit failed: 102 Command failed > Could not open user-level Click socket: Connection refused > [ 2009/11/26 19:20:51 ERROR xorp_rtrmgr:4063 RTRMGR +261 > master_conf_tree.cc config_done ] Configuration failed: 102 Command failed > Could not open user-level Click socket: Connection refused > [ 2009/11/26 19:20:51 INFO xorp_rtrmgr:4063 RTRMGR +2233 task.cc run_task > ] No more tasks to run > [ 2009/11/26 19:20:51 INFO xorp_rtrmgr:4063 RTRMGR +176 module_manager.cc > terminate ] Terminating module: fea > [ 2009/11/26 19:20:51 INFO xorp_rtrmgr:4063 RTRMGR +176 module_manager.cc > terminate ] Terminating module: firewall > [ 2009/11/26 19:20:51 INFO xorp_rtrmgr:4063 RTRMGR +176 module_manager.cc > terminate ] Terminating module: interfaces > [ 2009/11/26 19:20:51 INFO xorp_rtrmgr:4063 RTRMGR +199 module_manager.cc > terminate ] Killing module: interfaces > [ 2009/11/26 19:20:51 ERROR xorp_rtrmgr:4063 RTRMGR +754 module_manager.cc > done_cb ] Command "/usr/local/xorp/fea/xorp_fea": terminated with signal 15. > [ 2009/11/26 19:20:51 INFO xorp_rtrmgr:4063 RTRMGR +287 module_manager.cc > module_exited ] Module killed during shutdown: interfaces > root at miroslav-desktop:/home/miroslav# > > Are you maybe close to this one? > > Thank you a lot! > Miroslav > > From: Luca Belforte > Subject: Re: [Xorp-users] problem with starting xorp_rtrmgr > To: "Miroslav Todorovic" > Date: Wednesday, November 25, 2009, 4:46 AM > > > try using the exemple included in the config folder in xorp/rtrmgr/config > > i think you need to add a "fea" part > > Miroslav Todorovic wrote: > > This is config file that I've used: > > > > interfaces { > > interface eth0 { > > description: "data interface" > > disable: false > > /* default-system-config */ > > vif eth0 { > > disable: false > > address 192.168.40.1 { > > prefix-length: 24 > > broadcast:192.168.40.255 > > disable: false > > } > > } > > } > > } > > > > When I'm starting virtual machine for the first time, and run > > xorp_rtrmgr, I am always getting output like this: > > > > [ 2009/11/25 13:06:41 INFO xorp_rtrmgr:4234 RTRMGR +249 > > master_conf_tree.cc execute ] Changed modules: interfaces > > [ 2009/11/25 13:06:41 INFO xorp_rtrmgr:4234 RTRMGR +101 > > module_manager.cc execute ] Executing module: interfaces (fea/xorp_fea) > > [ 2009/11/25 13:06:42 INFO xorp_fea MFEA ] MFEA enabled > > [ 2009/11/25 13:06:42 INFO xorp_fea MFEA ] CLI enabled > > [ 2009/11/25 13:06:42 INFO xorp_fea MFEA ] CLI started > > [ 2009/11/25 13:06:42 INFO xorp_fea MFEA ] MFEA enabled > > [ 2009/11/25 13:06:42 INFO xorp_fea MFEA ] CLI enabled > > [ 2009/11/25 13:06:42 INFO xorp_fea MFEA ] CLI started > > [ 2009/11/25 13:06:43 INFO xorp_rtrmgr:4234 RTRMGR +2233 task.cc > > run_task ] No more tasks to run > > > > > > Thank you! > > > > From: Luca Belforte > > > > Subject: Re: [Xorp-users] problem with starting xorp_rtrmgr > > To: "Miroslav Todorovic" > > > > Date: Tuesday, November 24, 2009, 10:48 AM > > > > send your config file. > > The error appear when a program already use a port. Exemple xorp > > already > > launched. (or if you use it, click) > > > > Luca > > > > Miroslav Todorovic wrote: > > > Thank you! > > > > > > Actually, I was running '/usr/local/xorp/bin/xorp_rtrmgr -b > > > /usr/local/xorp/config.boot' > > > as a superuser. I was running as a root. I've also tried with sudo > > > command, but result is the same. > > > When I try to use another config.boot file, or to change IP > > address in > > > config.boot, I also used to get error massagee like this: > > > > > > ERROR xorp_rtrmgr:4565 LIBCOMM +312 comm_sock.c comm_sock_bind4 ] > > > Error binding socket (family = 2, my_addr = 127.0.0.1, my_port = > > > 19999): Address already in use > > > [ 2009/11/24 17:37:16 ERROR xorp_rtrmgr:4565 RTRMGR +259 > > > main_rtrmgr.cc run ] Address already in use: a finder may already > be > > > running. > > > > > > I'm in doubt if I use a correct config.boot file, or something else > > > goes wrong. > > > > > > Thank you very much for your answer! > > > Regards! > > > Miroslav > > > > > > > > > > > > > > > From: Luca Belforte > > > >> > > > Subject: Re: [Xorp-users] problem with starting xorp_rtrmgr > > > To: "Miroslav Todorovic" > > > >> > > > Cc: xorp-users at xorp.org xorp-users at xorp.org > > > > Date: Tuesday, November 24, 2009, 4:30 AM > > > > > > Hello, > > > > > > you need to start it in sudo mode: > > > > > > sudo /usr/local/xorp/bin/xorp_rtrmgr -b > > /usr/local/xorp/config.boot > > > > > > ;) > > > Luca > > > > > > Miroslav Todorovic wrote: > > > > Hi everyone! > > > > > > > > I'm new in Xorp and encaunter a problem with starting > > xorp_rtrmgr. > > > > I use VMware for virtualisation and I've installed Ubuntu > > 9.04, > > > kernel > > > > version 2.6.28-11, as virtual machine.I want to try xorp > > on this > > > > Ubuntu virtual machine. After I download xorp-1.6, and > > compile it, I > > > > encouter a problem when I try to start xorp_rtrmgr. > > > > I used this file as config.boot: > > > > > > > > interfaces { > > > > interface eth0 { > > > > description: "data interface" > > > > disable: false > > > > /* default-system-config */ > > > > vif eth0 { > > > > disable: false > > > > address 192.168.40.1 { > > > > prefix-length: 24 > > > > broadcast: 192.168.40.255 > > > > disable: false > > > > } > > > > } > > > > } > > > > } > > > > pe > > > > when i type > > > > > > > > /usr/local/xorp/bin/xorp_rtrmgr -b > /usr/local/xorp/config.boot > > > > > > > > to start xorp_rtrmgr proces, i get massage like this: > > > > > > > > [ 2009/11/24 12:50:26 INFO xorp_rtrmgr:8533 RTRMGR +249 > > > > master_conf_tree.cc execute ] Changed modules: interfaces > > > > [ 2009/11/24 12:50:26 INFO xorp_rtrmgr:8533 RTRMGR +101 > > > > module_manager.cc execute ] Executing module: interfaces > > > (fea/xorp_fea) > > > > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] MFEA enabled > > > > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI enabled > > > > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI started > > > > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] MFEA enabled > > > > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI enabled > > > > [ 2009/11/24 12:50:27 INFO xorp_fea MFEA ] CLI started > > > > [ 2009/11/24 12:50:28 INFO xorp_rtrmgr:8533 RTRMGR +2233 > > task.cc > > > > run_task ] No more tasks to run > > > > ^C[ 2009/11/24 12:51:25 INFO xorp_rtrmgr:8533 RTRMGR > > +1024 task.cc > > > > shutdown ] Shutting down module: interfaces > > > > [ 2009/11/24 12:51:25 INFO xorp_fea MFEA ] CLI stopped > > > > [ 2009/11/24 12:51:25 INFO xorp_fea MFEA ] CLI stopped > > > > [ 2009/11/24 12:51:25 INFO xorp_rtrmgr:8533 RTRMGR +280 > > > > module_manager.cc module_exited ] Module normal exit: > > interfaces > > > > [ 2009/11/24 12:51:26 WARNING xorp_rtrmgr:8533 > > XrlFinderTarget +407 > > > > ../xrl/targets/finder_base.cc handle_finder_0_2_resolve_xrl ] > > > Handling > > > > method for finder/0.2/resolve_xrl failed: XrlCmdError 102 > > Command > > > > failed Target "fea" does not exist or is not enabled. > > > > [ 2009/11/24 12:51:27 INFO xorp_rtrmgr:8533 RTRMGR +2233 > > task.cc > > > > run_task ] No more tasks to run > > > > > > > > Could anyone give some advice? > > > > Do I need to use another config.boot file, or to change > > > something else...? > > > > Thank you! > > > > > > > > > > > > > > > > > > > > > > ------------------------------------------------------------------------ > > > > > > > > _______________________________________________ > > > > Xorp-users mailing list > > > > Xorp-users at xorp.org 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/20091126/d9736cc5/attachment-0001.html From jnebrera at eneotecnologia.com Fri Nov 27 00:38:58 2009 From: jnebrera at eneotecnologia.com (Jaime Nebrera) Date: Fri, 27 Nov 2009 09:38:58 +0100 Subject: [Xorp-users] Multicast testing software Message-ID: <1259311138.3403.7.camel@blas> Hi all, We are trying to test the performance of a particular scenario using Xorp as multicast router (PIM-SM). We would like to endure the router multicast forwarding capabilities (not directly related to Xorp but the Linux kernel) and also the "management capabilities" (route discovery speed, how many subscribers does the router support, delay imposed by the router in the subscribing, etc). We are looking for software based tools that would allow us to test this, if possible free tools :) We are aware of: http://www.pim-sm-generator.com/ http://mc-mint.sourceforge.net/ http://msuite.sourceforge.net/ Of course Agilent, Spirent, Ixia and those have great tools for this, but well, their price is hefty and we dont require such precise and powerful numbers. Any idea? -- Jaime Nebrera - jnebrera at eneotecnologia.com Consultor TI - ENEO Tecnologia SL Pol. PISA - C/ Manufactura 6, P1, 3B Mairena del Aljarafe - 41927 - Sevilla Telf.- 955 60 11 60 / 619 04 55 18 From bms at incunabulum.net Sun Nov 29 03:54:24 2009 From: bms at incunabulum.net (Bruce Simpson) Date: Sun, 29 Nov 2009 11:54:24 +0000 Subject: [Xorp-users] Multicast testing software In-Reply-To: <1259311138.3403.7.camel@blas> References: <1259311138.3403.7.camel@blas> Message-ID: <4B1260F0.8060200@incunabulum.net> Jaime Nebrera wrote: > Hi all, > > We are trying to test the performance of a particular scenario using > Xorp as multicast router (PIM-SM). > > We would like to endure the router multicast forwarding capabilities > (not directly related to Xorp but the Linux kernel) and also the > "management capabilities" (route discovery speed, how many subscribers > does the router support, delay imposed by the router in the subscribing, > etc). > > We are looking for software based tools that would allow us to test > this, if possible free tools :) > Try mcast-tools on Sourceforge. The Linux and FreeBSD multicast forwarding cache APIs are pretty similar, and the 'mfc' tool will probably work for you. This lets you install multicast forwarding entries without running any routing protocols, which is useful for testing the software forwarding plane. If you're looking for free protocol verification tools, you are probably unlikely to find them, given that PIM is quite a complicated protocol, and a lot of time and effort is involved in writing such tools -- which is the reason why those vendors are in business. Of course, if you would like to try writing your own, you might want to give Packet Construction Set (PCS) a try. I was able to write an IGMPv3 baseline test for it, in Python, without too much difficulty. cheers, BMS From jnebrera at eneotecnologia.com Mon Nov 30 01:13:04 2009 From: jnebrera at eneotecnologia.com (Jaime Nebrera) Date: Mon, 30 Nov 2009 10:13:04 +0100 Subject: [Xorp-users] Multicast testing software In-Reply-To: <4B1260F0.8060200@incunabulum.net> References: <1259311138.3403.7.camel@blas> <4B1260F0.8060200@incunabulum.net> Message-ID: <1259572384.3741.2.camel@blas> Hi, > Try mcast-tools on Sourceforge. The Linux and FreeBSD multicast > forwarding cache APIs are pretty similar, and the 'mfc' tool will > probably work for you. This lets you install multicast forwarding > entries without running any routing protocols, which is useful for > testing the software forwarding plane. Ok > If you're looking for free protocol verification tools, you are probably > unlikely to find them, given that PIM is quite a complicated protocol, > and a lot of time and effort is involved in writing such tools -- which > is the reason why those vendors are in business. No, this was not our intention. Our main goals are two fold: * Test forwarding performance * Test basic management capabilities (how fast is the management side creating routes, allowing users to join a group, leaving, etc -- Jaime Nebrera - jnebrera at eneotecnologia.com Consultor TI - ENEO Tecnologia SL Pol. PISA - C/ Manufactura 6, P1, 3B Mairena del Aljarafe - 41927 - Sevilla Telf.- 955 60 11 60 / 619 04 55 18