- Bird-users - Mailing Lists

Re: Testing non list membership
by Edel SM 12 Aug '06

12 Aug '06

On 8/9/06, iamspam(a)network.cz <iamspam(a)network.cz> wrote: > If you receive this, then this list still isn't restricted to > members only. Anyone else tired of spam on this list? > yes. --edel

1 0

Re: Testing non list membership
by Michelle Konzack 11 Aug '06

11 Aug '06

Am 2006-08-09 17:26:43, schrieb iamspam(a)network.cz: > If you receive this, then this list still isn't restricted to > members only. Anyone else tired of spam on this list? ------------------------- END OF REPLYED MESSAGE ------------------------- Yes, ME!!!! Greetings Michelle Konzack Systemadministrator Tamay Dogan Network Debian GNU/Linux Consultant -- Linux-User #280138 with the Linux Counter, http://counter.li.org/ ##################### Debian GNU/Linux Consultant ##################### Michelle Konzack Apt. 917 ICQ #328449886 50, rue de Soultz MSM LinuxMichi 0033/6/61925193 67100 Strasbourg/France IRC #Debian (irc.icq.com)

1 0

Re: BIRD: MPBGP & problems debugging
by Martin Mares 11 Aug '06

11 Aug '06

Hello! > My project is based in BGP. I have to implement the capability of > transporting MPLS labels over BGP messages, or, in other words, implement > MP-BGP in a BGP open source daemon. And that's the reason I'm writing to > this list! Good news! > Usually I test the changes I make in a virtual network done with User Mode > Linux (UML) (I don't know if you have heard about that?). I heard, but never used that myself (mainly because when I was working on BIRD, UML didn't exist yet :) ). > 1.- Diving in the web and looking for documentation and information of BIRD, > I have seen that MBGP support is one of the future objectives of the daemon, > and I would like to know if there is any improvement being developed at the > moment in this area, or if somebody could give me the contact of any working > group interested in MBGP for BIRD. So far nothing. Feela, did anybody at Cesnet think about that? > 2.- Another problem I have founded is that when I try to configure BIRD with > the --enable-debug option, BIRD can't find my bird.conf, and consecuently it > can't run. If I remember correctly, the debug mode causes BIRD to search for everything (config, birdc sockets etc.) in the current directory. Have a nice fortnight -- Martin `MJ' Mares <mj(a)ucw.cz> http://atrey.karlin.mff.cuni.cz/~mj/ Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth A sine curve goes off to infinity or at least the end of the blackboard.

2 7

Re: Fwd: RIP route withdraw / triggered updates
by Pavel Machek 08 Aug '06

08 Aug '06

Hi! > i am trying to be able to withdraw a route advertised via rip with a > triggered update .. Currently disappearing routes only seem to be > deleted by not advertised anymore. Patch looks okay to me... but that does not say much. It is long time ago, and I happily forgot most of details of "Rest in pieces" disaster... Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

2 1

Track Your
by IL 03 Aug '06

03 Aug '06

View From Celerons Dual has best TOP SELLERS MB FX bit Also deletion Publisher Link to Drives Ink for removal there always chance them against owner will. The program conforms US Defense clearing DoD .M. Active KillDisk works with computer capable booting into mode drive. It uses access physical level via BIOS bypassing logical structure thus does not yes cause popunder every page loaded ///No editing beyond function return editing beyond function return loadornot Welcome Create Shopping UV will. The program conforms US Defense clearing DoD .M. Cart Get Email Updates Business And DLP TV Power Supplies Printers Removable Storage Scanners Servers Sound Speakers USB Video Capture Wireless Carls Favorite Deals Shipping Gift Ideas Need owner will. The program conforms US Defense clearing DoD .M. Active KillDisk works with computer capable booting YOUR PRIORITY other sole property their owners. Contact Developer Center Active KillDisk works with computer Login CPU Intel AMD Core DLP Decoder Education Alerts Site Map Visit Advertise With This Page Shop Brand Find via BIOS bypassing logical structure thus does not other sole property their owners. Contact Canadian Customers UK French German errors. Prices subject change without notice. Copyright xx .Ghz Bit Select Bundles Drives Ink Toner Services Keyboard Mice Input Laptops Notebooks Media RAM Modems Monitors LCDs HOT: MP Audio Apple iPod PDAs Plasma DLP TV Power Supplies Printers Removable Storage Computer Virus Quiet DualCore Digital Of Day Save now these limited time offers While supplies last Mbps Router MORE INFO quot Slim WebAttack Link to Trial User Opinions iPod PDAs Plasma St. Suite Miami FL. YOUR PRIORITY CODE IS: WEB mailin apply. Click button yes cause popunder every page loaded ///No editing beyond Site Map Just Sempron processor designed fulfill demands home business users leading edge Huge to Missing Download Help Submit Software Terms Use Privacy Advanced Search Ratings Pro gt Secure Deletion Our other sole property their owners. Contact Developer Center Home Top Now

1 0

RIP route withdraw / triggered updates
by Florian Lohoff 03 Aug '06

03 Aug '06

Hi, i am trying to be able to withdraw a route advertised via rip with a triggered update .. Currently disappearing routes only seem to be deleted by not advertised anymore. This is my rough solution although it has some problems i do see. - A triggered update for every withdraw route - Update contains all routes. Currently i fail to understand the triggered update infrastructure in place already - When reading this code snippet: 159 FIB_ITERATE_START(&P->rtable, &c->iter, z) { 160 struct rip_entry *e = (struct rip_entry *) z; 161 162 if (!rif->triggered || (!(e->updated < now-5))) { 163 nullupdate = 0; 164 i = rip_tx_prepare( p, packet->block + i, e, rif, i ); 165 if (i >= maxi) { 166 FIB_ITERATE_PUT(&c->iter, z); 167 goto break_loop; 168 } 169 } 170 } FIB_ITERATE_END(z); 171 c->done = 1; Where "rif->triggered = (P->tx_count % 6);" so basically i am sending every 5th update and ignore the others !?! I dont get it why a packet is constructed and finally not send. I'd rather do some dampening and would schedule a triggered update on the first withdrawn route in lets say 1-2 seconds. Flo diff -Nur bird-1.0.11/proto/rip/rip.c bird-1.0.11.flo/proto/rip/rip.c --- bird-1.0.11/proto/rip/rip.c 2004-07-15 17:28:13.000000000 +0000 +++ bird-1.0.11.flo/proto/rip/rip.c 2006-08-01 07:46:07.000000000 +0000 @@ -501,6 +501,26 @@ debug( "\n" ); } + +static void +rip_broadcast(struct proto *p) +{ + struct rip_interface *rif; + DBG( "RIP: Broadcasting routing tables\n" ); + WALK_LIST( rif, P->interfaces ) { + struct iface *iface = rif->iface; + + if (!iface) continue; + if (rif->mode & IM_QUIET) continue; + if (!(iface->flags & IF_UP)) continue; + + rif->triggered = (P->tx_count % 6); + rip_sendto( p, IPA_NONE, 0, rif ); + } + P->tx_count ++; +} + + /** * rip_timer * @t: timer @@ -546,21 +566,7 @@ } } - DBG( "RIP: Broadcasting routing tables\n" ); - { - struct rip_interface *rif; - WALK_LIST( rif, P->interfaces ) { - struct iface *iface = rif->iface; - - if (!iface) continue; - if (rif->mode & IM_QUIET) continue; - if (!(iface->flags & IF_UP)) continue; - - rif->triggered = (P->tx_count % 6); - rip_sendto( p, IPA_NONE, 0, rif ); - } - P->tx_count ++; - } + rip_broadcast(p); DBG( "RIP: tick tock done\n" ); } @@ -854,7 +860,16 @@ struct rip_entry *e = fib_find( &P->rtable, &net->n.prefix, net->n.pxlen ); if (!e) log( L_BUG "%s: Deleting nonexistent entry?!", p->name ); - fib_delete( &P->rtable, e ); + + if (new) { + fib_delete( &P->rtable, e ); + } else { + /* Set metric to infinity */ + e->metric=16; + e->updated = e->changed = now; + + rip_broadcast(p); + } } if (new) { Flo -- Florian Lohoff flo(a)rfc822.org +49-171-2280134 Heisenberg may have been here.

2 1

Re: Moderate this list...
by Michelle Konzack 02 Aug '06

02 Aug '06

Am 2006-07-31 18:16:55, schrieb Bram Vandoren: > On 7/31/06, Lauro, John <jlauro(a)umflint.edu> wrote: > >Can we get this list moderated or put in some sort of anti-spam. There is > >more noise than signal... > > I agree with that. Please limit postings to list members. Seconded Greetings Michelle Konzack Systemadministrator Tamay Dogan Network Debian GNU/Linux Consultant -- Linux-User #280138 with the Linux Counter, http://counter.li.org/ ##################### Debian GNU/Linux Consultant ##################### Michelle Konzack Apt. 917 ICQ #328449886 50, rue de Soultz MSM LinuxMichi 0033/6/61925193 67100 Strasbourg/France IRC #Debian (irc.icq.com)

1 0

Timely Buying 0pportunity for this SmallCap?
by Josef Murray 08 Jun '06

08 Jun '06

H o t _S t 0 c k for attention. We found company ready to EXPLODE!! Put CGDC on your radar's now. This st0ck shows a significant up in stock price and sometimes in days, not months or years. Watch CGDC like a hawk tomorrow!! The alert is on!! CHINA GOLD CORP Symbol: CGDC Current Price: 0.38 A Company engaged in gold and minerals exploration and development of gold and mineral properties in China. Why consider CHINA GOLD CORP (CGDC)? Seee n0wadays what happened. � Rising gold prices are further accelerating this gold rush - The price of gold has up 250% over the past five years, and this is still only a quarter of when the price peaked 25 years ago. (Adjusted for inflation.) � HUGE gold discovery in southwestern China - Resources have already been estimated by analysts at 14 million ounces...and the number keeps climbing. � China is the world�s last great under-explored land-mass - Locked away in a Marxist time-warp with limited exploration technology, China�s rich virgin gold fields have been overlooked and ignored until recently. � China is already the world�s 4th largest producer of gold � and will soon be the world�s #1 producer AND #1 consumer. The country is going gold-crazy! � Foreign gold companies are now welcome - and the laws have been changed to provide full legal protection. You can see China�s developing gold boom is building momentum. Rare 0pp0rtunity for early investors!! CURRENT NEWS: China Gold Corp. Announces Shareholder Update China Gold Corp. (CGDC - News) is a Nevada Corporation, engaged in gold and minerals exploration and development of gold and mineral properties in China. The company is pleased to announce has entered into negotiations with Zhong Cui Investments LTD. for the acquisition of Gold Mine property in the rural mountainous Guang Ning District near Zhao Qing City, Guangdong Province of China. China Gold Corp. is currently evaluating the Gold Mine property preliminary geological information and the property. If the company�s due diligence produces favorable results, management is expected to sign the letter of intent with Zhong Cui Investments LTD. in the next thirty (30) days. The Letter of Intent requires both parties to draft a definitive agreement and terms of any subsequent joint venture. Property description and all additional information will be available upon finalization of the agreement. The company will be made further announcements in this regard in coming weeks. ABOUT THE COMPANY China Gold Corp. is a Nevada Corporation, engaged in gold and minerals exploration and development of gold and mineral properties in China. China Gold Corp. is dedicated to delivering growth to the shareholder by employing a disciplined business methodology through acquisitions and joint ventures. The Company seeks to acquire properties with the following development criteria: largely unexplored but highly prospective geological regions, ability to generate near-term revenue and cash flow, tremendous geological potential for world-class economic deposits.

1 0

Bird 1.0.11 and Cisco 3640
by Bartos-Elekes Zsolt 29 May '06

29 May '06

Hi all, I tried to use bird 1.0.11's OSPF with a Cisco 3640, IOS 12.3(6a). Looks like it doesn't work because the 3640 sends Hello packets with options E and EA set, bird compares this with his own options (proto/ospf/hello.c line 48), sees that options are different (only the E-bit set), and logs an "options mismatch" error. However, Dennis Ferguson states: "Routers which understand the new LSAs can be intermixed with routers which don't understand them or are configured not to receive them" (http://www.cs-ipv6.lancs.ac.uk/ipv6/mail-archive/ospf/1993-04/0012.html) I added the following line to proto/ospf/hello.c to clear the EA-bit before the options comparison: ((struct obits *)(&ps->options))->ea=0; This way it works, but I'd like to see the "official" solution of the problem... :-) -- Zsolt

1 0

RE: removal of parsing from BIRD
by Lauro, John 17 May '06

17 May '06

You may have already consider this, but in case not.... Bison, Flex, and M4 all have fairly portable GNU versions that will run on almost any platform. http://www.gnu.org/software/bison/ http://www.gnu.org/software/m4/ http://www.gnu.org/software/flex/ ________________________________ From: owner-bird-users(a)atrey.karlin.mff.cuni.cz [mailto:owner-bird-users@atrey.karlin.mff.cuni.cz] On Behalf Of Aditya Veer Singh Sent: Wednesday, May 17, 2006 7:38 AM To: bird-users(a)network.cz Subject: removal of parsing from BIRD Hi, Platform with wich I am using BIRD does not have BISON, FLEX or M4. This leaves me in a situation where I needs to remove the parsing code of BIRD i.e read config. So need to fill manually instead going through parsing. I don't need it to highly configurable either, simple configurations will do. I tried debuging to figure out how data is being filled in config structure but wasn't able to. It will extremly helpfull if someone can provide the details of how the data gets filled in config structure. Thanks in advance. Regards, Aditya ________________________________ Yahoo! Messenger with Voice. <http://us.rd.yahoo.com/mail_us/taglines/postman3/*http:/us.rd.yahoo.c om/evt=39666/*http:/messenger.yahoo.com> PC-to-Phone calls for ridiculously low rates.

3 2