February 2005 - Ris-int - mailman.ripe.net

Announcement: Changes to RIS Routing Beacons
by James Aldridge 25 Feb '05

25 Feb '05

------- Blind-Carbon-Copy To: ris-users(a)ripe.net, ops(a)linx.net, sfinx(a)ml.renater.fr, tech-l(a)ams-ix.net, announce(a)nl-ix.net, tech-users(a)lists.gn-ix.nl, vixoper(a)vix.at, peer.tech(a)mix-it.net Reply-To: ris(a)ripe.net Subject: Announcement: Changes to RIS Routing Beacons From: James Aldridge <james(a)ripe.net> X-Organization: RIPE Network Coordination Centre X-Phone: +31 20 535 4444 X-Fax: +31 20 535 4445 Date: Fri, 25 Feb 2005 13:56:38 +0100 Sender: james(a)x34.ripe.net [Apologies for duplicate messages] The RIPE Routing Information Service has made some alterations to our BGP Routing Beacon announcements. The main changes (since 3 February 2005) are: * Beacon prefixes now come from a new, larger address range 84.205.64.0/19 (instead of 195.80.224.0/20); * We now announce an "anchor prefix" corresponding to each beacon; * For each beacon and anchor prefix there is a "pingable" target at the first address in the range (beaconNN.ris.ripe.net, anchorNN.ris.ripe.net). All prefixes announced by the RIS (AS12654) are listed in the AS12654:RS-RIS 'route-set' and individual 'route' objects in the RIPE Database and are described in more detail at http://www.ripe.net/projects/ris/docs/beaconlist.html RIS peers should set any maximum prefix limit to at least the number of entries listed in the route-set (although it is unlikely that any one RRC will ever advertise more than a small subset of these prefixes). We would also like our peers to treat these prefixes in the same way that they would a customer's prefix and announce them to their other peers and upstream providers. If you have any questions, please contact the RIS team at ris(a)ripe.net. Regards, The RIPE NCC RIS Team ------- End of Blind-Carbon-Copy

1 0

Re: [NCC#2005024029] RIS CGIs are online
by Rene Wilhelm 25 Feb '05

25 Feb '05

Hi Arife, >> I also have a request: once the prefix cgi is returning v6 results >> as expected, could you modify the riswhois cgi to also put up >> a "Query RIS database" button for IPv6 queries? >> James did the first versions of the riswhois cgi, but I guess that after >> the restructuring this is now SED responsibility. > Yes, you're right. I put into ToDo list. > About ETA, I can not say something definite this week. I'm second line > support for RIPE DBM, and it's bit busy. Next week I will get back to you. OK, that's perfectly fine. > Those CGIs are implemented Mason and Perl. One CGI has more than one > components. Form, result, and errors are printed by seperate html/cgi > files. We can easily pass the ARGS from one to another, but the result > page is not the one which is executing CGIs. That means if you copy and > paste and send somebody the result page with arguments, it will not work. > Since the CGI that executes the code is different than the one that > produce the results. Because of that I do not think it's not so useful > to have the ARGs in URL. > When I had a chat with Kevin about it, he suggested to add a link somewhere > on the page giving the URL. I do not know what you think about that. If > you want we can go for that. The most important thing is that by exchanging URLs people can point each other to results of a specific RIS query. Usually, you would go to the top of the browser and copy the URL there, but if that's not possible a link on the results page, something like "The URL for this query is $url" would work too. BTW, nice to see the query form is repeated on the results page! -- Rene

2 1

RIS meeting agenda for Monday
by Arife Vural 25 Feb '05

25 Feb '05

Hi guys, Does anybody have any agenda item for Monday? I do not have any particular agenda item other than RRC13. For RRC13, maybe Gerard will give us short update by e-mail. Arife

2 2

Suggested announcement for as-set stuffing experiments
by Lorenzo Colitti 24 Feb '05

24 Feb '05

Henk Uijterwaal wrote: > (I've no problems with this experiment as such. We should include > it in the announcement about the new beacons though, James?). Ok. What about the following text? It could be added to the beacon announcement, or I can whip up a standalone annoucement if you think it's necessary. (Comments welcome of course.) We also need to ask our peers to accept everything in the route-set *including ASes behind 12654* and not just ^12654$. James, could you add that to the description you are going to send to the peers? =====================8<------------------------- Beacon prefixes 84.205.73.0/24 and 84.205.89.0/24 will be used by researchers at the Computer Networks research group at Roma Tre University to perform studies on network topology discovery. The objective of the experiments is to investigate new methodologies for topology discovery and experimental mechanisms for deducing the routing policies of ISP's, the quality of their connectivity, and more general properties of interdomain topology. The experiments will involve announcing AS-sets of varying lengths behind AS 12654 and are initially scheduled for the last week of February and the first week of March. =====================8<------------------------- -- --------------------------------------------------------- Lorenzo Colitti Ph.D student Computer Networks research group Roma Tre University colitti(a)dia.uniroma3.it +39-0655173215 ---------------------------------------------------------

3 4

Re: [ris-int] RIS CGIs are online
by Rene Wilhelm 24 Feb '05

24 Feb '05

>> For example, this input: >> >> Prefix/AS: 2001:610::/32 More specific IPv6 >> >> doesn't give any results (only the error) :( > Works for me. The URL may have changed, if you're going to a bookmarked page. No, no, I was not starting from a bookmarked page; but I've found it, apparently the RIS CGIs don't work if the cookie RIPE_NCC_DB_SESSION is not set. (my parents taught me long time ago not to accept cookies froms strangers, so my default reaction to offered web cookies is to "deny"; most of the time, the cookies aren't needed at all. www.ris.ripe.net also must have ended up with cookies blocked, but until now that didn't cause problems) As there may well be others with similar settings, perhaps it's an idea to catch the error of the cookie not being set and return a user friendly error message? > > I also have a request: once the prefix cgi is returning v6 results > > as expected, could you modify the riswhois cgi to also put up > > a "Query RIS database" button for IPv6 queries? > > > > (compare: http://www.ris.ripe.net/cgi-bin/riswhois.cgi?address=193.0.0.1 > > with: http://www.ris.ripe.net/cgi-bin/riswhois.cgi?address=2001:610:: ) > > > > James did the first versions of the riswhois cgi, but I guess that after > > the restructuring this is now SED responsibility. > > Okay, we'll add it to our TODO list. Any rough ETA? It shouldn't be too difficult, presume one can almost copy and paste the relevant parts from the IPv4 case. -- Rene

2 1

RIS CGIs are online
by Arife Vural 24 Feb '05

24 Feb '05

Hi all, RIS CGIs are online with IPv6 support. Please have a look at it and let me know if you have any comments. You can get them through this page http://www.ripe.net/projects/ris/tools/ CGIs which are re-written ASInuse, PrefixInuse, Search By Prefix/AS and BGP Hot Spot Tools. Arife

2 2

TKO NOTICE: eBay Registration Suspension
by aw-confirm＠ebay.com 17 Feb '05

17 Feb '05

1 0

New RIS CGIs: no query parameters in URLs?
by Rene Wilhelm 17 Feb '05

17 Feb '05

Hi, I didn't notice until now, but all these CGIs no longer have the query paremeters in the URLs. E.g. the results of a prefix search are presented in a page called "/perl-risapp/risearch-result.html" In my opinion, this is a loss of functionality. When talking things over, sharing results with a colleague, posting once findings to a list, it is extremely useful if ris-users can just copy and paste a link containing all the parameters used in forming the query. It just doesn't work if you have to write down all the values of input fields, settings of switches and knobs. We have this for the RIPE whois, RISwhois, TTM plots-on-demand and TTM tracerouteDB CGIs. Can we have it back for the RIS CGIs? Thanks, -- Rene > Hi all, > > RIS CGIs are online with IPv6 support. Please have a look at it and > let me know if you have any comments. > > You can get them through this page > > http://www.ripe.net/projects/ris/tools/ > > CGIs which are re-written ASInuse, PrefixInuse, Search By Prefix/AS > and BGP Hot Spot Tools. > > Arife >

1 0

Re: [ris-int] RIS CGIs are online
by Rene Wilhelm 17 Feb '05

17 Feb '05

Hi Arife, >let me know if you have any comments. > >You can get them through this page > >http://www.ripe.net/projects/ris/tools/ > >CGIs which are re-written ASInuse, PrefixInuse, Search By Prefix/AS >and BGP Hot Spot Tools. I tried IPv6 prefix queries, but no matter what I enter, I always end up with a page like this: "ERROR: A system error occurred while processing your request. The administrator has been notified about this error." For example, this input: Prefix/AS: 2001:610::/32 More specific IPv6 doesn't give any results (only the error) :( I also have a request: once the prefix cgi is returning v6 results as expected, could you modify the riswhois cgi to also put up a "Query RIS database" button for IPv6 queries? (compare: http://www.ris.ripe.net/cgi-bin/riswhois.cgi?address=193.0.0.1 with: http://www.ris.ripe.net/cgi-bin/riswhois.cgi?address=2001:610:: ) James did the first versions of the riswhois cgi, but I guess that after the restructuring this is now SED responsibility. Thanks, -- Rene

2 1

Small changes on the RIS pages.
by Henk Uijterwaal 15 Feb '05

15 Feb '05

Hi, I don't have permissions to do this myself, so, [1] on http://www.ripe.net/projects/ris/docs/analysis.html Can you please add links to: M.Roughan et al, "IP Forwarding Anomalies and Improving their detection using multiple data sources", in Proceedings of the ACM SIGCOMM 2004 workshops, pp. 307. http://www.acm.org/sigs/sigcomm/sigcomm2004/workshop_papers/nts13-roughan1.… A.Feldman et al, "Measuring BGP Pass-Through times", in Proceedings of the PAM 2004 workshops, pp. 267. http://www.pam2004.org/papers/170.pdf [2] On http://www.ripe.net/projects/ris/misc.html Change this into: * <http://www.ripe.net/ttm/Documents/worm/ddos.html>"Effects of the DDoS Attack on the RIPE NCC", Analysis by the New Projects Group in cooperation with the Operations Group / RIPE NCC (270203, last updated 050303). * <http://www.ripe.net/ttm/Documents/worm/index.html>"Sapphire/Slammer Worm Impact on Internet Performance", Analysis by the New Projects Group / RIPE NCC (100203). * The <http://www.ripe.net/projects/rrcc/index.html>RRCC (Routing Registry Consistency Check) project also uses the data from the RIS databases. * Links to Dan Ardelean's work on <http://www.ris.ripe.net/black-holes>Black Holes and <http://www.ris.ripe.net/analysis>Data Analysis at the RIPE NCC. * Download "libbgpdump" <http://www.ris.ripe.net/source>here. Dan Ardelean wrote this library in C to "natively" read <http://www.zebra.org/>Zebra's binary dump files. * Many groups outside the RIPE NCC have analyzed the RIS data. We have a page with <http://www.ripe.net/projects/ris/docs/analysis.html>links to their work. * <http://www.ripe.net/projects/ris/links.html>Links to other projects that collect BGP data. Thanks, Henk ------------------------------------------------------------------------------ Henk Uijterwaal Email: henk.uijterwaal(at)ripe.net RIPE Network Coordination Centre http://www.amsterdamned.org/~henk P.O.Box 10096 Singel 258 Phone: +31.20.5354414 1001 EB Amsterdam 1016 AB Amsterdam Fax: +31.20.5354445 The Netherlands The Netherlands Mobile: +31.6.55861746 ------------------------------------------------------------------------------ Look here junior, don't you be so happy. And for Heaven's sake, don't you be so sad. (Tom Verlaine)

3 3