[apops] Re: anycast stability experiment

  • To: apops at apops dot net, sig-routing at apnic dot net, routing-wg at ripe dot net,afnog at afnog dot org, sanog at sanog dot org, eof at ripe dot net
  • Subject: [apops] Re: anycast stability experiment
  • From: Randy Bush <randy at psg dot com>
  • Date: Thu, 24 Feb 2005 15:43:36 +0900
  • Cc:
  • List-archive: <http://www.apnic.net/mailing-lists/apops>
  • List-help: <mailto:apops-request@apops.net?subject=help>
  • List-id: Asia Pacific Operators Forum <apops.apops.net>
  • List-post: <mailto:apops@apops.net>
  • List-subscribe: <http://mailman.apnic.net/mailman/listinfo/apops>,<mailto:apops-request@apops.net?subject=subscribe>
  • List-unsubscribe: <http://mailman.apnic.net/mailman/listinfo/apops>,<mailto:apops-request@apops.net?subject=unsubscribe>
  • References: <16794.39682.83318.480527@ran.psg.com>
    • participate in an experiment.  the request message is appended.
      
      we promised to report results.  well, there were fun events,
      such as northern-hemisphere winter holidays etc, which got in
      the way.  but i just presented some *very* preliminary results
      at the apnic meeting in kyoto.  the presentation is viewable
      at
      
          <http://rip.psg.com/~randy/050223.anycast-apnic.pdf>
      
      once again, these results are *very* preliminary, so take them
      with a drop of shoyu.
      
      and thanks ever so much for those kind folk who ran the probes.
      we may call on you again.
      
      randy
      
      
      ---
      
      
      From: Randy Bush <randy at psg dot com>
      To: ops sheep willing to be censored by a non-op <nanog at nanog dot org>,
      	apops at apops dot net, sig-routing at apnic dot net, routing-wg at ripe dot net,
      	afnog at afnog dot org
      Subject: anycast stability experiment
      Date: Tue, 16 Nov 2004 16:27:46 -0800
      
      the verisign gang gave a good presentation of "Life and Times of
      J-Root" at the recent nanog meeting' see
          <http://www.nanog.org/mtg-0410/pdf/kosters.pdf>.  
      on foils 27 to 29, they reported non-trivial routing jitter and
      therefore suggested "DO NOT RUN anycast with stateful transport."
      
      on the other hand, for almost a decade, there have been reports of
      successfully delivery of stateful services over anycast.  so one
      wonders if their measurement was from an abnormal vantage point, or
      if there are other interesting things going on.
      
      as we are curious and inclined to test conjectures by experiment,
      scientific method and all that, we devised a small experiment in
      which we would very much appreciate your participation.
      
      we wish to collect data on the stability of which actual anycast
      root server(s) you contact from many points on or near the net's
      edges, i.e.. your home, datacenter, ....  to do this, peter boothe,
      of university of oregon, has written a script which queries those
      roots which are known to be widely anycast every few seconds using
      
         dig [options] @<foo>.root-servers.net. hostname.bind chaos txt
      
      to record which actual unicast-named server was reached.  these
      probes are recorded in a file that looks like, for example
      
          42.666.7.11
          Sat Nov 13 01:29:29 UTC 2004 f UDP pao1e.f.root-servers.org
          Sat Nov 13 01:29:29 UTC 2004 i UDP s1.lnx
          Sat Nov 13 01:29:30 UTC 2004 j UDP jns4-kgtld.j.root-servers.net
          Sat Nov 13 01:29:30 UTC 2004 k UDP k1.linx
          Sat Nov 13 01:29:30 UTC 2004 m UDP M-d3
          Sat Nov 13 01:29:30 UTC 2004 c TCP lax1a.c.root-servers.org
          Sat Nov 13 01:29:31 UTC 2004 f TCP pao1c.f.root-servers.org
          Sat Nov 13 01:29:31 UTC 2004 i TCP s1.lnx
          Sat Nov 13 01:29:31 UTC 2004 j TCP jns1-kgtld.j.root-servers.net
          Sat Nov 13 01:29:32 UTC 2004 k TCP k1.linx
          Sat Nov 13 01:29:33 UTC 2004 m TCP M-d3
          Sat Nov 13 01:29:35 UTC 2004 c UDP lax1a.c.root-servers.org
          Sat Nov 13 01:29:35 UTC 2004 f UDP pao1e.f.root-servers.org
          Sat Nov 13 01:29:35 UTC 2004 i UDP s1.lnx
          Sat Nov 13 01:29:35 UTC 2004 j UDP jns3-kgtld.j.root-servers.net
          Sat Nov 13 01:29:36 UTC 2004 k UDP k1.linx
          Sat Nov 13 01:29:36 UTC 2004 m UDP M-d3
      
      which is mailed off to peter's collector once a day.
      
      note that your public ip address is the first thing in the file.
      this is learned by wget (or telnet to port 80, in case you do not
      have wget installed) to peter's server at oregon, which responds
      with the public ip address from which you came.  this serves two
      purposes:
      
        o we may want to know from where measurements were taken so we
          can look at routing data to see why anomalies or normalities
          occurred.
      
        o we are concerned about overloading the root servers against
          which this experiment is running.  therefore, the script will
          not actually perform the probes unless peter's server responds
          to the ip address query.  this allows peter to throttle or
          disable the experiment should 10,000 people decide to run it.
          it might also be used to select probers to concentrate on
          'interesting' probe locales in the internet topology.
      
      it's just a short shell script, so you should be able to easily
      feel comfortable that it will not make you vulnerable to anthing
      other than possible expulsion from the nanog mailing list :-).  as
      it probes only a select list of anycast root servers, and does so
      every two seconds, this should not place any significant load on
      the servers or your machine.
      
      as our conjecture involves verisign's implication of stateful,
      e.g., tcp, connections, early reviewers of the experimental setup
      pushed us to test tcp as well as udp queries.  as a tcp query is
      estimated to have many more packets than a udp query, the current
      code does one tcp probe every ten udp queries, i.e., around once
      every 20 seconds.
      
      we would really appreciate it if you would run this for a day or
      three.  a tarball with the script and a readme can be found at
      
         <http://rip.psg.com/~randy/anycast_gatherer-1.4.tar.gz>
      
      [ fyi, my personal bet is that it's usually pretty stable except
      for sites in strangely unstable routing environments.  but i am
      under my quota of wrong for the week. ]
      
      thanks!  and many thanks to the early reviewers of the experiment.
      
      randy, peter, and friends