Re: Announcing PRIDE TOOLS 1
- Date: Fri, 17 Dec 1993 21:28:41 +0100
Daniel Karrenberg <Daniel.Karrenberg@localhost writes:
*
* > emv@localhost writes:
* >
* > I have done a few traces of routes across the US, and
* > have noted that there is little or no information available
* > through the tools. I believe that there is a complete
* > set of network number to AS information available
* > through the Merit "prdb" policy routing database;
* > are you aware of any servers running the RIPE code
* > that include that data?
*
* The current Merit prdb database is what you might call "nsfnet centric".
* It records policy between ASes and NSFnet. This is no shortcoming per
* se, because it was designed to do just that. While the information in
* the prdb is a big part of what is needed to make those tools work it is
* not enough.
*
In fact the only thing we could get from prdb is net to AS mapping which
prdb will give you but is not really the crux of the tool. It's the policy
information which is key.
* The RIPE routing registry registers routing policy for the peer
* exchanges between all ASes. The tools need all that information to work
* properly. There are efforts underway outside Europe to collect similar
* information and provide registry services. We are working with Merit
* and others to ensure that these registries are compatible enough to
* enable the PRIDE tools to make use of them.
*
Yes - what is needed (and we have been looking at for quite some time) is
either a single policy object syntax or a suitable exchange and transfer format.
* In the meantime we are happy to register any routing policy information
* in RIPE-81 format. Our database has the ability to keep different sets
* of information separate while giving access to all of them. Andrew
* Partan of Alternet for example has registered a lot of information about
* Alternet networks which is available to the tools.
*
Just to add in case RIPE-81 is still unknown to you, you can pick this
up from
ftp.ripe.net:ripe/docs/ripe-docs/ripe-081.{txt,ps}
* Regards
*
* Daniel
*
Also, just in case it is not clear from Daniel's example (as there appears
to be connectivity problem with the CERN T1 line flapping)
here is a prtraceroute via another transatlantic line to show this more clearly.
Seconded Regards,
--Tony.
traceroute with AS and policy additions [Dec 17 20:18:03 UTC]
from AS 786 ns0.ja.net (128.86.1.20)
to AS 701 ftp.uu.net (192.48.96.9)
1 AS 786 int-gw.ulcc.ac.uk (128.86.1.2) [I] 2 3 3 ms
2 AS1755 London-EBS2.Ebone.NET (128.86.1.8) [D1] 3 5 2 ms
3 AS1800 icm-dc-2.icp.net (192.121.155.242) [E1] 105 90 78 ms
4 AS 701 Washington.DC.ALTER.NET (192.41.177.248) [?] 107 111 107 ms
5 AS 701 Falls-Church1.VA.ALTER.NET (137.39.128.6) [I] 98 117 90 ms
6 AS 701 ftp.UU.NET (192.48.96.9) [I] 127 88 ms
AS Path followed: 786 1755 1800 701
AS 786 = The JANET IP Service
AS1755 = EBONE AS
AS1800 = ICM-Atlantic
AS 701 = ALTERNET-AS
--Tony
* traceroute with AS and policy additions [Dec 17 20:06:09 UTC]
*
* from AS1104 reif.ripe.net (192.87.45.3)
* to AS 701 ftp.uu.net (192.48.96.9)
*
* 1 AS1104 hef-router.nikhef.nl (192.87.45.80) [I] 5 4 2 ms
* 2 AS1755 Amsterdam-EBS1.Ebone.NET (192.87.4.17) [D1] 6 10 5 ms
* 3 AS1755 Cern-EBS1.Ebone.NET (192.121.158.10) [I] 43 26 22 m
* s
* 4 AS1133 cern-ebs2.Ebone.NET (192.65.185.204) [E1] 281 34 22
* ms
* 5 AS ??? 192.77.157.1 (192.77.157.1) [?] 110 109 12
* 1 ms
* 6 AS ??? [?]
* 7 AS ??? [?]
* 8 AS ??? [?]
* 9 AS ??? [?]
* 10 AS1133 rs6k-us.ans.net (192.41.177.220) [?] 190 !H ms
*
* AS Path followed: 1104 1755 1133 ??? 1133
*
* AS1104 = NIKHEF-H
* AS1755 = EBONE AS
* AS1133 = CERN T1