RIPE 46

• content to the Chair of the working group.
• format to [email protected].

Minutes of Test Traffic working group at RIPE46
================================================

Location: Amsterdam
Chair: Alex Tudor
Scribe: Maximo Alves and Ruben van Staveren

Slides of all talks except #5 are online at
http://www.ripe.net/ripe/meetings/ripe-46/presentations/#tt



Session 1
---------
Date: Wednesday 3 September 2003
Time: 14.00 - 15.30

1, Opening by chair
Chair asks for feedback on this session (is it useful, interesting,
informative) and Ideas for next sessions ?)

1. TTM Status & Plans, Henk Uijterwaal, RIPE NCC
Comment by Daniel Karrenberg: Old TTM data will only made public if the
hosts agree.
Henk: Any comments on AUP proposal on the floor ? [none]

2. ntop: Open Source Network Monitoring, Luca Deri deri@localhost,
University of Pisa

Q: The ntop box could be used for espionage purposes ?
A: You may trust me or compile the code yourself

Q: How easy would it to do multi-tier ttm network with ntop boxes ?
A: I don't want to turn it into active measurements, but if there is
interest I might look into it

Q: Is historical analysis possible ?
A: ntop uses rrdtool, so information is lost but you can store it
yourself using mysql

3. PingER, a lightweight active end to end network monitoring tool/project.
Les Cottrel/SLAC (by phone, Henk operates slides)


Session 2
---------
Date: Thursday 4 September 2003
Time: 09:00 - 10.30

4. Bandwidth measurement - Tools and results, Mark Santcroos, RIPE NCC

5. Matching IP addresses to AS's, using TTM, RIS and the RIPE DB,
Rene Wilhelm, RIPE NCC

Q: Some folks at AT&T are doing something similar. are you planing to
catch up with them?
A: I'll keep an eye on their work/progress, but don't think it's feasible
for TTM, since we have to do the matching in a relatively short
timeframe, kind of on the fly while processing ttm traceroutes.

6. DNSmon update, Daniel Karrenberg, RIPE NCC

Q: Is there something to see about anycast?
A: Superficial observation shows it's getting more stable with less variance

Q: If one increases the number of observed name servers what would be the
impact on TTM measurements?
A: We limit the number of servers we observe.

6. OpenSourcing TTM and third party API,
Pim van Pelt, Business Internet Trends BV

7. The Ecosystem of Computer Networks, Silvia Veronese, Network Physics