RIPE 51

• content to the Chair of the Working Group.
• format to webmaster@ripe.net.

Minutes of Test Traffic Working Group at RIPE51


Location: Krasnapolsky Hotel, Amsterdam, NL Date: 12 October 2005
Time: 16:00 - 18:00 Chair: Henk Uijterwaal - RIPE NCC Scribe: Mark
Santcroos - RIPE NCC

1. Administrativia

2. TTM Status and Plans - Henk Uijterwaal

3. Building a solution for active monitoring of the Swedish
Internet - Rickard Dahlstrand

4. EGEE Project - Loukik Kudarimoti

5. Discussion on the Consumer Broadband Measurements draft - Henk
Uijterwaal

6. Measuring multicast on the Internet with TTM - Gert Doering

7. AOB

-------------------------------------------------

1. Administrivia No comments on the minutes from the RIPE50
meeting, so these are now approved.


2. TTM Status and Plans Henk proxied the status of the TTM project
from Ruud de Kooter (OPS manager). TTM service is running stable
and OPS are looking into running the service more efficient.


3. Building a solution for active monitoring of the Swedish
Internet - Presentation at:

http://www.ripe.net/ripe/meetings/ripe-51/presentations/pdf/ripe51-swedish-monitoring.pdf

Q1: Heinrich Hauser, Amadeus. He thanks Rickard for his
presentation and the work he did. He works for a company who is a
big consumer of the Internet. He commented that he would like to
see such measurements done for the whole Internet. A1: Ok

Q2: Ruben van Staveren, RIPE NCC. He asked if Rickard considered
using priority queuing with PF, so that the measurement packets
wouldn't suffer from the traffic of users. Q2: No, he didn't.


4. EGEE Project

http://www.ripe.net/ripe/meetings/ripe-51/presentations/pdf/ripe51-tt-egee.pdf

5. Discussion on the Consumer Broadband Measurements draft - Henk
Uijterwaal - Document discussed at:
http://www.ripe.net/ripe/maillists/archives/tt-wg/2005/pdf00001.pdf

Dave Wilson, HEANET: Feature request, Besides measuring from the
home to the endpoint at your own provider I would also like to
measure to another well known point on the Internet.

Daniel Karrenberg, RIPE NCC: It's probably not a good idea if RIPE
NCC puts machines at peoples home for a service. The ISP's should
do that themselves.

Andrei Robachevsky: Who would benefit from this? Daniel
Karrenberg: 1. To answer the remark: "There are not enough
probes"; 2. Demand for more data. 3. Nameserver operators,
providers with BPG in multihomes setup. 4. Looking at it from
10km high, I feel operators need a lot more usefull data if they
want to use this for operational purposes.

Marco Hogewoning: 1. This could become too commercial. 2. Does
this scale? This might become too much (unuseful) data.

Daniel Karrenberg: These kind of things needs to be measurered
independently, and the RIPE NCC can do that.

Marco Hogewoning: We need to make agreements about QOS. That
operators don't give the measurement traffic a high priority than
the normal traffic.

Rickard Dahlstrand: In general, what exactly is the scope of this?

Henk Uijterwaal: We all need to think about that indeed, let's
take that to the mailinglist.


6. Measuring multicast on the Internet with TTM - Gert Doering

Gert had proposed on the mailinglist to think about measuring
multicast with TTM. It was agreed that this is interesting and
that discussion should continue on the list. Possible first step
would be to look at how well the current set of testboxes are
multicast connected.

7. AOB

Rickard Dahlstrand: Is interested in more measurement nodes for
DNSMON. Too euro centric currently. Offers help to get more
machines in Africa, etc.

Meeting closes at 18:00.

Test Traffic WG @ RIPE51 - Amsterdam.

Actionpoints:

* RIPE NCC: research the Multicast connectivity of the testboxes *
All: Think about how to get more measurement nodes in TTM

Highlights:

* Proposal to measure Multicast using TTM * Consumer Broadband
Measurements draft send to the list and shortly discussed in WG *
Discussion on how to measure DNSMON from more nodes worldwide